CHAFJTER1
Introduction
to the Molecular of Baculoviruses
Biology
Jorge E. Vialard, Basil M. AriE and Christopher L D. Richardson 1. Introduction Over the last 10 years, baculovirus expression vectors have become a very popular and effective means with which to produce recombinant proteins in large quantities (1-S). Posttranslational modifications of the gene products of these insect viruses closely parallel glycosylation, fatty acid acylation, and phosphorylation in mammalian cells (reviewed in 6). Scaleup of insect cells in culture has also been largely perfected, making purification of large quantities of recombinant proteins a reality (7). In addition, baculoviruses offer an ecologically acceptable and effective alternative to chemicals for the control of forest and agricultural insect pests (8,9). Their demonstrated safety as expression vectors and pest management tools is the result of limited host specificity and lack of resemblance to mammalian viruses. The development of the baculovirus expression system was facilitated by the establishment of insect cell lines that support the replication of one subgroup, the nuclear polyhedrosis viruses (NPVs). The ability to propagate baculoviruses in cell culture has also allowed extensive study of their molecular biology (IO). The model virus in these studies is the Autographa californica NPV (AcNPV). Although it was first isolated from the alfalfa looper (Autographa californica), it multiplies readily in cell lines derived from both the fall armyworm (Spodoptera frugiperda) and the cabbage looper (Trichoplusia ni). Most expression vectors are basedon AcNPV infection of Spodopterafrugiperda From: Methods in Molecular Biology Vol. 39: Baculovirus Expression Protocols Edited by: C. D. Richardson Q 1995 Humana Press Inc., Totowa, NJ
1
2
Vialard, Arifi and Richardson
cells. However, the production of heterologous proteins in silkworm (Bomby~ mori; Bm) larvae relies on infection with recombinant BmNPV (4). The baculovirus expression system is based on introduction of the foreign gene into nonessential regions of the viral genome through allelic replacement. Production of the recombinant protein is achieved following infection of insect cells or larvae with the newly engineered virus. 2. Classification
The Baculoviridae are a family of double-stranded DNA viruses that infect a variety of arthropods. They can be divided into two subfamilies (II): the Eubaculovirinae (occluded baculoviruses) and the Nudibaculovirinae (nonoccluded baculoviruses). Eubaculovirinae infect the larvae of Lepidoptera, Coleoptera, Diptera, Hymenoptera, Neuptera, Siphonoptera, Thysanura, and Trichoptera, and even some crustaceans, such as shrimp and crabs (12,13). Members of the Nudibaculovirinae include the palm rhinoceros beetle (Orcytes rhinoceros) virus, the Hz-l virus, and the cricket (Gryllus campestris) virus (14). The Eubaculovirinae produce crystalline proteinaceous structures called occlusion bodies (OBs) (Figs. 1 and 2), which are absent in the Nudibaculoviridae. Virions embedded within the OBs are protected from environmental inactivating factors, such as UV light, desiccation, and nucleases. The Eubaculovirinae subfamily is made up of two genera (granulosis and nuclear polyhedrosis viruses) distinguished by the major protein that constitutes the OB matrix. The granulosis virus OBs are generally small (0.25-0.5 pm), contain a single virion, and are composed of a protein called granulin (15). The NPV OBs are much larger (1-15 p,rn diameter) and are composed of the closely related polyhedrin protein (16). NPV OBs, or polyhedra, usually contain a large number of virions embedded within the matrix. NPVs can be further separatedinto subgenera depending on the number of nucleocapsids surrounded by a common membrane; MNPVs and SNPVs contain multiple and single nucleocapsids, respectively. However, this difference does not seem to be phylogenetically important. For this reason, the abbreviations AcNPV and AcMNPV are often used interchangeably in the literature. Most baculoviruses isolated thus far are very host-specific, and the majority of Eubaculovirinae have been isolated from larvae of the Lepidoptera family. A survey of different baculoviruses with excellent electron photomicrographs can be found in the Atlas of Invertebrate Viruses (12-15).
Molecular
Biology of Baculoviruses
Secondary Infection Cells and Tissues
3
of
Ingestion
Primary Infection of Insect
Fig. 1. Life cycle of a baculovirus in an infected insect cell. Two populations of virus are formed-occluded virions (PDVs) accumulate in the nucleus and budded virions mature at the plasma membrane of the host cell. In nature, occlusion bodies serve to protect the virus from the environment (UV light and desiccation); they are ingested by larvae and become solubilized in the gut, releasing virions that attach and fuse with the cells of the midgut. The nucleocapsid is targeted to the nucleus, where replication and transcription occur. Budding virus promotes secondary infection to adjacent cells. The virus spreads to the ovaries, fat bodies, and most endothelial cells via the tracheal system.
3. Natural
Infection
of Insect
Larvae
Baculovirus infection is characterized by the production of two structurally and functionally distinct types of virions, the occluded or polyhedra-derived virion (PDV) and the extracellular or budded virion (BV).
4
Vialard, Arifi and Richardson
Molecular
Biology of Baculoviruses
5
The PDV type, which is responsible for primary infection, is embedded within the matrix of newly formed OBs (Fig. 2) and is required for dissemination in the environment. In a natural infection, larvae ingest PDVcontaining OBs that contaminate their food. The alkaline environment of the insect midgut causes the polyhedra to dissolve releasing the embedded virions. The liberated PDV infect midgut columnar epithelial cells by a process of receptor-mediated membrane fusion (I 7). These infected cells produce the BV type, which is required for secondary infection. The BV is responsible for systemic spread within insects and is also the type that infects cells in culture. Although it was previously thought that the spread of infection within the insect occurred via hemocytes in the hemocoel (l&19), this role has been recently ascribed to cells of the tracheal system (20). The tracheal system provides access to various tissues such as the ovaries, fat bodies, and most endothelial cells where both BV and PDV are produced. Cellular entry of the BV occurs through receptor-mediated adsorptive endocytosis (21,22). Studies of baculovirus infections in cell culture have revealed a series of landmark events. Following penetration of the plasma membrane, the nucleocapsids move toward the nucleus by a process that appears to require the formation of actin microfilaments (22). At the nucleus, the nucleocapsids are uncoated, and the DNA is released. At about this time, the nucleus becomes enlarged, and a distinct electron-dense granular structure, called the virogenic stroma, is formed (Fig. 2). This structure is associated with the nuclear matrix and is thought to be the site of nucleocapsid assembly (18-23). Viral transcription and replication may also take place at the Fig. 2. (opposite page) Autographu califomica nuclear polyhedrosis virus infected Sf9 insect cells. PanelsA and B show various featurescommon to a baculovirusinfection in Sf!Jcells. Occlusionbodies(OB) containingvirions (V) arepresentin the nuclei. Surroundingeachocclusionbody is a polyhedralenvelope. Replication and assembly of viral nucleocapsidsoccur in the nucleus in associationwith the virogenic stroma(VS). PlO is associatedwith fibrillar structures (FS), which are found both inside and outsidethe nucleus.Electron-dense spacers(ES) are associatedwith FS within the nucleus.ES are believed to be involved in the formation of the polyhedralenvelope,whereasFS favor lysis of the cell following virus maturation. Spindle bodies (S) that contain gp37 are diamond-like structuresthat are associatedwith the nuclearmembraneand can sometimesbe found in the cytoplasm. Their function is currently unknown.
6
Vialard, Arif, and Richardson
virogenic stroma. By 12 h postinfection, progeny BVs are produced and are released into the extracellular compartment. Polyhedra begin to be formed soon thereafter, and mature PDVs (surrounded by an envelope) become occluded. Feeding continues throughout infection (5-7 d) during which large numbers of OBs (up to 25% of the dry weight of the caterpillar) accumulate in the infected cells. Production of large numbers of OBs results from hyperexpression of the polyhedrin gene. The polyhedrin protein is generally essential for in vivo infections of larvae, but is expendable for infections in cultured cells. Most baculovirus expression vectors exploit this phenomenon by substituting a foreign gene for the coding sequence of polyhedrin. Eventually, the caterpillar stops feeding and undergoes several rapid physiological changes. Its cuticle melanizes, the musculature becomes flaccid, and the larva liquefies. Larval disintegration results in release of the OBs, which are subsequently dispersed in the environment. The baculovirus lifecycle is summarized in Fig. 1. 4. Virus Structure and Assembly The AcNPV nucleocapsid is bacilliform in shape, measures 3540 x 200-400 nm, and contains a circular, double-stranded DNA genome of approx 134 kb, which has been recently sequenced in its entirety (23a). Baculovirus DNA is tightly associated with a protamine-like protein known as p6.9 (24,25). The resulting complex forms the core of the nucleocapsid. In addition to p6.9, several other genes encoding nucleocapsid proteins have been identified. The most abundant protein in purified nucleocapsids is p39, the major capsid protein (26). Immunoelectron microscopy studies demonstrated its distribution throughout the length of the nucleocapsid (27). A similar localization is observed with ~24, a minor nucleocapsid protein (28). In contrast, p78/83, a proline-rich phosphoprotein, is associatedwith end structures of the nucleocapsids (29,301. The precise localization of ~87, a fourth nucleocapsid protein, has not been established (31). A model for nucleocapsid morphogenesis proposes that viral DNA is condensed by association with the basic p6.9 protein to form the core, whereas the capsid is assembled independently. The nucleoprotein complex enters the capsid through one end to form the mature nucleocapsid (23). A baculovirus encoded phosphoprotein, pp3 1, binds DNA nonspecifically, colocalizes with the virogenic stroma (Fig. 2), and is tightly associated with the nuclear matrix. It may play a role in packaging or, alternatively, in viral transcription and/or replication (32,33).
Molecular
Biology of Baculoviruses
7
Following assembly, nucleocapsids destined to become BV pass through the nuclear membrane and acquire a temporary envelope containing the virus-encoded protein, ~16 (34,35). This envelope is associated with the BV as it passes through the cytosol, but is lost when the virus buds through the plasma membrane. At the cell surface, the nucleocapsid acquires a loosely fitting envelope that contains the BV envelope glycoprotein, gp67 (36,37). This protein, which may be present in peplomerlike structures at one end of the virion, is required for BV infectivity by pH-dependent fusion (38). The nucleocapsids destined to become PDVs remain in the nucleus and acquire a de novo envelope of unknown origin, In MNPVs several nucleocapsids may be included within a single tightfitting envelope. At least three distinct proteins are associated only with PDV, but not with BV virions. Two of these, ~25 and gp41, appear to be associated with the PDV envelope (39-42). The other protein, ~74, is not essential to viral replication in cell culture, but is required for larval infection following ingestion of OBs (43). Its precise location is not known, As previously mentioned, the major component of the OB is polyhedrin, a protein that is highly conserved among the NPVs (12). Surrounding the matrix of the OB is a structure called the polyhedral envelope (PE) or calyx (Fig. 2). This structure has been reported to be rich in carbohydrate (44), but also contains a proteinaceous component called pp34 or PE protein (45-47). The PE may increase stability of the OB. Interruption of the pp34 gene produces OBs that are more sensitive to weak alkali conditions than wild-type OBs (48). A third gene that is involved in OB formation is ~25, also called few polyhedra (FP). Insertions of cellular DNA that interrupt this gene result in an FP phenotype (49). However, it is not known whether this protein participates in OB formation directly or indirectly. A second hyperexpressed protein, ~10, forms fibrous networks in the nucleus and cytoplasm of infected cells (5051) (Fig. 2). These plO-containing structures are associated with electron-dense spacers that form in the infected-cell nucleus. The spacers contain pp34 and are thought to be developing PE (47,51a). An association between p10 and microtubules has also been reported (52). Disruption of the ~10 gene results in mutants with varying phenotypes. Some mutants displayed aberrant attachment of PE, which resulted in the production of OBs sensitive to mechanical stress (51). Other ~10 deletion mutant studies suggest that it is involved in cell lysis late in infection (53). Deletion of the p10 protein prevented release of polyhedra from
Vialard,
Arifi
and Richardson
infected cells, presumably because of impaired nuclear disintegration. A protein, called gp37 or SLP, shares homology with a major OB component of another insect virus family, the Entomopoxviridae. It forms spindle-shaped inclusions that are found both in the cytoplasm (54~56), and in association with the nuclear membrane. SLP (gp37) may possess proteolytic activity (56a). In addition to the structural proteins described above, the baculoviruses encode a number of regulatory proteins. These include a ubiquitin-like factor, Z&u dismutase, protein kinases, PTPase, egt, proliferating cell nuclear antigen (PCNA), DNA polymerase, helicase, chitinase, cysteine protease, and a protein that blocks apoptosis. A summary of the genes encoded by the AcNPV that are known at this point is shown in Fig. 3 and Table 1. For a more detailed description of these genes and references pertaining to these genes, the reader is referred to Ayres et al. (23a) and a review by Kool and Vlak (57). 5. Baculovirus Gene Expression and Replication Baculovirus gene expression is regulated in a cascade-like fashion where activation of each set of genes relies on the synthesis of proteins from previous gene classes (reviewed in 58). This temporal regulation allows the grouping of baculovirus genes into three phases during infection: early (E), late (L), and very late (VL). Although most baculovirus genes can be placed into one of the above classes, some may be transcribed in more than one phase. The E genes are transcribed prior to viral DNA replication, whereas the L and VL genes are activated during or after replication, The late classes are not synthesized in the presence of aphidicolin, an inhibitor of replication. The reason for this dependence on viral replication is not yet known. The L genes are activated before the VL genes and are maximally transcribed over a short period of time (between 12 and 24 h postinfection). The VL genes are hyperexpressed following activation of the L genes and remain active well after L transcription has diminished (from 48 h postinfection onwards). The early genes generally encodeproteins with regulatory functions, such as transcription, replication, and modification of host processes. Late genes include BV and PDV structural proteins, whereas VL proteins are those involved in the processes of occlusion and cell lysis. The AcMNPV genome contains several regions called homologous repeats (hrs), which contain repeated sequencesharboring EcoRI recog-
Molecular
Biology of Baculoviruses
Fig. 3. Genes and open reading frames on the genome of Autographa caZifornica nuclear polyhedrosis virus that have been identified to date. The Kpn I, BamH I, BglII, PstI, Hi&III, and EcoRI restriction fragments of the circular dsDNA genome of AcNPV are classified using alphabetical letters. The locations of the genes on the 12%kb genome are indicated as positions from O-100 map units. Further information concerning the genes and their proposed functions is listed in Table 1.
nition sites (59). These elements appear to have two functions. As mentioned above, they act as enhancers for a number of E and L genes when present in plasmids (in transient expression assays) and within the viral
Vialard, Arif, and Richardson
10
Table 1 AcNPV Genes and Open Reading Framesa Class VL L L E V L L VL L VL ? E E ? E L E L ? L ? L E/L ? ? ? ? ? L ? ? L E/L E ? E E ? E/L E E
Designation PTPase (19 kDa) ORF 984 (38 kDa) ctl(5.6 kDa) ORF 453 (17 kDa) ORF 327 lef-2 (24 kDa) ORF 603 (24 kDa) polh (29 kDa) ~78183 Pk ORF 1020 lef-1 (3 kDa) egt (57 kDa) da13 da26 da18 da41 ORF 324 ORF 975 ORF 963 ORF 276 ORF 648 ORF 2070 ORF 405 ORF 615 ORF 858 ORF 507 ORF 1062 SOD (16 kDa) 17 kDa ORF 25 kDa ORF v-ubi PP31 13 kDa ORF 14 kDa ORF ORF 1089 P47 P79 ets (10 kDa) etm pcna (etl; 28 kDa)
Direction R L L L R R L R L R
L L R L R R L R L R L R R L L R R L R L L R L L R L L R L L L
Function Protein-tyrosine phosphatase ? Conotoxin ? ? Late expression factor ? Polyhedrin Nucleocapsid-associated phosphoprotein Protein knase? ? Late expression factor Ecdysosteroid UDP-glucosyltransferase ? ? ? ? ? ? ? ? ? 7 ? ? ? ? ? Cu/Zn superoxide dismutase ? ? Ubiquitin-like nuclear matrix-associated protein ? ? ? Late expression factor ? ? ? Proliferating cell nuclear antigen
Molecular
Biology
11
of Baculoviruses Table 1 (continued)
Class
Designation 21 kDa ORF 7kDaORF lef-8 10 kDa ORF 16kDaORF 11 kDaORF 9.6 kDa ORF 6kDaORF FP (25 kDa) ORF 474 gp37 (spindolin) DNApol(ll4 kDa) lef-3 ORF 252 ORF 1137 ORF 327 ORF 312 gp41 ORF 699 ORF 540 ORF 2541 cg30 p39 (cap) lef-4 ~25 ~143 (hel) 18-kDa ORF 15-kDa ORF 19-kDa ORF 38-kDa ORF lef-5 p6.9 ~48 ~80 HE65 kinase lef-7 ci cath is64 ORF 381
Direction
Function
R L L L R R L R L R L L L L L L L L L L R L L R R L R R R L R L L R L L L L R L R
? ?
Late expression factor ? ? ? 7 ? Few polyhedra phenotype ? Entomopoxvirus spindolin homolog DNA polymerase Late expression factor 7 ? 7 ? OV-associated glycoprotein ? ? ? Zinc finger/leucine zipper Major capsid protein Late expression factor OV envelope protein Helicase ? ? ? ? Late expression factor Basic DNA binding protein ? Capsid associated protein ? tyr/ser kinase Late expression factor Chitinase Cysteine protease (cathepsin-like) BV major envelope protein ? (continued)
12
Vialard, Arif
and Richardson
Table 1 (continued) Class ?
L L IJLV L L E E E/L VL L E E/L ? ? ? ? E/L E E ?
Designation
Direction
ORF 951 ~24 ml6 PP34 25-kDa ORF 48-kDa ORF P94 P35 ~26 PlO P74 ME53 IE-0 lo&Da ORF 49-kDa ORF 43-kDa ORF 23-kDa ORF IE-1 (67 kDa) IE-N (47 kDa) pe38 ORF 246
R R R R R R L R R R L L R R R R L R L R R
Function ?
Capsid-associated protein Nuclear membrane protein Polyhedral envelope protein ? ? ? Blocks apoptosis ? Cytoplasmic/nuclear fibrous structures Essential for OB infectivity in larvae Zinc finger First exon of IE-1 ? ? ? ? Transactivates early genes Modulates IE- 1 expression Zinc finger/leucine zipper ?
aTime of transcription is classified as early (E), late (L), or very late (VL). Direction of transcription is indicated rightward (R) or leftward (L) in respect to the O/100 point in Fig. 3.
genome (60-62). More recently, they were shown to act as origins of replication for plasmids when cotransfected together with various fragments of the baculovirus genome or introduced into infected cells (6365). Regions of the baculovirus genome that may encode factors required for replication have also been identified in a plasmid-based replication system (66). Some genespresent in these regions include the baculovirusencoded DNA polymerase, ~143 (helicase), and proliferating cell nuclear antigen (PCNA) genes. Baculovirus E genes are transcribed by the host RNA polymerase 11. Consequently, transcription from the E promoters is abolished in the presence of a-amanitin, an inhibitor of RNA polymerase II (67). The involvement of RNA polymerase II in E gene transcription was demonstrated by accurate initiation of mRNAs in an in vitro transcription system using nuclear extracts from uninfected cells (68). Thus, the E
Molecular
Biology of Baculoviruses
13
promoters resemble typical eukaryotic RNA polymerase II responsive promoters that contain DNA elements that are recognized by host transcription factors (69,70). Previously, the E genes were often subdivided into immediate early (IE) and delayed early (DE) components depending on their requirement for viral protein synthesis in transient expression assays where reporter genes were placed under the control of IE or DE promoters and transfected into insect cells (71). Under these conditions, IE genes d&onot require viral proteins for their activation and are transcribed in uninfected cells. However, this distinction cannot be made in infected cells even in the presence of cycloheximide, an inhibitor of protein synthesis; both IE and DE genes are expressed. This suggests that factors required for DE promoter activation in transient expression are provided during the initial phase of infection by proteins associated with the virion. For example, IE-1, a transactivator of DE genes, has recently been shown to be a component of the BV (72). Most IE proteins identified thus far appear to be involved in the regulation of viral transcription. IE-1 is a 66.9~kDa polypeptide capable of transactivating a number of early and late promoters in transient expression (71,73-7.5). IE-l-mediated activation requires the presence of hr elements in cis with the responsive promoter (60) and has been recently demonstrated to impart its activity through binding to hr (homologous repeat) elements (76,77). The IE-0 protein is a product of alternative splicing, which results in the utilization of an exon 5’ to the IE-1 promoter. This results in fusion of 54 amino acids to the N-terminus of IE-1 (7879). The IE-0 gene contains its own promoter, which is regulated differently than the TIE-1promoter. Also, the IE-O transacting functions differ from those of IE-1. For example, in addition to activating a number of genes, IE-1 negatively regulates IE-0 transcription (79). In contrast, IE-1 expression is stimulated by IE-0 (80). A third lE protein, IE-N (or lE-2), augments IE-l-mediated transactivation moderately, exhibits an autoregulatory activity, and is downregulated by IE-1 (81,82). IE-N contains a zinc finger and a leucine zipper, motifs characteristic of some transcription factors. Two other baculovirus genes, pe-38 and cg-30, also encode these motifs (83,84). PE38 stimulates IE-N and ~143 (helicase) transcription in transient expression assays (81,85). A truncated form of PE38, that does not have stimulatory activity and appearsto be the product of alternative transcriptional initiation, has been recently identified (86).
14
Vialard, Arif
and Richardson
Several DE genes encode components of the DNA replication machinery. For example, the PCNA (ETL) gene product was shown to be involved in both replication and late gene transcription (87,88). A mutation in pcna produces virus that exhibits delayed replication and late gene expression. In addition, two ts mutants defective in late gene expression have been mapped. One, which is also defective in replication, has a mutation in the ~143 (helicase) gene (89). The second ts mutant is rescued by wild-type p47, a protein whose function is not yet known (90). The L and VL genes are under the control of an a-amanitin-resistant RNA polymerase that is induced during infection (67,91,92). This polymerase activity is also resistant to tagetitoxin, an inhibitor of insect RNA polymerase III (93). Partial purification of the a-amanitin-resistant activity suggests that its protein composition is different from the three host RNA polymerases (94). It is not known whether its components are virus encoded, host encoded, or a combination of the two. The specificity of the virus-induced polymerase may be dictated by the unique baculovirus L and VL promoters. They differ from most RNA polymerase II promoters in that they are very compact and do not contain DNA elements, such as the TATA box, present in most eukaryotic promoters. The only element that seems to be present in all L and VL promoters is a consensus core sequence, TAAG, which contains the transcriptional start site and is essential for activity (12,25,26,95). This element is present in the promoters of the hyperexpressed VL genes as part of a very well-conserved sequence, TAATAAGT/AATT. This sequence is responsible for the very high levels of expression observed from the VL promoters (95). Sequences in the leader region between the TAAG element and the start codon may influence levels of transcription somewhat (96). The factors that interact with the L and VL promoters to regulate transcription are not known. The development of an in vitro transcription system that utilizes nuclear extracts from late in infection may help in their identification (93). A transient expression system utilizing a mixture of successively smaller fragments of the baculovirus region that are able to activate L and VL promoters has resulted in the identification of a number of genes encoding late gene expression factors (lef) (75,97-99b). Although some of these genes have not been previously identified (lef l-S>, some of them were previously described as regulators of early transcription or replication (ie-1, ie-n, and ~143).
Molecular
Biology
of Baculoviruses
15
The predicted amino acid sequence of lef-8 contains a conserved motif of RNA polymerases (99b) and the protein copurifies with the virusinduced RNA polymerase activity (99c) suggesting that it is a component of the a-amanitin-resistant polymerase. 6. Baculoviruses as Expression Vectors and Engineered Insecticides Two important features of baculoviruses account for the success of this virus as an expression vector. First, the virus contains a number of nonessential genes that can be replaced by an exogenous gene. Second, many of these genes, particularly the very late ones, are under the control of powerful promoters that allow abundant expression of the passenger recombinant gene. Most of the expression systems in baculoviruses make use of the polyhedrin or p10 promoters together with their associated flanking sequences. Both polyhedrin and p10 are nonessential, since deletion of these genes does not affect the replication of the virus in cell culture (100,101). The p6.9 promoter appears to be as efficient as the p10 and polyhedrin promoters and may be harnessed for recombinant protein expression (IOla). For reviews concerning the use of baculoviruses in the expression of recombinant proteins important in the pharmaceutical industry and in basic research, the reader is referred to Luckow and Summers (102) and O’Reilly et al. (2). The samebasic principles apply to the utilization of these viruses in pest management strategies where the wild-type virus is ineffective in producing the desirable control of an insect pest. When the virus is used in pest management strategies, a number of important criteria must be considered. In contrast to the use of expression vectors in cell culture where synthesis of polyhedrin in not necessary, the formation of OBs is important for the viral insecticide to survive in nature long enough for the insect to ingest it. Without the protection afforded by the OBs, the virus is quickly inactivated, Although wild-type baculovuuses have been used as insecticides, the lethal dose and time can be improved by genetic engineering (10lb). A number of candidate genes with potential insecticidal properties have been inserted into baculoviruses, and the engineered viruses have been tested against the target insects (lOlb,lO3114). An insect-specific toxin that appears to be effective in enhancing AcNPV as an insecticide is derived from the venom of the North AFrican scorpion, Androctonus austrah Hector (104,106). The gene product produced the desired neurotic effects, and reduced both the median survival
Vialard,
Arif
and Richardson
time of the infected insect and the median lethal dose of virus (104). This modified baculovirus was used recently in a field trial and was shown to be more effective in reducing crop damage as a result of its increased lethality (107). A toxin (TxP-I) derived from the venom of female mites, Pyemotes tritici, was also shown to be effective against insects. The potential of this toxin was investigated by engineering a cDNA encoding the toxin into AcNPV (108). Larvae infected with the virus containing the engineered gene became paralyzed during infection. Other genes that have been introduced into baculoviruses for insecticidal purposes include juvenile hormone esterase (jhe; 110, II 1) and an insect diuretic hormone (112). Deletion of the ecdysosteroid UDPglucosyltransferase (egt) gene of AcNPV also increased lethality of the virus by interfering with insect metamorphosis and moulting (113,114). In short, the baculovirus expression system has made a great impact in both academic and applied pharmaceutical research. It has become a major workhorse in most expression laboratories. References 1. Luckow, V. A and Summers, M. D. (1988) Trends in the development of baculovirus expression vectors. Bioflechnology 6,47-55. 2. O’Reilly, D. R., Miller, L. K., and Luckow, V. A. (1992) BacuZovirus Expression Vectors: A Laboratory Manual. W. H. Freeman, New York. 3. King, L. A. and Possee, R. D. (1992) The Baculovirus Expression System: A Laboratory Guide. Chapman and Hall, London. 4. Maeda, S. (1989) Expression of foreign genes in insects using baculovirus vectors. Annu. Rev. Entomol. 34,351-372. 5 Kidd, M. and Emery, V. C. (1993) The use of baculoviruses as expression vectors. Appl. Biochem. Biotech. 42, 137-159. 6. Luckow, V. A. (1991) Cloning and expression of heterologous genes in insect cells with baculovirus vectors, in Recombinant DNA Technology and Applications (Prokop, A., Bajpai, R. K., and Ho, C., eds.), McGraw-Hill, New York, pp. 97-152. 7. Van Lier, F. L. J., Vlak, J. M., and Tramper, J. (1992) Production of baculovirusexpressed proteins from suspension cultures of insect cells. Animal Cell Biotech. 5,169-188. 8. Wood, H. A. and Granados, R. R. (1991) Genetically engineered baculoviruses as agents for pest control. Annu. Rev. Microbial. 45,69-87. 9. Huber, J. (1986) Use of baculoviruses in pest management programs, in The Biology ofBaculoviruses, vol. II (Granados, R. R. and Federici, B., eds.), CRC, Boca Raton, FL, pp 181-202. 10. Blissard, G. W. and Rohrmann, G. F. (1990) Baculovirus diversity and molecular biology. Annu. Rev. Entomol. 35, 127-155.
Molecular
Biology of Baculoviruses
17
11. Fran&, R. I. B., Fauquet, C, M., Knusdon, D. L, and Brown, F. (199 1) Classification and nomenclature of viruses. Fifth report of the International Committee on Taxonomy of Viruses. Virology Division of the International Union on Microbiological Societies. Arch. Viral. 2, 117-123. 12. Adams, J. R. and McClintock, J. T. (1991) Baculoviridae. Nuclear polyhedrosis viruses. Part 1. Nuclear polyhedrosis viruses of insects, in Atlas of Invertebrate Viruses (Adams, J. R. and Bonami J. R., eds.), CRC, Boca Raton, FL, pp. 87-204. 13. Couch, J. A. (1991) Baculoviridae. Nuclear polyhedrosis viruses. Part 2. Nuclear polyhedrosis viruses of invertebrates other than insects, in Atlas af Invertebrate Viruses (Adams, J. R. and Bonami J. R., eds.), CRC, Boca RatIn, FL, pp. 205-226. 14. Huger, A. M. and Krieg, A. (1991) Baculoviridae. Nonoccluded baculoviruses, in Atlas of Invertebrate Viruses (Adams, J. R. and Bonami, J. R., eds.), CRC, Boca Raton, FL, pp. 287-3 19. 15. Tanada, Y. and Hess R. T. (1991) Baculoviridae. Granulosis viruses, in Atlas of Invertebrate Viruses (Adams J. R. and Bonami J. R., eds.), CRC, Boca Raton, FL, pp. 227-257. 16. Rohrmann, G. F. (1986) Polyhedrin structure. J. Gen Virol. 67,1499--1513. 17. Horton, H. M. and Burand, J. P. (1993) Saturable attachment sites Ibr polyhedron-derived baculovirus on insect cells and evidence for entry via direct membrane fusion, J. Viral. 67,1860-l 868. 18. Granados, R. R. and Lawler, K. A. (1981) In vivo pathway of Autographa californica baculovirus invasion and infection. Virology 108,297-308. 19. Keddie, B. A., Aponte, G. W., and Volkman, L. E. (1989) The pathway of infection of Autographa californica nuclear polyhedrosis virus in an insect host. Science 243,1728-1730. 20. Engelhard, E. K., Kam-Morgan, L. N. W., Washburn, J. O., and Volbman, L. E. (1994) The insect tracheal system: A conduit for the systemic spread of Autographa californica M nuclear polyhedrosis virus. Proc. Natl. Acad. Sci. USA 91,3224-3227. 21. Volkman, L. E. and Goldsmith, P. A. (1985) Mechanism of neutralization of budded Autographa californica nuclear polyhedrosis virus by monoclonil antibody inhibition of entry by adsorptive endocytosis. Virology 143, 185-195. 22. Charlton, C. A. and Volkman, L. E. (1993) Penetration ofAutographa californica nuclear polyhedrosis virus nucleocapsids into IPLB Sf 21 cells induces actin cable formation. Virology 197,245-254. 23. Fraser, M. J. (1986) Ultrastructural observations of virion maturation in Autographa californica nuclear polyhedrosis virus infected Spodoptera frugiperda cell cultures. J. Ultrastruct. Mol. Struct. Res. 95, 189-195. 23a. Ayres, M. D., Howard, S. C., Kuzio, J., Lopez-Ferber, M., and Poaee, R. D. (1994) The complete DNA sequence of Autographa califomica nuclear polyhedrosis virus. Virology 202,586-605 24. Tweeten, K. A., Bulla, L. A., and Consigli, R. A. (1980) Characterization of an extremely basic protein derived from granulosis virus nucleocapsid. J. Viral. 33, 866-876.
18
Vialard, Arif
and Richardson
25. Wilson, M. E, Mainprize, T. H., Friesen, P. D., and Miller, L. K. (1987) Location, transcription and sequence of a baculovuus gene encoding a small arginine-rich polypeptide. J. Viral. 61,661-666. 26. Thiem, S. M. and Miller, L. K. (1989) Identification, sequence and transcriptional mapping of the major capsid protein gene of the baculovirus Autographa californica nuclear polyhedrosis. J. Virol. 63,2008-2018. 27. Russell, R. L. Q., Pearson, M. N., and Robrmann, G. F. (1991) Immunoelectron microscopic examination of Orgyia pseudotsugatu multicapsid nuclear polyhedrosis-virus infected Lymantriu dispar cells: time course and localization of major-polyhedron-associated proteins. J. Gen. Virol. 72,275-283. 28. Wolgamot, G. M., Gross, C. H., Russell, R. L. Q., and Rohrmann, G. F. (1993) Immunocytochemical characterization of ~24, a baculovuus capsid-associated protein. J. Gen Virol 74, 103-107. 29 Possee, R. D , Sun, T.-P., Howard, S. C., Ayres, M. D., Hill-Perkins, M , and Gearing, K. L. (1991) Nucleotide sequence of the Autogrupha californica nuclear polyhedrosis 9.4 kbp EcoRI-I and -R (polyhedrin gene) region. Virology 185,229-241.
30. Vialard, J. E. and Richardson, C. D. (1993) The 1,629-nucleotide open reading frame located downstream of the Autographa californica nuclear polyhedrosis virus polyhedrin gene encodes a nucleocapsid-associated phosphoprotein. J. Virol. 67,5859-5866.
31. Mtiller, R., Pearson, M. N., Russell, R. L. Q., and Rohrmann, G. F. (1990) A capsid-associated protein of the multicapsid nuclear polyhedrosis virus of Orgyiu pseudotsugata: genetic location, sequence, transcriptional mapping, and immunocytochemical characterization. Virology 176, 133-144 32. Guarino, L. A. and Smith, M. W. (1990) Nucleotide sequence of the 39K gene region of Autographa californica nuclear polyhedrosis virus. Virology 179, l-8. 33 Guarino, L. A., Dong, W., Xu, B , Broussard, D R., Davis, R. W., and Jarvis, D. L (1992) Baculovirus phosphoprotein pp3 1 is associated with virogenic stroma. J. Viral. 66,7113-7120.
34. Oellig, C., Happ, B., Muller, T., and Doerfler, W. (1987) Overlapping sets of viral RNAs reflect the array of polypeptides in the EcoRI J and N fragments (map positions 81.2 to 85.0) of the Autographa californica nuclear polyhedrosis vnus genome. J. Viral. 61,3048-3057 (erratum, 63,1494, 1989). 35. Gross, C. H., Wolgamot, G. M., Russell, R. L. Q., Pearson, M. N., and Rohrmann, G. F. (1993) A baculovirus encoded 16-kDa glycoprotein localizes near the nuclear membrane of infected cells. Virology 192,386-390. 36. Whitford, M., Steward, S., Kuzio, J , and Faulkner, P. (1989) Identification and sequence analysis of a gene encoding gp67, an abundant envelope glycoprotein of the baculovirus Autographa californica nuclear polyhedrosis virus. J. Virol. 63,1393-1399.
37. Blissard, G. W. and Rohrmann, G. F. (1989) Location, sequence, transcriptional mapping and temporal expression of the gp64 envelope glycoprotein gene of the Orgyiapseudotsugata multicapsid nuclear polyhedrosis virus, Virology 170, 537-555.
Molecular
Biology of Baculoviruses
19
38. Blissard, G. W. and Wenz, J. R. (1992) Baculovirus p64 envelope glycopt otein is sufficient to mediate pa-dependent membrane fusion. J. Virol. 66,6829--6835. 39. Russell, R. L. Q. and Rohrmann, G. F. (1993) A 25kDa protein is associated with the envelopes of occluded baculovirus virions. Virology 195,532-540. 40. Nagamine, T., Sugrmorr, H., Nakamura, K., Saga, S., and Kobayashi, M. (1991) Nucleotide sequence of the gene coding for ~40, an occluded virion-specific polypeptide of Bombyx mori nuclear polyhedrosrs virus. J. Znverfebr. Puthol. 58, 290-293.
41. Whitford, M. and Faulkner, P. (1992) Nucleotide sequence and transcrlptional analysis of a gene encodmg ~41, a structural glycoprotein of the bacrlovirus Autographa californica nuclear polyhedrosis virus. J. Virol. 66, 47634768 (erratum, 67,2427, 1993). 42. Whitford, M. and Faulkner, P. (1992) A structural polypeptide of the bacu lovirus Autogrupha californica nuclear polyhedrosis virus contains O-linked Nacetylglucosamine. J. Virol. 66,3324-3329. 43. Kuzio, J., Jaques, R., and Faulkner, P. (1989) Identification of p74, a gene essential for virulence of baculovirus occlusion bodies. Virology 173,759-76X 44. Minion, F. C., Coons, L. B., and Broome, J. R. (1979) Characterizatiorl of the polyhedral envelope of the nuclear polyhedrosis virus of Heliothis virescens. J. Invert. Pathol. 34,303-307.
45. Whitt, M. A. and Manning, J. S. (1988) A phosphorylated 34kDa protehl and a subpopulation of polyhedrin are thiol-linked to the carbohydrate layer surrounding a baculovirus occlusion body. Virology 163,33-42. 46. Gombart, A. F., Pearson M. N., Rohrmann, G. F., and Beaudreau, G. S. (1989) A baculovirus polyhedral envelope-associated protein: genetic location, nucleotide sequence and immunocytochemical characterization. Virology 169, 182- 193, 47. Russell, R. L. Q. and Rohrmann, G. F. (1990) A baculovirus polyhedron envelope protein: immunogold localization in infected cells and mature pol!/hedra. Virology 174, 177-184. 48. Zuidema, D., Klinge-Roode, E. C., van Lent, J. W. M., and Vlak, J. M. (1989) Construction of an Autographa californica nuclear polyhedrosis virus mutant lacking the polyhedral envelope. Virology 173,98-108. 49. Beames B. and Summers, M. D. (1989) Location and nucleotide sequence of the 25K protein missing from baculovirus few polyhedra (FP) mutants. Virology 168, 344-353. 50. van der Wilk, F., van Lent, J. W. M., and Vlak, J. M. (1987) Immunogold detection of polyhedrin, p10 and virion antigens in Autographa californica nuclear polyhedrosis virus-infected Spodopterufrugiperda cells. J. Gen. Viral. 68,2615-:!623.
51. Williams, G. V., Rohel, D. Z., Kuzio, J., and Faulkner, P. (1989) A cytopathological investigation of Autographa californica nuclear polyhedrosis virus p10 gene function using insertion/deletion mutants. J. Gen. Virol. 70, 187-202. 51a. van Lent, J. W. M., Groenem, J. T. M,, Klinge-Roode, E. C., Rohrmann, G. F., Zuidema, D., and Vlak, J. M. (1990) Localization of the 34kDa polyhedron envelope protein in Spodoptera frugiperda cells infected with Autographa calijomica nuclear polyhedrosis virus. Arch. Virol. 111, 103-l 14.
20
Vialard,
Arifi
and Richardson
52. Volkman, L. E. and Zaal, K. J. M. (1990) Autographa californica M nuclear polyhedrosis virus: microtubules and replication. Virology 175,292-302. 53. van Oers, M. M., Flipsen, J. T. M., Reusken, C. B. E. M., Sliwinsky, E. L., Goldbach, R. W., and Vlak, J. M. (1993) Functional domains of the p10 protein of Autogruphu culifornicu nuclear polyhedrosis virus. J. Gen. Virol. 74,
563-574. 54. Dall, D., Sriskantha, A , Vera, A., Lai-Fook, J., and Symonds, T. (1993) A gene encoding a highly expressed spindle body protein of Heliothis urmigeru entomopoxvirus. J. Gen. Virol. 74, 18 1 l-l 8 18. 55. Gross, C. H., Wolgamot, G. M., Russell, R. L. Q., Pearson, M. N., and Rohrmann, G. F. (1993) A 37-kilodalton glycoprotein from a baculovirus of Orgyiu pseudotsugutu is localized to cytoplasmic inclusion bodies. J. Viral. 67, 469-475.
56. Vialard, J. E., Yuen, L , and Richardson, C D. (1990) Identification and characterization of a baculovirus occlusion body glycoprotein which resembles spheroidin, an entomopoxvirus protein. J. Virol. 64,5804-5811. 56a. DeMoerlooze, L N. A., Vialard, J. E., and Richardson, C. F. (1995) Structural and functional characterization of baculovirus and entomopox virus spindle proteins (submitted for publication). 57. Kool, M. and Vlak, J. M. (1993). The structural and functional organization of the Autograph culijornicu nuclear polyhedrosis virus genome. Arch Vol. 130,1-16. 58. Friesen, P. D. and Miller, L. K. (1986) The regulation of baculovirus gene expression, in The Molecular Biology of Buculoviruses (Doerfler, W. and Boehm P., eds.), Springer-Verlag, Berlin, pp. 31-49. 59. Guarino, L. A., Gonzales, M. A., and Summers, M. D. (1986) Complete sequence and enhancer function of the homologous DNA regions of Autographa culifornicu nuclear polyhedrosis virus. J. Virol. 60,224-229. 60. Guarino, L. A. and Summers, M. D. (1986) Interspersed homologous DNA of Autographa culijornicu nuclear polyhedrosis virus enhances delayed-early gene expression. J. Virol. 60,2 15-223. 61. Rodems, S. M. and Friesen, P. D. (1993) The hr5 transcriptional enhancer stimulates early expression from the Autographa culijornicu nuclear polyhedrosis virus genome but is not required for virus replication. J. Viral. 67,5776-5785. 62. Guarmo, L. A. and Summers, M. D. (1988) Functional mapping of Autographa culifornicu nuclear polyhedrosis vnus genes required for late gene expression. J. Viral. 62,463-471.
63. Pearson, M., Bjornson,
R., Pearson, G., and Rohrmann, G. (1992) The baculovirus genome: evidence for multiple replication origins. Science 257, 1382-1384. 64. Kool, M., Voeten, J. T. M., Goldbach, R. W., Tramper, J., and Vlak, J. M. (1993) Identification of seven putative origins of Autographa culifornicu multiple nucleocapsid nuclear polyhedrosis virus DNA replication. J. Gen Virol. 74,2661-2668. 65. Leisy, D. J. and Rohrmann, G. F. (1993) Characterization of the replication of plasmids containing hr sequences in baculovirus-infected Spodopterufrugiperdu cells. Virology 196,722-730. Autographa
culifonicu
Molecular
Biology
of Baculoviruses
21
66. Kool, M., Voeten, J. T. M., Goldbach, R. W., and Vlak, J. M. (1994) Functional mapping of regions of the Autographa culifornica nuclear polyhedrosis viral genome required for DNA replication. Virology 198,680-689. 67. Huh, N. E. and Weaver, R. F. (1990) Identifying the RNA polymerases 1hat synthesize specific transcripts of the Autographa culifornicu nuclear poly hedrosis virus. J. Gen. Virol. 71, 195-201. 68. Hoopes, R. R. and Rohrmann, G. F. (1991) In vitro transcription of baculovirus immediate early genes: accurate mRNA initiation by nuclear extracts from both insect and human cells. Proc. N&l. Acud Sci. USA 88,45 13-45 17. 69. Krappa, R., Behn, K. A., Jahnel, F., Doerfler, W., and Knebel, M. D. (1992) Differential factor binding at the promoter of early baculovirus gene PI138 during viral infection: GATA motif is recognized by an insect protein. J. Wrol. 66, 3493-3503.
70. Kogan, P. H. and Blissard, G. W. (1994) A baculovirus gp64 early promoter is activated by host transcription factor binding to CACGTG and GATA elements. J. Virol. 68,8 13-822. 71. Guarino, L. A. and Summers, M. D. (1986) Functional mapping of a trms-activating gene required for expression of a baculovirus delayed-early gene. J. Virol. 57,563-57 1. 72. Theilmann, D. A. and Stewart, S. (1993) Analysis of the Orgyiu pseudotsugutu multicapsid nuclear polyhedrosis virus trans-activators IE-1 and IE-2 using monoclonal antibodies. J. Gen. Viral. 74, 1819-1826. 73. Nissen, M. S. and Friesen, P. D. (1989) Molecular analysis of the transcriptional regulatory region of an early baculovirus gene. J. Virol. 63,493-503. 74. Guarino, L. A. and Summers, M. D. (1987) Nucleotide sequence and temporal expression of a baculovirus regulatory gene. J. Viral. 61,2091-2099. 75. Passarelli, A. L. and Miller, L. K. (1993) Three baculovirus genes involved in late and very late gene expression: ie-1, ie-n, and lef-2. J. Virol. 67,2149-2158. 76. Guarino, L. A. and Dong, W. (1991) Expression of an enhancer-binding protein in insect cells transfected with the Autographa culifornicu nuclear poly hedrosis virus IEl gene. J Viral. 65,3676-3680. 77. Kovacs, G. R., Choi, J., Guarino, L. A., and Summers, M. D. (1992) Functional dissection of the Autographa culifornicu nuclear polyhedrosis virus immediateearly transcriptional regulatory protein. J. Virol. 66,7429-7437. 78. Chisholm, G. E. and Henner, D. J. (1988) Multiple early transcripts and splicing of the Autogruphu culifonicu nuclear polyhedrosis virus IE-1 gene. J. Virol. 62,3 1!13-3200. 79. Kovacs, G. R , Guarino, L. A., and Summers, M. D. (1991) Novel regulatory properties of the IEl and IEO transactivators encoded by the baculovirus Autographa culifornicu multicapsid nuclear polyhedrosis virus. J. V;rol. 65, 5281-5288. 80. Carson, D. D., Summers, M. D., and Guarino, L. A. (1991) Molecular analysis of a baculovirus regulatory gene. Virology 182,279-286 81. Carson D. D., Summers, M. D., and Guarino, L. A. (1991) Transient expression of the Autographa culifornicu nuclear polyhedrosis virus immediate-early gene, IE-N, is regulated by three viral elements. J. Viral. 65,945-95 1.
Vialard, Arif
22
and Richardson
82. Carson, D. D., Summers, M. D., and Guarino, L. A. (1988) Functional mapping of an AcNPV immediate early gene which augments expression of the IE- 1 trunsactivated 39k gene. Virology 162,44445 1. 83. Krappa R. and Knebel-Morsdorf, D. (1991) Identification of the very early transcribed baculovirus gene PE-38. J. Virol. 65,805-812. 84 Thiem, S. M. and Miller, L. K. (1989) A baculovirus gene with a novel transcription pattern encodes a polypeptide with a zinc finger and a leucine zipper. J. Virol. 63,4489-4497.
85. Lu, A. and Carstens, E. B. (1993) Immediate-early baculovirus genes transactivate the ~143 promoter of Autographa califomicu nuclear polyhedrosis virus. Virology 195,710-718.
86. Wu, X., Stewart, S., and Theilmann, D. A. (1993) Alternative transcriptional initiation as a novel mechanism for regulating expression of a baculovirus tram activator. J Virol. 67,5833-5842. 87. Crawford, A. M. and Miller, L. K. (1988) Characterization of an early gene accelerating expression of late genes of the baculovirus Autographa californica nuclear polyhedrosis virus. J. Virol. 62,2773-278 1. 88. O’Reilly, D. R., Crawford, A. M., and Miller, L. K. (1989) Viral proliferating cell nuclear antigen. Nature 337,606. 89. Lu, A. and Carstens, E. B. (1991) Nucleotide sequence of a gene essential for viral DNA replication in the baculovirus Autographa californica nuclear polyhedrosis. Virology 181,336-347. 90. Carstens, E. B., Lu, A., and Chan, H. B. (1993) Sequence, transcriptional mapping, and overexpression of ~47, a baculovirus gene regulating late gene expression. J. Virol. 67,2513-2520. 91. Grula, M. A., Buller, P. L, and Weaver, R. F. (1981) a-Amanitin-resistant viral RNA synthesis in nuclei isolated from nuclear polyhedrosis virus-infected Heliothis zea larvae and Spodopteru frugiperdu cells. J. Virol. 38,9 19-92 1. 92. Fuchs, L. Y., Woods, M. S., and Weaver, R. F. (1983) Viral transcription during Autographa californica nuclear polyhedrosis virus infection: a novel RNA polymerase Induced in infected Spodoptera frugiperda cells. J. Virol. 48,641-646. 93. Glocker, B., Hoopes, R. R., Jr., Hodges, L., and Rohrmann, G. F. (1993) In vitro transcription from baculovirus late gene promoters: accurate mRNA initiation by nuclear extracts prepared from infected Spodopterufrugiperda cells. J. Virol. 67, 3771-3776. 94. Yang, L., Stetler, D. A., and Wever, R. F. (1991) Structural comparison of the Autographa califomica nuclear polyhedrosis virus-induced RNA polymerase and the three nuclear RNA polymerases from the host, Spodopteru frugiperda Virus Res. 20,25 l-264. 95. Possee, R. D. and Howard, S. C. (1987) Analysis of the polyhedrin gene promoter of the Autographa califomica nuclear polyhedrosis virus. Nucl. Acids Res. 15, 10,233-10,248. 96. Ooi, B. G., Rankin, C., and Miller, L. K. (1989) Downstream sequences augment transcription from the essential initiation site of a baculovirus polyhedrin gene. J. Mol. Biol. 210,721-736.
Molecular
Biology
of Baculoviruses
23
97. Passarelli, A. L. and Miller, L. K. (1993) Identification of genes encoding late expression factors located between 56.0 and 65.4 map units of the Autographa californica nuclear polyhedrosis virus genome. Virology 197,704-7 14. 98. Passarelli, A. L. and Miller, L. K. (1993). Identification and characterization of lef-1, a baculovuus gene involved in late and very late gene expression. J. Viral. 67,3481-3488.
99. Li, Y., Passarelli, A. L., and Miller, L. K. (1993). Identification, sequence, and transcriptional mapping of lef-3, a baculovirus gene involved in late and very late gene expression, J, Virol. 67,5260-5268. 99a. Passarelli, A. L. and Miller, L. K. (1994) Identification and transcriptional regulation of the baculovirus lef-6 gene. J. Virol. 68,4458-4467. 99b. Passarelli, A. L., Todd, J. W., and Miller, L. K. (1994) A baculovixus gene involved in late gene expression predicts a polypeptide with a conserved motif of RNA polymerases. J. Virol. 68,4673-4678. 99c. Iorio, C., Vialard, J. E., Lagace, M., and Richardson, C. D (1995) Pol:/peptide components associated with the Autographa californica nuclear polyhedrosis vital RNA polymerase (in preparation). 100. Smith, G. E., Fraser, M. J., and Summers, M. D. (1983) Molecular engineering of the Autographa californlca nuclear polyhedrosis virus genome: deletion mutations within the polyhedrin gene. J. Virol. 46,584-593. 101. Weyer, U., Knight, S., and Possee, R. D. (1990) Analysis of very late gene expression by Autographa californica nuclear polyhedrosis virus and the further development of multiple expression vectors. J. Gen. Virol. 71, 1525-1534. 101a. Bonning, B. C., Roelvnik, P. W., Vlak, J. M., Possee, R. D., and Hammock, B. D. (1994) Superior expression of juvenile hormone esterase and p-galac iosidase from the basic promoter of Autographa californica nuclear polyhedror is virus compared to the p10 and polyhedrin promoter. J. Gen. Virol. 75, 1551-I 556. 101b. Bonning, B. C. and Hammock, B. D. (1992) Development and potential of genetically engineered viral insecticides. Biotechnol. Genet. Eng. Rev. 10,45!3-489. 102. Luckow, V. A. and Summers, M. D. (1988) Trends in the development of baculovirus expression vectors. Biotechnology 6,47-55. 103. Merryweather, A. T., Weyer, U., Harris, M. P. G., Hirst, M., Booth, T., and Possee, R. D. (1990) Construction of genetically engineered baculovirus insecticides containing the Bacillus thuringiensis subsp. kurstaki HD-73 delta endotoxin. J. Gen. Virol. 71, 1535-1544. 104. Stewart, L. M. D., Hirst, M., Ferber, M. L, Merryweather, P., Cayley, P J., and Possee, R. D. (1991) Construction of an improved baculovirus insecticide containing an insect-specific toxin gene. Nature 352,85-88. 105. Maeda, S., Volrach, S. L., Hanzlik, T. N., Harper, S. A., Majima, K., Maddox, D. W., Hammock, B. D., and Fowler, E. (1991) Insecticidal effects of an insect-specific neurotoxin expressed by a recombinant baculovirus. Virology 184,7;‘7-780. 106. McCutchen, B. F., Choudary, P. V., Crenshaw, R , Maddox, D., Kamita, S. G., Palekar, N., Volrath, S., Fowler, E., Hammock, B. D., and Maeda, S. (1991) Development of a recombinant baculovirus expressing an insect-selective neurotoxin: potential for pest control, &o/Technology 9,848-852.
Vialard,
24
Arifi
and Richardson
107. Cory, J. S., Hirst, M. L., Williams, T., Hailo, R. S., Goulson, D., Green, B. M., Carty, T. M., Possee, R D., Cayley, P J., and Bishop, D. H. L. (1994) Field trial of a genetically improved baculovuus insecticide. Nature 370, 138-140. 108. Tomalski, M. D. and Miller, L. K. (1991) Insect paralysis by baculovirus-mediated expression of a mite neurotoxin gene. Nature 352,82-85. 109. Tomalski, M. D. and Miller, L. K. (1991) Expression of a paralytic neurotoxin gene to improve insect baculoviruses as biopesticides. Bio/Technology 10,
545-549. 110. Hammock, B. D., Bonning, B. C., Possee, R. D., Hanzlik, T. N., and Maeda, S. (1990) Expression and effects of the juvenile hormone esterase in a baculovirus vector. Nature 344,458-561. 111. Hammock, B. D., McCutchen, B. F., Beetham, J., Choudary, P. V., Fowler, E., Ichinose, R., Ward, V. K., Vikers, J. M., Bonning, B. C, Harshman, L. G., Grant, D., Uematsu, T., and Maeda, S. (1993) Development of recombmant viral insecticides by expression of an insect-specific toxin and insect-specific enzyme in nuclear polyhedrosis viruses. Arch. Insect Biochem. Physiol. 22,3 15-344. 112. Maeda, S. (1989) Increased insecticidal effect by a recombinant baculovirus carrying a synthetic diuretic hormone gene. Biochem. Biophys. Res. Comm. 165, 1177-1183. 113. O’Reilly, D. R. and Miller, L. K. (1989) A baculovirus blocks insect molting by producing ecdysteroid UDP-glycosyl transferase. Science 245, 1110-l 112. 114. O’Reilly, D. R. and Miller, L K. (1991) Improvement of a baculovirus pesticide by deletion of the egt gene. Biotechnology 9, 1086-1089
CHAPTER2
Baculovirus William
Transfer
Vectors
Miguel Lbpez-Ferber, P. Sisk, and Robert D. Possee
1. Introduction The aim of this chapter is to give an overview of the baculovirus transfer vectors currently available, and their applications in the production of recombinant viruses and synthesis of heterologous proteins. Other-chapters in this book provide detailed protocols of cotransfection and selection methods for the isolation of recombinant viruses. Here, we will only refer to these processeswhere it is essential to understand the function of the transfer vector. Transfer vectors have been developed using different baculoviruses. In this chapter, attention will be focussed on Autographa californica nuclear polyhedrosis virus (AcNPV). The other bacul ovirus that has been broadly used is Bombyx mori NPV. The interested readers may refer to Maeda’s work (I-5) or Chapter 14 in this book. These vectors will not be discussed here in detail. We will try to guide the reader through the many options now available for expressing foreign genes using the baculovirus system and recommend the transfer vectors that should be used for particular applications. 1.1. The Role of the Baculovirus Transfer Vector The baculovirus genome is too large to permit easy manipulation for the insertion of foreign genes. However, some reports have been published (6) describing the direct insertion of pieces of DNA into the genome via enzymatic ligation (7) through use of large bacterial plasmids and a transposable element (8), or inserting yeast replication From
Methods m Molecular B!ology, Vol. 39: Baculovms Expression Protocols Edlted by. C. D. Rchardson Q 1995 Humane Press Inc., Totowa, NJ
25
26
Ldpez-Ferber, Sisk, and Possee
elements (ARS and CEN) and a URA3 selection marker into the viral genome followed by homologous recombination with a vector in Saccharomyces cerevisiae (9). These methods are effective, but cumbersome for the novice, and the usual way to construct a recombinant baculovirus is by introducing the desired foreign coding region into a transfer or transplacement vector. This vector contains a bacterial plasmid (usually one of the pUC series), and a portion of the baculovirus genome spanning the gene promoter and transcription terminator to be used for foreign gene expression. The transfer vector must also have a suitable restriction enzyme site after the gene promoter to facilitate insertion of the heterologous sequences.Theoretically, any promoter may be used to drive foreign gene expression, provided that the promoter can be utilized by the transcription machinery of the baculovirus-infected cell. In practice, most transfer vectors have used the polyhedrin or p10 gene promoters because products from these genes are expressed in very large quantities. The use of a transfer vector enables preliminary manipulations to be performed in vitro and subsequently amplified in bacterial cells. The foreign coding sequences are finally inserted within the baculovirus genome by cotransfecting insect cells with infectious virus DNA and the plasmid transfer vector. Recombination between the homologous sequences in the virus and the transfer vector effects exchange of genetic information. Detailed studies to determine the optimum length of these flanking sequencesrequired to drive recombination have not been conducted. The first transfer vectors tended to have very long regions flanking the foreign gene insertion site, which resulted in plasmids >lO kbp in size. More recently, the sizes of the vectors have been reduced to accommodate larger insertions of foreign coding sequences. This is particularly important in the construction of multiple gene expression vectors. Two regions of the AcNPV genome have been used to construct expression vectors. These are the 7.3-kbp fragment generated by digestion with EcoRI (EcoRI I fragment), in which the polyhedrin gene is located, and the 2.0-kbp EcoRI P fragment, containing the p10 gene. Extensive studies to determine which of these locations is optimal for foreign gene expression or recombinant virus stability have not been performed. It has been shown that the p10 gene promoter can function efficiently when inserted at the polyhedrin gene locus. It has also been demonstrated that the polyhedrin gene region is deleted naturally when
Baculovirus
Transfer
Vectors
27
the virus is subjected to serial passage in cell culture at a high multiplicity of infection (10). Many different transfer vectors have be:enconstructed, but a common and logical nomenclature for them has yet to emerge. In this chapter, we have classified them according to the number of gene promoters available for expressing foreign genes and the location of the insertion site within the virus genome. 1.2. The Parental Baculovirus Genome (Linearized or Circular DNA?) It is difficult to consider the range of transfer vectors available for inserting foreign genes into baculovirus genomes without briefly discussing the virus that is to be modified. Early studies to express foreign genes in baculoviruses simply used the wild-type AcNPV. In the case of polyhedrin gene promoter-based transfer vectors, the selection method for recombinant viruses consisted of identifying polyhedrin-negative plaques (see Chapter 6 of this book). More recently, improved methods for the selection of recombinant viruses have been devised by Kitts et al. (11,12) (see Chapter 7 of this book). They describe the use of linearized baculovirus DNA to enhance the proportion of recombinants in the progeny virus from a cotransfection. It had previously been shown l-hat linearized viral DNA was not infectious. This group isolated a recombinant baculovirus with P-galactosidase cloned at the polyhedrin locus and also introduced Bsu36 I restriction sites in the flanking sequenceupstream of the promoter and another Bsu36 I site in the essential 1629 gene:,which is contained in the downstream polyhedrin flanking sequence. The modified viral genome has been named BacPAK6. This viral DNA is linearized with Bsu36 I and cotransfected with a conventional polyhedrin-based expression vector. A process of homologous recombination repairs the deletion incurred by cutting with the restriction enzyme and substitutes the new foreign gene in place of P-galactosidase. New recombinants form white plaques, whereas BakPak6 gives a blue background. Both Clontech (Palo Alto) and PharMingen (San Diego) market this system. Linearized viral DNA containing the same 1629 gene lethal mutation is marketed by PharMingen under the trade name “Bacul oGold” linearized viral DNA. Invitrogen (San Diego) markets another form of linearized viral DNA containing an Sse I site near the beginniqg of the polyhedrin ORF, which does not contain the P-galactosidase gene. Cotransfection with BlueBac vectors yields viral plaques of which 90%
28
Lbpez-Ferber, Sisk, and Possee
are blue in color in contrast to the clear recombinant plaques made with the Clontech and PharMingen systems. Both types of linear viral DNA appear to work equally well. Other transfer vectors, which incorporate a functional polyhedrin gene in addition to the foreign coding sequences, require the use of a polyhedrin-negative virus DNA in the cotransfection. 2. Transfer Vectors for Expression of Single Genes 2.1. Polyhedrin Gene Locus Expression Vectors Vectors which recombine with wild-type AcNPV and replace the polyhedrin gene with a foreign gene were the first transfer vectors to be developed, and have proven the most popular for researchers concerned with the expression of foreign genes in the baculovirus system. Many groups have developed their own versions of these vectors, which differ in only a few nucleotides in the promoter regions; they may also have alternative restriction enzyme sites for the insertion of foreign coding regions (13-22). Other variations include the amount of residual polyhedrin gene coding sequences (out of frame) and the length of the virus sequences flanking the polyhedrin promoter, which facilitate homologous recombination in virus-infected cells. The vectors pBac 1, pVL1392/1393, pEV, and BlueBac II (also known as pETL) are now routinely used and usually yield comparable levels of expression. However, there have been limited comparisons of the relative abilities of various vectors to express foreign genes. We will indicate the most important features of each transfer vector. The integrity of the polyhedrin gene promoter is the most important characteristic of the transfer vector. The early vectors did not retain the complete polyhedrin promoter. For example, pAc373 (13,15) lacks seven nucleotides of the 5’ noncoding region before the translation initiation codon (Table 1). Although pAc373 has been used very successfully to produce many foreign gene products, later examples of the transfer vectors have included the extra seven nucleotides and proven that they are very important for foreign gene expression (15-18,23). A second generation of transfer vectors (Table 1) includes pEV55 (14,24), pAcYM1 (18), pAcRP23 (19), and pAcRP25 (20). All of these vectors retain the complete polyhedrin promoter and give significantly higher levels of foreign gene expression than pAc373. Some vectors have the ATG start codon embedded in a unique NcoI site, such as pAcC4 and pAJcC5 (I5,17) (described in Table 1). Other studies suggest that the
Table 1 Cloning Regions Adjacent to the Polyhedrin Promoter of Some Common Expression Vectors for Single Gene Expression -8
R-~aI/XhoII/SstII/BamHI
-CCcgagatccgcgga~cc..-.-...-----..-.-...--...-......
pAcRPl8
-CCTATAAATCcggatCC...-.-...---.........---..-
pAcRP23
-1 BamliI -CCTAT-Tccgg~tcc.-~-..-.--..---...-..~-~..~.......-...-.-.-.----.-.-.-----.-.-.---.--~..........~~~~.~~CcT
pAcRP25
+2 BamIiI -CCTAT-TATCC~~~~~~.--...--.-......-..-...-........
pAcyM1
-CCTAT-TA~~~~~~~~~---.----....--..-..---..--.....-...
.
-1
+l
pAcC0 pAcC5
%
+177
pAc373
pEv55
.
..-.-.-
._...-._-..-.-.___._...
BamRI
+177 . . . . . .
.
..---..--...--.----.--.-.-.-...........-...------CT
+177 +752 __
__
.-.--.---.------..-...~.~................-.-.~~A~
Bad41
+752 . . . . ..-
.-.....-.----------------------------.----~~A~
-3 NcoI/SmaI/KpnI/PstI/BglII/XbaI/EcoRI/BamHI AAACCTATnnccataggcggcccgggtacctgcagatctagaattcggatcctgatcaccggg-----------------------------------------3 NcoI/BamHI/EcoRI/XbaI/BglII/PstI/KpnI/SmI AAACCTATAAccatagcggcccbvbtgatcaggatccgaattctagatctgcaggtacccggg---------------------------+l BglII/XhoI/EcoRI/XbaI/KpnI
+l
pvL1392 pvL1393 pVL670
pBlueBac2 pAC360
+654
--GGA +654
--------------GGA +631
-CCTAT-TAgatctcgagaattctagatcgat---...----..-----......-.-.--..------.----.-.--....-.--.-------.-.~~~~~TACC
BglII/XhoI/EcoRI/XbaI/KpnI
+631
-CCTAT-TAgatctcgagaattctagatcgat-..-.-.-....--.-..--..--...-----.....-.-----.--~~~
pvL945
..-............-qyl.cCT
+l Ssp I BspM II AAACCTATAAATATtCC'lTCATACCGTCCCACCA~ +1 ARACCTATAAAT~CCGGA-ITA~TACCGTCCCACCA~ +1 AAACCTATAAAT~CCTACCGTCCCACCA~ +1 AAACCTATi=AAT~CGZAlTATTCATACCGTCCCACCA~ +1 AAACCTATMAT~CCGZATTATTCATACCGTCCCACCA~ +l AAACCTATAAAT~CCCACCATCGSZC +1 BamH I
-..---..--.--.----.GGTAccGA
BamHISinaI ggatcttaaggatccacccgggttaagatcc------ - ---+35 BglII/PstI/NotI/EcoRI/XbaI/KpnI/SmaI/BiI ggatcagatctgcagcggccgctccagaattctagaaggtacccg-------------mCCT +35 BamlfI/SmeI/KpnI/XbaI/EcoRI/NotI/PstI/BglII ggatccccgggtaccttctagaattccggagcggccggagcggccgctgcagatcctgatcc~+35 l ** BamH I/Sma I ggatcttaaggatccacccgggttaagattccc---------------------+35 Nhe I BamH I .-.--------------..---.---..-------.--CT Tgctagcggatcc--------BamB I +35
+32
+177
- - ---
-- - ---
- - --mC +177 +177 ---TlTCCT +17-l
---TlT.C +177 +177
cggatcc--..--.--.---.----------.-----------.----.-------...---.-CT
-CCTAT-Tccggatcc..SV40 p,,ly (A) _ Ia&’ _ &p-Jo pro~ter.--..-.--.----....-----.---.-..-.---.-.---...-.G-CC - - .pBacpm1 -CCTAT-~~~gg~~~~g~~~~~...~..~~.~...~.~~~.~~~~~~~.---..~~.~.~~.~~~~----~---.---------.-~
+670
pAcDZ1
-.---.--.---.-G~A~
-7%
30
Lbpez-Ferber,
Sisk, and Possee
first nucleotides of the coding region of the polyhedrin gene may play a role in determining the level of geneexpression (16-l 8). Therefore, some transfer vectors (pVL941, pVL945, and pVL985) (18) retain a variable number of nucleotides after the polyhedrin gene translation start codon by mutating the ATG codon to ATT. However, these vectors did not direct expression at significantly higher levels when compared with pAcYM1 (18). These authors also observed an extra protein in cells infected with recombinant virus derived from pAcYM1, containing the CAT reporter gene. This was consistent with alternative translation initiation at the ACG codon generated by the manipulations to construct the vector. The observation has not been reported by other users of pAcYM1, possibly because a protein initiated at this alternative codon will not be in the correct reading frame. Other vectors, such as pAcRP23 (19) or pAcRP25 (20), in which there is not an ACG in a favorable position should not present such a problem. A similar observation was made with pVL941, where initiation was also observed at the altered ATT codon when the foreign gene was in frame with the polyhedrin leader (2). Thus, it is recommended that the foreign gene be cloned out of frame from the altered ATT or ACG initiation codons. Recently, Sanchez-Martinez and Pellet (26) published a report describing a vector that conserves all of the structure near the polyhedrin ATG, so that the protein start codon of the foreign gene replaces the original ATG without the introduction of any other residues. A higher level of expression was observed in comparison with pAc373, but vectors, such as pAcYM1 or pAcRP25, were not included in the study. Rankin et al. (23) produced a series of linker scanning mutations of the polyhedrin gene promoter by inserting a Hind III linker in lieu of various regions of the promoter. They observed that one of the mutants increased the activity of the original promoter by about 50%. This modified synthetic polyhedrin gene promoter was designated PXIV. A series of transfer vectors incorporating this modification has been developed (27,28). Most baculovirus transfer vectors based on the polyhedrin gene locus are large plasmids (e.g., pAcYM1,9.2 kbp, pVL941,9.8 kbp). This was a consequence of the convenient DNA fragment, generated by EcoRI (7.3 kbp), used asthe starting point for constructing the vectors. The large flanking regions either side of the polyhedrin gene locus also ensured that homologous recombination took place in virus-infected cells. Other transfer vectors have a reduced size (e.g., pEV55,6.2 kbp; pPAK1, 5.5 kbp),
Baculovirus
Transfer
Vectors
31
which may facilitate easier insertion of foreign DNA into the plasmid. Variable amounts of the 3’-coding region from the polyhedrin gene have been conserved and the expression vector generally contains the polyhedrin termination and polyadenylation signal (Table 1). A further refinement has been the integration of the Ml3 phage or Fl phage origins of replication into some transfer vectors, thus enabling the production of single-stranded DNA in the bacterial host, after superinfection with Ml3 K07. Such vectors include pAcC129 (29), pPAK1 (Clontech, Palo Alto, CA), and pJVNheI(30). These plasmids are particularly useful when performing a series of mutagenesis steps on a foreign gene. 2.2. Transfer Vectors Utilizing Alternative Gene Promoters The use of other baculovirus gene promoters as expression systems is complicated by uncertainties whether their normal gene products are dispensable for virus replication. For example, most of the late gene products identified to date serve as structural virus proteins, and it is unlikely that they can be deleted from the virus genome. However, there is interest in using the promoters associated with these genes for synthesis of recombinant proteins at earlier times in the virus replication cycle. Foreign gene expression prior to the very late phase may be advantageous for efficient posttranslational modifications. The development ofgenetitally engineered virus insecticides may also benefit from the use of expression vectors that can produce insecticidal proteins at an earlier stage in the virus infection of the insect. To circumvent the problems associated with using the late gene loci to insert foreign sequences, the polyhedrin gene locus has been employed as a site for adding a copy of the preferred late gene promoter to the virus genome. This approach preserves the native gene and leaves virus replication unaffected by the manipulations. Hill-Perkins and Possee (31) replaced the polyhedrin gene promoter with the basic protein gene promoter to derive pAcMP1. The basic protein is a late gene product that associates with virus DNA within the nucleocapsid. It is produced in the late phase of gene expression, between about 8 and 24 h postinfection. Using pAcMP1, it was shown that P-galactosidase could be produced at the same time as the normal basic protein in virus-infected cells. Thiem and Miller (27) produced a similar vector that replaced the polyhedrin
32
Lbpez-Ferber, Sisk, and Possee
promoter with the p39 capsid gene promoter. The vector was tested using expression of the CAT gene. A vector using the p39 promoter instead of the polyhedrin promoter is currently marketed by PharMingen as pAcJP1. In the following section, some other vectors that use ~10, basic synthetic promoters will be discussed. 2.3. Transfer Vectors for Positive Selection of Recombinant Viruses Elsewhere in this volume, protocols will be described for the efficient selection of recombinant viruses after cotransfection of insect cells. Virus DNA is linearized at the desired site of insertion of the foreign gene, prior to cotransfection of insect cells (Chapter 8), and the plasmid transfer vector carries a marker gene that facilitates positive selection of the
recombinant virus. One group of these vectors contains the Escherichiu coli ZucZ gene under the control of the p10 promoter (pJVNheI[30], pPl0 [27]), ETL promoter (pJVETL [32]) or the Drosophila heat-shock promoter (33). Recombinant viruses may be identified as blue plaques in the presence of X-gal. Use of these types of vectors is discussed in Chapter 9. All of these vectors retain the characteristic of single-gene expression at the polyhedrin gene locus; the extra promoter and marker gene serve as an aid to recombinant virus selection. Thymidine kinase selection has also been used to select recombinant baculoviruses (34). Baculoviruses do not normally contain a thymidine kinase gene (35,36). However, recombinant virus expressing the herpes simplex virus thymidine kinase does not grow in Sf9 cells in the presence of the nucleoside analog Guanciclovir (100 CLM).Recombination with a polyhedrin locus expression vector produces virus that can grow in the presence of the inhibitor (34). The polyhedrin gene itself has been used as a selectable marker, mainly for production of recombinant viruses to be used as biopesticides. Recombinant viruses are selected as polyhedrin-positive plaques against a background of polyhedrin-negative parental viruses. In these vectors, the foreign gene expression is driven by a modified polyhedrin promoter, p10 promoter, basic protein promoter, or synthetic hybrid promoters. Weyer et al. (37) developed two vectors, pAcUW2A and pAcUW2B, in which the polyhedrin promoter drives the expression of the polyhedrin gene, and the p10 promoter is used to express the foreign gene. pAcUW2B has been successfully used by Stewart et al. (38) and by
Baculovirus
Transfer
Vectors
33
McCutchen et al. (39) to increase the insecticidal activity of Ac:NPV by expressing Andoctonus australis Hector insect toxin. Later modifications of this vector to reduce the size and permit single stranded DNA production in E. coli have been made (pAcUW21). A further development is the use of an earlier promoter, the basic promoter, or a synthetic hybrid (tandem basic protein and ~10) promoter, in lieu of just the p10 prolmoter, in order to obtain an earlier expression as in the vectors pAcMLF7 or pAcMLF8 (M. Lbpez-Ferber, unpublished). In a similar approalch,some transfer vectors developed by Wang et al. (28), pSYN XIV VI+ and pSYNVI+wp, also retain the polyhedrin gene. Expression of the foreign gene is controlled by the polyhedrin-modified promote:r PXIV, the Psyn synthetic promoter (40), or from a tandem synthetic hybrid of both. Tomalski and Miller (41) used the PSynXIV tandem promoter to drive the expression of the Pyemotes tritici mite toxin, in a recombinant polyhedrin-positive virus, and described its potential use as a bioinsecticide. The selection of recombinant viruses must be done following recombination with a polyhedrin-negative or a polyhedrinnegative lZacZ-positive parental virus. 2.4. ~10 Gene Locus
The p10 protein is dispensable for virus replication in insect cells (4244). However, the absence of a visible phenotype for the ~10 protein in baculovirus-infected cells has made the construction o Fexpression vectors at this locus difficult. Weyer et al. (37) constructed a transfer vector for the insertion of foreign coding regions having their own translation initiation codon. This vector has to be used with a virus containing the ZacZ gene at the p10 locus; recombinant viruses are selected by the loss of the blue plaque phenotype. When used in conjunction with the linear virus DNA method (12), the isolation of virus recombinants is easy. Vlak et al. (45) adopted an approach for the insertion of foreign genes at the p10 locus that was analogous to the method used for positive selection of recombinant viruses with genes substituted at the polyhedrin gene locus. In the transfer vector pAcAS3, the ZacZ gene was inserted in association with the Drosophila melanogaster hsp 70 promoter, which is constitutively expressed in Spodoptera frugiperdu infected cells. This provided for the facile selection of recombinant viruses by staining plaque assays with X-gal.
LcSpez-Ferber, Sisk, and Possee 2.5. Fusion Vectors Fusion vectors are useful for the expression of a coding region that lacks a translation initiation codon. For AcNPV, these have all utilized the polyhedrin gene promoter at the polyhedrin gene locus. Such vectors are probably now becoming redundant with the advent of the polymerase chain reaction, which can be used to quickly tailor a foreign gene to incorporate an ATG codon. One feature of these vectors that may still be useful, is that some foreign proteins synthesized as a fusion product are more stable in baculovirus-infected cells. If the recombinant gene product is only required for the production of antibody, the additional sequences may not present a problem. Recently, some vectors that produce fusion proteins with the S-glutathione transferase have been developed. In these vectors, a protease cleavage site is present at the junction of the fused protein, allowing the recovery of an almost intact protein (46,47; R. F. Weaver and M. Lopez-Ferber, unpublished data [pAcRW4]). Depending on the protease (e.g., thrombin) used, some GST residues may remain attached to the cleaved recombinant protein. The advantage of the system is that proteins maybe easily purified from virus-infected cells using glutathione affinity chromatography. In a similar fashion, histidine residues have been fused to either the carboxy or amino terminal ends of a number of recombinant proteins. A (pBlueBacHis) vector specifies six histidine residues that can be fused to the amino terminus of a recombinant protein. It has been marketed by Invitrogen. (His),-tagged 12-lipoxygenase has been expressed and isolated from infected insect cells using nickel-ion-chelation chromatography followed by imidazole elution (48). Fusion vectors that add amino terminal secretory sequences to the recombinant protein in order to facilitate their secretion have also been developed and marketed. The human placental alkaline phosphatase (pPBac), honey bee melittin (pVTBac, pMBac), myelinassociated glycoprotein (see Chapter 21 of this book), ecdysteroid UDP glycosyltransferase (49), and the envelope glycoprotein gp67 (pAcGP67 A,B,C) (49). When the HIV-l gp120 amino terminus was replaced with these sequences, a 6- to 20-fold increase in expression and secretion of gp120 was observed (49). pPBac and pMBac are currently marketed by Stratagene (San Diego, CA), while pAcGP67 A,B,C plasmids are sold by PharMingen.
Baculovirus
Transfer Vectors
35
3. Transfer Vectors for Multiple Gene Expression The ability to produce two or more foreign proteins in insect cells is attractive in studies to investigate protein structure-function and to produce subviral particles for vaccine development. 3.1. Polyhedrin Gene Locus The first vector of this type was constructed by Emery and Bishop (50), where a duplicated polyhedrin promoter and transcription termination region was inserted upstream of the native polyhedrin gene. This vector, pAcVC2, was used to produce the lymphocytic choriomeningtitis virus nucleoprotein in addition to the polyhedrin protein. Furthe]. developments derived pAcVC3, where the native polyhedrin gene coding region was deleted to permit its replacement with a second foreign gene coding sequence. pAcVC3 was used to produce Bluetongue virus corelike particles (51) and secreted antibodies (52). pAcVC3 has been largely superseded by the duo-expression vector pAcUW3, where a copy (230 bp) of the p10 gene promoter was inserted upstream of the polyhedrin gene promoter, in association with SV40 transcription termination signals (53). The amount of repetitive sequences has been kept to a minimum in this vector and should preclude some of the virus instability problems associated with pAcVC3 (D. H. L. Bishop, personal communication). When pAcUW3 was first tested by expression of the influenza virus hamagglutinin and neuraminidase genes (5.3), the recombinant virus was shown to be stable for at least 12 passages in cell culture. The transfer vector pAcUW3 has been subsequentl!l modified to reduce the amount of virus sequences flanking the two transcription units and to incorporate the Ml3 intergenic region for the production of single-stranded DNA (pAcUW3 1, M. Lopez- Ferber, unpublished data). Even more recent multiple gene expression vectors (pAcAB and pAcAB4) were developed by David Bishop’s laboratory and used to express three or four Bluetongue virus proteins, which in turn yielded Bluetongue virus-like particles in insect cells (54). Another set of vectors (p2Blue, p2Bac, pAcUW51) contain both a pollrhedrin promoter and a p10 promoter that lie adjacent to each other, but transcribe foreign genes in opposite directions. The p2Blue and p2Bac vectors are marketed by Invitrogen (San Diego), whereas pAcUW51, which was developed in Robert Possee’s laboratory, is sold by PharMingen. These vectors have been used to synthesize a humlan IgG
36
Lrjpez-Ferber, Sisk, and Possee
from insect cells (55). Wang et al. (28) have also constructed a transfer vector mediating expression of two foreign genes in insect cells. This vector (pSynXIV VI-) used a polyhedrin promoter and a tandem Syn/ PX14 synthetic promoter to drive the expression of two different genes. 3.2. pl0
Gene Locus
Expression
Vectors
To date, the reports of dual expression at the p10 gene locus have been confined to the use of a marker gene (e.g., ZacZ) to ease the selection of recombinant viruses in the plaque assays subsequent to cotransfection. Transfer vectors analogous to pAcUW3, capable of synthesizing two experimental gene products, have not been reported in the literature. One such vector was developed in our laboratory. This has been designated pAcUW41 and is marketed by PharMingen. It consists of a copy of the polyhedrin gene promoter inserted in tandem with the p10 gene promoter. The transcripts initiating at the p10 promoter are terminated by SV40 signals inserted between the two promoters. The transcripts initiated at the polyhedrin promoter are terminated by the plO- gene signals. The vector also has the facility to produce single-stranded DNA in bacterial cells. Similar vectors (pAcUW42 and pAcUW43) with polylinkers adjacent to the p 10 promoter are also available from PharMingen. 4. Tandem 4.1. PoZyhecZrin
Hybrid
Gene Promoters
Gene Locus
Expression
Vectors
Transfer vectors are available to synthesize proteins utilizing gene promoters active in the very late phase of virus gene expression, such as the polyhedrin and ~10, or the late phase, such as the basic protein or capsid gene promoters. Some attempts have now been made to combine these regulatory elements to either increase the level of protein production or to expand the time in which the protein is synthesized. Above we have discussed the tandem vectors using Psyn and PXlV (28), and pAcMLF8, that use a tandem hybrid of basic and p10 promoters. A transfer vector pAcMP2 with the basic protein gene promoter arranged in tandem with the polyhedrin gene promoter and driving expression of the 1acZ gene was constructed by Hill-Perkins and Possee (31) The recombinant virus derived from this vector demonstrated P-galactosidase activity in both the late and very late phases of virus gene expression in virus-infected cells. This vector was later modified by M. Lopez-Ferber to give pAcMLF 2.
Baculovirus
Transfer Vectors
37
5. Transfer Vectors Based on Other BacuZoviruses The development of vectors using other baculoviruses has been restricted by the availability of suitable cell lines to support viral growth; therefore, it is not surprising that the other baculovirus that has been used for expression is the Bombyx mori nuclear polyhedrosis virus. Maeda et al. (3,4) described the use of the polyhedrin gene promoter of Bombyx mori NPV in the expression of a-interferon. Their original vector, p89B310, lacked 18 bp from the translational start in the S-flanking sequenceof the polyhedrin (4). Successive improvements allow the development of better vectors, with the whole promoter sequence(5). In analogy to AcNPV transfer vectors, there are also Bombyx mori vectors that facilitate expression of proteins as fusions with the polyhedrin. Readersare referred to reviews concerning the Bombyx mori-based vectors (1,2). Devauchelle and his group (56,57) developed their vectors on Gulleriu melonella NPV, which is almost identical to Autographa califomica NPV. This virus also g,rows in Sf9 and Sf21 cells. Baculovirus expression vectors that infect cell lines and two species of spruce budworm (Choristoneuru biennis and CrZoristoneurufimiferuna) are also being developedby the Forest Pest Management Institute in Sault Ste. Marie, Canada(Basil Arif, personal communication). 6. Summary of Commonly Used Baculovirus Expression Vectors Many of the vectors described in this chapter were briefly presented and surveyed at the Workshop on Baculovirus and Recombinant Protein Production Processes (22). A number of these vectors can now be acquired from commercial suppliers such as Invitrogen, Clontech, PharMingen, or Stratagene(San Diego). Restriction maps of the most commonly used vectors that have been described in the preceding pages are presentedin Figs. l-32. The DNA sequences from these plasmids are available from the suppliers, individual laboratories, the authors of this chapter, and the editor of this book. Cloning sites for insertion of foreign genesfrom a number of vectors are presented in Table 1. Restriction maps of the polyhedrin and p10 genes with their flanking sequencesare shown in Fig. 33. Since most baculovirus expression vectors are derived through the incorporation of these genes and their respective flanking sequencesinto DNA pl,asmids, the compete DNA sequencesand the coding regions for the pol*yhedrin and p10 genes of Autographu califomicu nuclear polyhedrosis vuus are presented in Figs. 34 and 35. This information (text continued on p. 59)
Ldpez-Ferber, Sisk, and Possee
38
AWN
I 8488.
, Hl”d P,amdd
name
Plasmid
aloe
8232
Constructed
by
Malsuura
R D
Poasee.
Polyhedrm
promoter.
uptoAUG
,418,
bp
Y
Commsnts~Refsrence,s pnlyhedrln
SnaB
11, 4530
pAcYM1
which
0amH gives
I cloning better
H A
locus
Overton.
replacement
611% full polyhedrlo
expression
than
D H L vector,
mRNA
pAc373.
J
S,shop
single non
Gen
gene
coding
“walagy
expressm,
leader
sequence
67 151.5
Sal
,529
($987,
I 2947
k
/ Ch
tnsp
I 5139 Hind
11, ,,,a\ Aat
P,asm,d
name
Plssmld
.Ize
Cawtruclad
9780 by
polyhedrin
G E
433,n,na
111 4110
I, 5022
bp Smith
and
Polyhedrln
promolsr, next
,111
Kp” ’ “” ala0 I4741
pAc373
Comments,Refermses leader
’
10 AUG
SamH
I olonlng
which
redwas
MD
Summers locus
slle.
replacment mtss,ng
exprewon,
veclor.
g nucleotldes Prw
Nat,
amgle I” mRNA
Acad
SC,
gene non USA
expression, coding BP
8404
84gg(,gS5)
Figs. l-32. Restriction maps of commonly used baculovirus expression vectors. Basic features of each vector are described in the individual legends.
Baculovirus
Transfer
Vectors
Pvu AhN
EcaR
17194
-.-/ I 6034 c,a
S”&
I 5711 Hind Aat
by
R D
Possse
Comment*,R~‘e,~nses polyhedrln BamH
and
Polyhedri”
promote,. I cloning
II ““‘9,
I 904q
\
.-’ c,a
Construetsd
39
full she,
NW,
mRNA Acid
,I, 51OP
I4739
0amtl
,409s
)
II 5029
SC
Howard
IX”8 leader Research
with
replaoemsnt addlllonal 15
19223
“*&Jr. 3’ sequence 10249
(1997)
smgle
gene
al polyhedrin
expre.5skm, gene
40
Ldpez-Ferber,
PVU
II
PIamId
“ams
Plasmld
size
Constructed
pVL841 9822 by
VA
bp Luckow
Comm~nts/Refsrencss palyhedrl” mRNA
promoler, arf~ fused
PlanmId
“mna ‘1~.
AUG
MD
to foreign
ORF.
Summers
locus
of polyhedrl”
replacement
gene
EamH
wctor.
mutated
I cloning
to ATT EWJ. Virology
snngle and 170
35
gene
expression,
“ucleotldes
of polyhedri”
31 39 (1989)
pVLl392/1393 by
9660 bp ” Luck”“,
CommenWReferencss polyhedrl”
and
Polyhedrln
PleamId Constructed
61tes.
Sisk, and Possee
promoter. BWTechnology
and
MD
Polyhedri” All
leader
6 47 55
Summers
locus from
(1988)
(marketed
replaCema”
polyhedron
mRNA.
by INWTRCGEN
vector,
single
forward
and
and
gene reverse
PHARMINGEN)
expression, mulliclonlng
Baculovirus
Transfer
ma
Ill 2999 Ban
Plasmld
name
PIesmId
e,ze
Construoted
by
polyhsdrln
PA
larger
genes
I/N.?,
I luslon
2693
\ I 2893
bp Km6
(marketed Polyhedrln
promoter, size
Act
/
I, 2825 NW
pPAKl 5462
Commants/Rslsrancss plasmid
41
Vectors
due
EamH to diminished
I” E COli
I cloning polylwrln
by locus 6118,
CLONTECH) replacement 11 orl
llanklng
lo enable sequences
veofor 8ynlhesis facll~tates
single
gene
expression
of ssDNA. cloning
sma,ler of
42
Ldpez-Ferber, Sisk, and Possee
Baculovirus
Transfer
Hmd
Vectors
Ill 9746 *a,
I 9704
Ptaemld
name
PIa8mld Co”.tr”cted
else
pJVPt0 13918 bp by C Rlchardean
CommadslRstwonces beta
43
11 OR,
lntervlrology
J
Polyhedrln
galsotosidase
@DNA
and
eelecllo”, Nhe 34
213
loous
strong
I and
BamH 227
VIalard
Lao,?
, cloning
replacement expmss,o”,
sites,
sector. ATT
no gap
single
polyhedrin
between
P,o
gene
expression.
mRNA and
leader.
polyh
promo,e,s,
(1892)
Hind
Ill
13988,
pJVETL
(ftluettac
I)
EooR
BmlH
PlasmId
name
PI.mnld
~12.
pJVETL
Conetruoted
bp
CD
Richardson
by
34
PI3
Nhe 227
(BlueEac
13980
CommentsfReiarenoo~ beta galaotosldase 11 ORI,
I7774 I Nhe I 7768 EOOR v 7839
8eIectlon.
I and (lesn)
BamH
88”36
I 7107
I)
and
M
Lalumlere
(marketed
Polyhedr,” lx,,8 replacement wsk la02 expression ATT I cloning
I 4328
sites.
no gap
between
by
vector, polyhedrin ETL
and
INVITROGEN) single mRNA
polyh
gene expression, leader, ssDNA
promoters.
lntewwology
Ltjpez-Ferber, Sisk, and Possee
a, I 1047 al I 1332 P72 -?
ECOR
v 5744
P,a.m,d
“*me
pETL
,Bl”eBac
PISSrn,d
.,.a
10271
bp
ConEdrusted
by
CD
Rlchardso”
CommenisIRsfersnoss with
polyh.
leader.
and
Polyhedrm
beta
Nhe
I,)
ga,actos,dase
I and
34 213.227
M
,ocus
se,ect~o”.
BamH
I clonmg
replacemenl
week
E&S.
veclor,
single
exPress,on,
S,ZB
favors
ATT clonmg
gene
expresslon
polyhedrin
mRNA
in E co,,,
,n,ew,m,ogy
(lgg2,
c,a
I 7424
P,aam,d
name
Plaamld
8129
Conslruciad
Nhe 227
$ph
I 235
pP10
C
bp D
Richardson
Polyhedrm
galactoadase 213
‘“‘“‘,
I
10200 by
11 ORI.
”
-7
CommenfslRsfersnsss
34
IacZ
8maller
sa’
be,a
Lalurlwre
se,ecl,a”. I and (1992)
BamH
slrong I cloning
and
locus ,acZ .stes,
M
Mumere
replacement expressmn. smaller
vector AT,
sue
favors
polyhednn donrng.
angle
gene leader,
lnterwrology
expression, ssCNA
Baculovirus
Transfer
Vectors
45 pph
CommenlsIRelarsncss beta
Polyhedrin
galsctosidase
signals,
selecbon.
multlcloning
IOCUB
polyhedrln
bite
for
foreign
replacement
and gene
I 230
PlO
insemn.
vector.
promolers, derived
SV40 from
single
gene
and
polyhedrln
pETL
pBlueBacHI~~
Em13
I Nco
Plasmid
nsma
PIa.mid
size
Constructed
I Ps, , Sg, I, EamH
pSlueBacH16 9302 by
CammantslAeforancss ~xp,ess,o”. beta ,us,on
LO (H,s)6
I EK (His)6
expresslon polyadenylation
(SlueSacll)
A,B.C
AT
A.0.C
bp
INVITROGEN
galaclosidase tag
wkh
Polyhedrin lows replacement selsotlon, polyhedrin and e”,e,ok,nase
(EK)
proleolylic
vector, single ETL promoters, 6118. 3 rsadmg
foreign amino frames
gene terminsl (A.E.C)
46
Ltjpez-Ferber, Sisk, and Possee
Nde
PlasmId Plawllld
I
ne.me s,ze
Cowlrucled
PlasmId
nams size
Kpn
pAcJPl 19749 by
PInamId Constructed
pAcJP1
PHARMINGEN
pSynXtV by
I 5495
bp
VI+
5947
bp
X
Wang.
BG
001
and
L K
Mdler
Baculovirus
Transfer Vectors Hind
47
III 9853
Pv Nde .Bs,E
QQ53
/ Phrnld
“ame
Planmld
slza
Cone.truoted
Kp”
I EooR
PVT
Teswr.
T
Verne,.
Polyhedrin
promo,er.
11 OdQin
I No, I Ss, I Pet I Sma
IOr "?akl"Q
foreign Si"Qb
gene
ORF
BIranded
PYU I 132QO-
soa
D IOCUB
I AYlYSlYVVMFVLAVNVLFK>
Thomaa replacement
fused
DNA,
Hind
PVU
I SamH
bp
D
CommanWReisrsnces polyhedrin
bp
sac
QQ53 by.
I Nhe
II 9.23
veclor.
to me,i,,ln
Gene
dQ”d
93 177-183
single peptide
gene
expression.
IO lava,
soo,e,,on,
(1991)
,,I
II 13824
II 1,731
sv‘lt
t term
EC&l v 4003
&%4% r
/
Human Alkaline
Phosphatase
\
P””
Signal Peptide
EamHIEGFs,ullerSmsIPNEEEVPIIGLSLOLRLGLLLLLLLMCPGGATGNheI P,.‘mld
name
P,asm,d
s,ze
Construoted
by
pPSsc 14138
bp
W
Lernhard,
(STRATAGENE)
J
Sm,
Chem
I, 4327
EcoR I4331 Pv” II 4680
(I” press)
Lbpez-Ferber, Sisk, and Possee
48
Pvu
II 1149
/
t.Mntln
signa,
PeptIde \
,st”l,erSma,GPRPDAY,YS,Y”“MF”LA”N”LFKATGNhe,
Phmlld
name
PInamid
size.
ptmac 13947
bp
Commmt.dReferenoea palyhadrln signal
Polyhedrln
promote, p%pbde.
beta
5ma
locus
gatactosidase
,/BamH
replacement
8elecbon,
I cloning
sne
Wllh
,ore,gn removeable
vector. ORF
alngle (used
gene 10 the
12 ““oleobde
expression. me,,,,,n
stuffer
sequence
pAcRW4
I AQ,
,I
P””
I, 4929
Sal
P,aunld
name
Plnsmld
elzn
I4913
pAoRW4 8525
bp
7 HIM
I III 4391 r\ c,a I 427
\ Aa, th”d
,I 3263 111 3345
Baculovirus
Transfer
Vectors
49
GP67
PIasmld
name
PlasmId Con.b”oted
slza
pAoGPB7 9760 bp by R Possee
A.B.C and
D
Whop
(marketed
by
,lled
Con.,,uo,.d
by
M
Lopez
Co,,,,,,~n,~,R~fs,ene~s polyhedrin Rgl
I, Elonlng
and
Ferber
U
Polyhedrln PI0 s,,**
promolers,
leads,
SV40
Weyer. locus terminator,
and
Xho
PHARMINGEN)
I 1
/6Ph
,236
R D
Possse
replacement 11 ORI
“eoto,. for
making
(marketed double ssDNA.
by PHARMINGEN) gene BamH
expresalon. I and
Lbpez-Ferber, Sisk, and Possee
50 Nde
I 9637 \
Baculovirus
Transfer Vectors
AlwN
I 6471.
51
_
-Xba
I 2’1 071 SBO I,
Nar I c,a
$I;,,’ I 4339 Nar I 4198 b
A,, II Aac I Aal
II 3844 Kp”
\ SnaB
Plaamld
name
Ptaemld
.,I. by
Lh” II Ill*” sty I Ei, Hl”d 111
\
INVITROGEN
Polyhedrin
expreaslon
s,gna,s.
\
bp
CommanttlRsInrencsa gene
I.3383
p2Ssc 7139
Constructed
‘3183
vector,
locus
polyhedrin
2 m”ltblo”~“g
and
replacement
PI0
expression
promoters.
SV40
vector.
and
double
polyhedrln
polyadenylatlon
eit.35
EcoRl
10098
v40 Terminator aa I 5438
Sam”, Plarmld
name
P,sm,,ld
&a
Conmtrueted
4990
10096 by
R 0
one
polyhedrl”. Aclds
I 4955
Xba
I 4709
S,u
I 4704
Sgl
II 44.54
EcoR,
bp Poesee
CommentsIRefetenc~~ NW,
Sma
pAoAB3 (marketed
Palyhedrin two Res
2,
PI0 1219
prorno,er8. 1223
by
PHARMINGEN)
IOCUB
replacement
3 olonlng
sites.
vector. polyhedrin
triple and
SW0
gene
expression,
terminators
4448
Ldpez-Ferber, Sisk, and Possee
52
%bI’1o78
Nar PleamId
nams
PlasmId
size
Constructed
I 2053
pAcUWl 4521 by
U
bp Weyer
and
R D
Possee
(marketed
by PHARMINGEN)
EcoR
I
El Asp
718
1100
Baculovirus
Transfer Vectors
53
pAcUW42143 7137
t.p
p’“uz--
54
a
I
Age I
Hmdl I I BamHl
53
56
63 62
SnaBl 47 44
49
Hlndll
Kpn I Hlndl I I 42 40L
25
T
I& 0 s =
:; 31 30
23 E 205 I It 17 16 13 1 25 12 & 081 071 061
803
:
ii
Ltipez-Ferber, Sisk, and Possee
Bell . Hind1 I I . B IIIWIMI ’ P&xl.
d
MM
19
22 21
092 060 074
g/s
38
Bell Sphl
Bst Eli sotl I XClll
058 045
Sall Hmdl I I:
Pmll . BstXl * Sal1 -
36 35 34
m
34
xhol
Pacl Agel
0
14
EcoRl
Baculovirus
Transfer
Vectors
TRANSLATED -190
SEOUENCE
OF POLYHEDRIN -170
-180
*
55
*
GENE -160
*
-150
*
GGT CTG CGA GCA GTT GTT TGT TGT TAA AAA TAA CAG CCA TTG TAA -140
-130
l
-120 *
ACT AAT AAT
CAC AAA CTG GAA AAT
-80 *
ATA
*
-70 *
ATT AAA ATG ATA
GTC
-90 EcoRV * GTT GCT GAT ATC ATG GAG *
TAT CAA TAT ATA -50 *
-60 *
ACC ATC TCG CAA ATA
GCA CAA
-100
-110 *
*
TGA GAC
AAT AA0
-40 *
-30 *
TAT TTT ACT GTT TTC
GTA 4C4
mRNA.... -20 *
GTT TTG TAA
1
-10 *
TAA AAA AAC CTA TAA
40
20 *
t
30
w
ATA ATG CCG GAT TAT TCA TAC CGT CCC ACC ATC M P D Y S Y R P T I Polyhedrin.... 60
50
*
10
*
*
*
GGG CGT ACC TAC GTG TAC GAC AAC AAG TAC TAC AAA AAT GRTYVYDNKYYKNLGAVIK
70
80
l
*
TTA GGT GCC GTT ATC A4G
90 100 110 120 130 140 * EBB1 l * * * * AAC GCT AAG CGC AAG AAG CAC TTC GCC GAA CAT GAG ATC GAA GAG GCT ACC CTC GAC N A K R K K H F A E H E I E E A T L D 150 *
160 *
170 *
180
190
200
* BarnHI * PDUldI * CCC CTA GAC AAC TAC CTA GTG GCT GAG GAT CCT TTC CTG GGA CCC GGC AAG AAC CAA PLDNYLVAEDPFLGPGKNQ 210 220 230 * t * AAA CTC ACT CTC TTC AAG GAA ATC CGT AAT KLTLFKEIRNVKPDTMKL 260 *
270 *
280 *
GTT GGA TGG AAA GGA AAA GAG TTC V G W K G K E F 320 *
AGC TTC S F
330 *
ccc P
ATT I
240
250
*
GTT AAA CCC GAC ACG ATG 290 t
300 *
350 *
G;
310 *
TAC AGG GAA ACT TGG ACC CGC TTC Y R E T W T R F 340 *
'Hind111 4AG CTT
360 *
ATG GAA GAC M E D 370 *
GTT AAC GAC CAA GAA GTG ATG GAT GTT TTC CTT GTT GTC AAC ATG V N D Q E V M D V F L V V N M
Fig. 34 (continued on next page). The complete DNA sequence of fhe polyhedrin gene and its neighboring flanking regions from Autographa californica nuclear polyhedrosis virus. The predicted open reading frame representing the polyhedrin protein sequence and principal restriction sites are indicated.
56
Ldpez-Ferber, Sisk, and Possee
380 390 400 * l * CGT CCC ACT AGA CCC AAC CGT TGT TAC I&A RPTRPNRCYKFLAQHALRC
410
420
*
l
Tl-C CTG GCC CAA CAC GCT CTG CGT TGC
430 440 460 470 480 450 * * * * l * GAC CCC GAC TAT GTA CCT CAT GAC GTG ATT AGG ATC GTC GAG CCT TCA TGG GTG GGC DPDYVPHDVIRIVEPSWVG 490
500
510
520
530
540
l
l
l
t
*
t
AGC AAC AAC GAG TAC CGC ATC AGC CTG GCT AAG AAG GGC GGC GGC TGC CCA ATA SNNEYRISLAKKGGGCPIM 550 560 570 580 * * + * AAC CTT CAC TCT GAG TAC ACC AAC TCG T-TC GAA CAG TTC N L H S E Y T N S F E Q F
ATG
590
600
l
*
ATC GAT CGT GTC ATC TGG I D R V I W
610 630 640 620 650 * * l * * KpnI GAG AAC TTC TAC AAG CCC ATC GTT TAC ATC GGT ACC GAC TCT GCT GAA GAG GAG GM ENFYKPIVYIGTDSAEEEE 660 670 680 * l * ATT CTC CTT GAA GTT TCC CTG GTG TTC ILLEVSLVFKVKEFAPDAP 720 730 * * CTG TTC ACT GGT CCG GCG TAT L F T G P A Y
890 *
GCG TCT TTA
840 * TTA TAA TCT
710
*
*
AAA GTA AAG GAG TTT GCA CCA GAC GCA CCT
800
810
*
*
TAT
770 * TAA
820 SnaBI
*
TTT AAT
AAT
850 860 870 880 * * * * TTA GGG TGG TAT GTT AGA GCG AAA ATC AAA TGA TTT
TCA
900 *
TAT CTG AAT
700
*
740 750 760 * t * TAA AAC ACG ATA CAT TGT TAT TAG TAC ATT
780 790 * l GCG CTA GAT TCT GTG CGT TGT TGA TTT 830 * TCA TTA AAT
690
910 *
TTA AAT
ACA GAC AAT TGT TGT ACG TAT
920 *
930 l
ATT AAA TCC TCA ATA GAT TTG TAA AAT
CGA
Fig. 34 (continued).
940 t
AGG TTT
Baculovirus
Transfer Vectors TRANSLATED
-360
-350
GAA
-290
t
AAA TGG CGA -240
ATT
CCC AAC ATG
-270
TCC GW
-230
-220
-170
ATT
-150
-100
-80
-40
-30 *
PacI*
-20 *
TTA AAA TAC TAT ACT GTA AAT 20 *
*
30 *
TTT ACA ATC ATG TCA AAG CCT AAC GTT TTG ACG CAA ATT TTA GAC GCC M S K P N V L T Q I L D A Pi0 protein.....
40 *
50 *
60 t
80 *
70 *
90 *
GTT ACG GAA ACT AAC ACA AAG GTT GAC AGT GTT CAA ACT CAG TTA VTETNTKVDSVQTQLNGLE 100
110 *
*
GAA TCA TTC E S F
GAA
t t CCC AAC ACA ATA TAT TAT AGT
10
*
t
-90
t
ATT C?A
-50 *
1
-14c *
ACT TAC AAC AAG GGG GAC TAT
ATT ATC AAA TCA TTT GTA TAT TAA
*
l
AAA ACG CCA AAC GCG TTG GAG TCT 'XT
-110
-10 TAC ATT TTA
-190 l
t
-60 *
t
-200 *
-160 l
-120
-250
*
-210
TAC AAA GAT TCA GAA ATA CGC ATC
-70 *
-260
*
GTT
TAC GGG ACT GTG CAA TTG CCG TAC GAT AAA
* NBiI * ATT ATG CAT TTG AGG ATG CCG GGA CCT TTA
A-A IllRNA.....
AAG GTG CTG AAC GGC GTC'CGT
t
l SphI * XhOI AAA CAG CAT GCG CTC GAG CAA GAA AAT
t
-310 t
-280
*
TAA
-320
CAG CTG TCG GGA CAT Ak
-180
-130
-330
t
t
GTG CTA TTT
GENE
*
CGA GAG GCG 'ICC
-300
OF PlO
-340 l
*ECORI
GA ATT
SEOUENCE
57
CAG CTT Q L
160
120 *
130 *
TTG GAC GGT T-X L D G L 170
180
230 *
140 *
150 *
CCC GCT CAA TTG ACC GAT CTT AAC P A Q L T D L N
* * BglII * -CA GAA ATT CAA TCC ATA TTG ACC GGC GAC ATT ISEIQSILTGDIVPDLPDS 220 *
AAC GGG CTG GAA
240
190
200
ACT A?? T 210
* * * GTT CCG GAT CTT CCA s:AC TCA 250
260
* * Hind111 * CTA AAG CCT AAG CTG AAA AGC CAA GCT TTT GAA CTC GAT TCA GAC GCT CGT I:GT GGT L K P K L K S Q A F E L D S D A R R G
Fig. 35 (continued on next page). The complete DNA sequence of the p10 gene and its neighboring flanking regions from Autographa californica nuclear polyhedrosis virus. The predicted open reading frame representing the polybedrin sequence and principal restriction sites are indicated.
Ldpez-Ferber, 270 280 290 * t * AAA CGC AGT TCC AAG TAA ATG AAT CGT TTT K R S S K 330 340 * * TAT TCG TAC GAT TCT T7'G ATT An:
AAT ATA
390 It AAA AAT
Al'2
300
310
l
f
320 *
TAA AAT AAC AAA TCA ATT GTT TTA TAA
350 360 370 * * * EC11 TAA TAA AAT GTG ATC ATT AGG AAG ATT
ACG AA.i
400 410 420 430 * l * AGT TCT GTG TGT ATA ACA AAT GCT GTA AAC GCC A:A
ATT GTG
440 450 460 * * * TTT GTT GCA AAT AAA CCC ATG ATT ATT 500 * GAC AAT
Sisk, and Possee
470 480 * * TGA Tl'A AAA TTG TTG TTT
380
490 f TCT TTG TTC ATA
540 550 510 520 530 * * l l * AGT GTG TTT TGC CTA AAC GTG TAC TGC ATA AAC TCC ATG CGA GTG TAT AGC
570 580 590 600 * * * * * GAG CTA GTG GCT AAC GCT TGC CCC ACC AAA GTA GAT TCG TCA AAA TCC TCA ATT TCA 560
610 620 630 640 * * * * TCA CCC TCC TCC AAG TTT AAC ATT TGG CCG TCG GAA TTA 670 * TAA TCT
680 * AAT AAA TGA AAT
690 * AGA GAT WA
650 660 * * ACT TCT A&A GAT GCC ACA
710 720 700 * * * AAC GTG GCG TCA TCG TCC GTT TCG ACC ATT
730 740 750 760 770 780 * * * l l * TCC GAA AAG AAC TCG GGC ATA AAC TCT ATG ATT TCT CTG GAC GTG GTG TTG TCG AAA 790 800 * l CTC TCA AAG TAC GCA GTC AGG I&C
810
820
830
*
*
*
GTG CGC GAC ATG TCG TCG GGA AAC TCG CGC GGA
840 850 860 870 880 890 * * * * * * AAC ATG TTG TTG TAA CCG AAC GGG TCC CAT AGC GCC AAA ACC AAA TCT GCC AGC GTC
AAT
900 910 920 930 * * * * AGA ATG AGC ACG ATG CCG ACA ATG GAG CTG GCT TGG ATA
Fig. 35 (continued)
940 * GCG ATT
950 * CGA GTT AAC
Baculovtrus ‘1%ansf&r Vectors
58
should permit the reader to understand more clearly how the individual expression vectors were constructed. A newcomer to the field of baculovirus expression vectors will probably choose a polyhedrin locus, single gene expression vector such as pVL1392/1393, pBacPAK1, pEVmV, or pBlueBac2, for his or her fist experiments. He/She will also wish to purchase linearized viral DNA from either Invitrogen, Clontech, or PharMingen to facilitate the selection of recombinant virus References 1. Maeda, S. (1989) Gene transfer vectors of a baculovirus, Bombyx mori nuclear polyhedrosls virus, and their use for expression of foreign genes in insect cells. Invertebrate cell system applications, vol. 1 (Mitsuhashi, J., ed.), CRC, Boca Raton, FL, pp. 167-181. 2. Maeda, S. (1989) Expression of foreign genes in insects using baculovirus vectors. Annu. Rev. Entomol. 34,35 I-372.
3. Maeda S., Kawai, T., Obinata, M., Chika, T. H. Horiuchi, T., Maekawa, K,, Nakasuji, K., Saeki,Y., Sato, Y., Yamada, K., and Furusawa, M. (1984 1Characterization of human interferon-a produced by a gene transferred by a baculovirus vector in the silkworm, Bombyx mori. Proc. Jpn. Acad. Ser. B60,423--426. 4. Maeda, S., Kawai, T., Obinata, M., Fujiwaa, H., Horiuchi, T., Saeki, Y , Sate, Y., and Furusawa, M. (1985) Production of human alpha-interferon m silkworm using a baculovirus vector. Nature 3 15592-594. 5. Horiuchi, T., Marumoto, Y., Saeki, Y., Sato, Y , Furusawa, M., Kondo, A., and Maeda, S. (1987) High-level expression of the human a-Interferon gene through the use of an improved baculovirus vector in the silkworm, Bombyx mori Agricultural Biol. Chem.
51,1573-1580.
6. Davies, A. H. (1994) Current methods for manipulating
baculovvuses. I.?ioffech-
nology12,47-50. 7. Peakman, T. C., Harris, R., and Gewert, D. R. (1992) Highly efficient generation of recombinant baculoviruses by enzymatically mediated site-specific in vitro recombination. Nucleic Acids Res. 20,495-500. 8. Luckow, V. A., Lee, S. C., Barry, G. F., and Olins, P. 0. (1993). Efficient generation of infectious recombinant baculoviruses by site-specific transposon-mediated insertion of foreign genes into a baculovirus genome propagated in Escherichia coli. J. Viral. 67,4566-4579.
9. Patel, G., Nasmyth, K., and Jones, N. (1992). A new method for the isolation of recombinant baculovirus. Nucleic Acids Res. 20,97-104. 10. Kool, M., Voncken, J. W., van Lier, F L. J., Tramper, J., and Vlak, J. M. (1991) Detection and analysis of Autographa californica nuclear polyhedrosis virus mutants with defective interfering properties. Virology 183,739-746 11. Kitts, P., Ayres, M. D., and Possee, R. D. (1990) Linearization of baculovirus DNA enhances the recovery of recombinant virus expression vectors. Nucleic Acid Res.
l&5667-5672
Lopez+ erber, msk, and Yossee
brJ
12 Kitts, P. A. and Possee, R. D. (1993). A method for producing recombinant baculovirus expression vectors at high frequency. Biorechniques 14,810-817. 13. Smith, G. E., Ju, G., Ericson, B. L., Moschera, J., Lahm, H.-W., Chissonite, R., and Summers, M. D. (1985) Modification and secretion of human interleukin 2 in insect cells by a baculovirus expression vector. Proc. Nut. Acad. Sci. USA 82,8404-8408. 14. Miller, L. K. (1988). Baculovirus as gene expression vectors. Ann. Rev. Microbial. 42,177-99. 15. Luckow, V. A. and Summers, M. D. (1988) Trends in the development of baculovirus expression vectors. &o/Technology 6,47-55. 16. Luckow, V. A., Summers, M. D. (1988) Signals important for high-level expression of foreign genes in Autographa californica nuclear polyhedrosis virus expression vectors. Virology 167,56-7 1, 17. Luckow, V. and Summers, M. D. (1989) High level expression of non fused foreign genes with Autographa californica nuclear polyhedrosis virus expression vectors. Virology 170,31-39. 18. Matsuura, Y., Possee, R. D., Overton, H., and Bishop, D. H. L. (1987) Baculovirus expression vectors: The requirements for high expression level of proteins, including glycoproteins. J. Gen. Virol. 67, 1515-1529. 19. Possee, R. D., Howard, S. C. (1987) Analysis of the polyhedrin gene promoter of Autographa californica nuclear polyhedrosis virus. Nucleic Acid Res. 15, 10,223-10,248. 20. Merrywheather, A. T., Weyer, U., Harris, M. P. G., Hirst, M., Both, T., and Possee, R. D. (1990) Construction of genetically engineered baculovirus insecticides containing the Bacillus thuringiensis subsp. kurstaki I-ID-73 delta endotoxin. J. Gen. Virol. 71,1535-1544. 21. O’Reilly, D. R., Miller, L. K., and Luckow, V. (1992). Baculovinrs Expression Vectors. A Laboratory Manual. W. H. Freeman, New York. 22. Bishop D. H. L., Hill-Perkins, M., Jones, L. M., Kitts, P., Lopez-Ferber, M., Clarke, A. T., Possee, R. D., Pullen, and Weyer, U. (1992). Construction of baculovirus expression vectors, in Baculovirus and RecombinantProtein Production Processes (Vlak, J. M., Schlaeger, E.-J., and Bernard, A. R., eds.), Editiones Roche, Basel, Switzerland, pp. 27-50. 23. Rankin, C., Ooi, B. G., and Miller, L. K. (1988) Eight base pairs encompassing the transcriptional start point are the major determinant for baculovirus polyhedrin gene expression, Gene 70,3949. 24. Miller, L. K. (1990) Improved baculovirus expression vectors. International patent application number C12N 15/86. 25. Beames, B., Braunagel, S., Summers, M. D., and Lanford, R. E. (1991) Polyhedrin initiator codon altered to AUU yields unexpected fusion protein from a baculovirus vector. Biotechniques 11,378-383. 26. Sanchez-Martinez, D. and Pellett, P. D. (1991) Expression of HIV-l and HIV-2 glycoprotein G in insect cells by using a novel baculovirus expression vector. Virology 182,229-238.
27. Thiem, S. M. and Miller, L. K. (1990) Differential gene expression mediated by late, very late and hybrid baculovirus promoters. Gene 91,87-94.
Baculovirus
Transfer
Vectors
61
28. Wang, X., Ooi, B. G., and Miller, L. K. (1991) Baculovirus vectors for multiple gene expression and for occluded virus production. Gene 100,131-137. 29. Livingstone, C. and Jones, I. (1989) Baculovirus expression vectors with single stranded capability. Nucleic Acid Res. 17,2366. 30. Vialard, J., LaLumiere, M., Vernet, T., Briedis, D., Alkhatib, G., Henning, D., Levin, D., and Richardson, C. (1990) Synthesis of the membrane fusion and hemagglutinin proteins of measles virus, using a novel baculovirus vector containing the P-galactosidase gene. J. Virol. 64,37-50. 31. Hill-Perkins, M. S. and Possee, R. D. (1990) A baculovirus expression vector derived from the basic protein promoter of Autographa califomica nuclear polyhedrosis virus. J. Virol. 71,971-976. 32. Richardson, C. D., Banville, M., Vialard, J., and Meighen, E (1992). Bacterial luciferase produced with rapid screening baculovnus vectors is a sensitive reporter for infection of insect cells and larvae which contain a P-galactosidase mdicator gene. Intervirology 34,213-227. 33. Zuidema, D., Schouten, A., Usmany, M., Maule, A. J., Belsham, G. J., Roosien, J., Klinge-Roode, E. L., van Lent, J. W. M., and Vlak, J. M. (1990) Expression of cauliflower mosaic virus gene I in insect cells using a novel polyhedrin-based baculovirus expression vector. J. Gen. Virol. 71,2201-2209. 34. Godeau, F., Saucier, C., and Kourilsky, P. (1992). Replication inhibition by nucleoside analogues of a recombinant Autographa californica multicapsid nuclear polyhedrosis virus harboring the herpes thymidine kinase gene driven by the IE-l(0) promoter: a new way to select recombinant baculoviruses. Nucleic Acids Res. 20, 6239-6246.
35. Lytvin V., Fortin, Y., Banville, M., Arif, B., and Richardson, C. (1992). Identification and characterization of the thymidine kinase genes from three different entomopoxviruses. J. Gen. Virol. 73,3235-3240. 36. Wang, X, Xie, W., Long Q., He, D., Lin, G., Pang, Y., and Pu, Z. (1992). Stimulation of thymidine kinase activity in baculovirus infected cells is not due t,o a virus coded enzyme. Arch Virol. 127,3 15-3 19. 37. Weyer. U. Knight, S., and Possee, R. D. (1990) Analysis of very late gene expression by Autographa californica nuclear polyhedrosis virus and the further development of multiple expression vectors. J, Gen. Virol. 71, 1525-1534. 38. Stewart, L. M. D., Hirst, M., L6pez-Ferber, M., Merrywheather, A. T., Cayley, P. J., and Possee, R. D. (1991) Construction of an improved baculovirus insecticide containing an insect-specific toxin gene. Nature 352,85-88. 39. McCutchen, B. F., Choudary, P. V., Crenshaw, R., Maddo, D., Kamita, S. G., Palekar, N., Volrath, E., Fowler, E., Hammock, D. B. D., and Maeda, S. (1991) Development of a recombinant baculovirus expressing an insect-selective neurotoxin: potential for pest control. &o/Technology 9,848-852. 40. Ooi, B. G., Rankin, C., and Miller, L. K. (1989) Downstream sequences augment transcription from essential initiation site of a baculovirus polyhedrin gene, J. Mol. Viol. 210,721-736. 41. Tomalki, M. D. and Miller, L. K. (1991) Insect paralysis by baculovirus-mediated expression of a mite neurotoxin gene. Nature 352,82-85.
62
Lc5pez-Ferber, Sisk, and Possee
42 Gonnet, P. and Devauchelle, G. (1987) Obtention par recombinaison dans le gene du polypeptide p10 d’un baculovirus exprimant le gene de resistance ?I la neomycine dans les cellules d’insecte. CR. Acad. Sci. (Paris) 305, 11 l-l 14. 43. Vlak, J. M., Klinkenberg, F. A., Zaal, K. J. M., Usmany, M., Klinge-Roode, E. C., Geervliet, J. B. F., Roosien, J., and Van Lent, J. W. (1988) Functional studies on the p10 gene of Autographa californicu nuclear polyhedrosis virus using a recombinant expressing a plO-beta-galactosidade fusion gene. J. Gen. Viral. 69, 765-776. 44. Williams, G. V., Rohel, D. Z., Kuzio, J., and Faulkner, P. (1989) A cytopathological investigation of Autogruphu culifornicu nuclear polyhedrosis virus p10 gene function using insertion/deletion mutants. J. Gen Viral. 70, 187-202. 45. Vlak, J. M., Chouten, A., Usmany, M., Belsham, G. J., Kinge-Roode, E. C., Maule, A. J., Van Lent, J. W. M., and Zuidema, D. (1990) Expression of cauliflower mosaic virus gene I using a baculovirus vector based upon the p10 gene and a novel selection method. Virology 17,9312-9320. 46. Davies, A. H. and Jones, L. M. (1991) Recombinant baculovirus vectors expressing glutathione-S-transferase (GST) fusion proteins. Bio/TechnoZogy 11,933-936.
47. Peng, S., Sommerfelt, M., Logan, J., Huang, Z., Jilling, T., Kirk, K, Hunter, E., and Sorscher, E. (1993) One-step affinity isolation of recombinant protein using the baculovirus/insect cell expression system. Protein Expr. Pur& 4,95-100. 48. Chen, X. S., Brash, R., and Funk, C. D. (1993) Purification and characterization of recombinant histidine-tagged human platelet 1Zlipoxygenase expressed in a baculovirus/insect cell system. Eur. J. Biochem. 214,845-852. 49. Murphy, C. L., M&tire, J. R., Davis, D. R., Hodgdon, H., Seals, J. R., and Young, E. (1993). Enhanced expression, secretion, and large-scale purification of recombinant HIV-l gp120 in insect cells using the baculovirus egt and p67 signal peptides. Protein Expr. Purif, 4,349-357. 50. Emery, V. C. and Bishop, D. H. L (1987) The development of multiple expression vectors for high level synthesis of eucaryotic proteins: expression of LCMVN and AcNPV polyhedrin protein by a recombinant baculovirus. Protein Eng. 1, 359-366.
51. French, T. J. and Roy, P. (1990) Synthesis of bluetongue virus (BTV) corelike particles by a recombinant baculovirus expressing the two major structural proteins of BTV. J. Vol. 64, 1530-1536. 52. Putlitz, J., zu, Kubasek, W. L, DuchQne, M., Marget, M., von Specht, B.-U., and Domdey, H. (1990). Antibody production in baculovirus infected cells. Bio/technology 8,65 1-654
53. Weyer, U. and Possee, R. D. (1991) A bacuiovirus dual expression vector derived from the AcMNPV polyhedrin and pl0 promoters: coexpression of two influenza virus genes in insect cells. J. Gen. Viral, 72,2967-2974. 54. Belyaev, A. S. and Roy, P. (1993). Development of baculovirus triple and quadruple expression vectors: coexpression of three or four bluetongue virus proteins and the synthesis of bluetongue virus-like particles in insect cells. Nucleic Acids Res. 21,1219-1223.
Baculovirus
Transfer Vectors
63
55. Richardson, C., Attia, J., Dunn, R., Gupta, S., O’Connor, M., Semeniuk, D., Tam, J., Hamel, M., Lambert, G., Dennis, M., Jacobs, F., Martin, L., Iorro, C., and Vialard, J. (1992) Engineering glycoproteins for secretion using the baculovirus expression system, in Baculovirus and Recombinant Protein Production Processes (Vlak, J. M., Schlaeger, E.-J., and Bernard, A. R., eds.), Editiones Roche, Basel, Switzerland, pp. 67-74. 56. Cerutti, M., Hue, D., Charlier, M., L’Haridon, R, Pernollet, J. C., Devauchelle, G., and Gaye, P. (1991) Expression of a biologically active ovine trophoblastic interferon using a baculovirus expression system. Biochem. Biophys. Res. Commun. 181,443-448. 57. Chaabihi, H., Ogliastro, H., Martin, M., Giraud, C., Devauchelle, G., and Cerutti, M. (1993) Competition between baculovirus polyhedrin and p10 gene expression during infection of insect cells. J. Viral. 67,2664-267 1.
CHAPTER3 Insect Cell-Culture in Serum-Containing
Techniques Medium
Stefan A Weiss, William G. Whitford, Stephen E Gomen, and Glenn I? Godwin 1. Introduction Almost 50 years ago, Grace reported the successful establishment of continuous cell lines from insects that were without defined biological and biophysical parameters (I). Since that time, the minimal distinctive parameters for the establishment, maintenance, and manipulation of invertebrate cell cultures have been identified and optimized (2,3). The establishment of new lines is now routine, and the culture of hundreds of cell lines from a variety of insect species have to date been reported (2,4,5,6). During the last 10 years, genetic engineering has revolutionized insect cell culture and baculovirology; this technology has become the method of choice for production of a variety of recombinant proteins of medical and agricultural importance (7,8). These scientific events have renewed previous interest in using cell culture for production of baculovirus for viral pesticides (9-14). Vertebrate sera,in place of insect hemolymph, have been used in insect cell culture for the past 30 years and, until recently were thought to provide an unidentified high-molecular-weight, growth-promoting substance not provided in more defined, protein-free formulations (15). Fetal bovine serum (FBS), which is most often utilized, can provide a growthpromoting effect, as well as other distinct values, to basal media formulations (16,17). These values include protection from sheer forces, toxin-masking effects, and cell/substrate attachment factors. From:
Methods In Molecular Biology, Vol. 39: Baculovlrus Ed&d
by C. D. Rchardson
CQ 1995
65
Humana
Expression Protocols
Press Inc , Totowa,
NJ
Weiss et al. It is now known, however, that each of these properties can be provided by defined, nonproteinaceous supplements (18). Furthermore, there are a number of undesirable effects associated with the use of animal sera supplemented growth media. These include excessive foaming in most bioreactors, introduction of exogenous agents determintal to cellular proliferation and/or recombinant protein production, hindrance to and increased cost of downstream processing, increased and fluctuating medium cost and serum availability, chronic suboptimal cell growth, product yields and recovery, distinct toxic effects in certain cell lines, introduction of adventitious infectious agents, and often regulatory obstacles. Despite the success obtained with many of the newer serum-free formulations in recent years, many researchers and industrial organizations still rely on serum supplementation of both their standard cellular growth and virus or recombinant protein production media. This is mainly because of the presence of established procedures and qualified production criteria that specify serum-supplemented growth medium, and the fact that there are still some specific cell lines and production configurations that have not yet been treated and/or certified in the newer serumfree media currently available. Since the introduction of Grace’s medium, a number of serum/hemolymph-dependent formulations have evolved, each providing distinct improvements in performance or having divergent applications. Of particular interest today are the more popular insect cell-culture media, such as Grace’s TNM-FH, IPL-41, TC-100, and Schneider’s drosophila formulations (19-25). High-quality commercial preparations of each of these formulations are now readily available, and their use is recommended rather than the time-consuming and less reproducible approach of in-house production. In this chapter, we will describe experimental methods using IPLSf21-AE and Sf9 cell lines. These experimental methods are, however, essentially applicable to many of the invertebrate lines described to date (8,15,20,26-29). Numerous other insect cells have been cultured in Grace’s, TNM-FH, IPL-41, TC-100, and Schneider’s media supplemented with fetal bovine or other animal sera. The most widely used established cell lines of the variety available besides Sf9 are Tn-368, Aedes Aegypti, Drosophila melanogaster (D.me1.2). Bombyx mori, Heliothis zea, IPLB-Hz-1075, Lymantria dispar (Ld), IPLD-LD-G4, and IPL-Ld (26-27).
Cell Culture
in Serum-Containing
Media
67
2, Materials Described in this section are all the equipment and reagents generally required for the methods described; specific items required for each particular procedure are listed in the respective section. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10.
2.1. Equipment and Supplies Incubator capable of maintaining 28 f 0.5”C and large enough to accommodate the desired culture configuration apparatus (Forma Scientific [Marietta, OH] Model 3919). Electronic cell counter (Coulter Model S.STD.11). Light microscope capable of hemocytometer application. Inverted light microscope for examination of flasks, roller bottles, and plates (Leitz # 020-441). Hemocytometer (
[email protected] mm, Reichert-Jung Model 1475). Low speed “clinical” type centrifuge (IEC Model CRU-5000). Automated pro-pipet (Drummond Pipet-Aid@‘,Model 400030). Lammar flow hood suitable for cell culture (NUAIRE@ N&425-400). 37OC water bath (Fisher [Pittsburgh, PA] Vera-Bath@, Model 138). Pipets: 1,2,5, 10 and 25 mL (preferably disposable plastic or glass), Corning #7075-( l-25 mL).
2. 3. 4.
2.2. Reagents and Solutions 2.2.1. Basal Cell Culture Medium of Choice Grace’s insect cell-culture medium, supplemented (TNM-FH), (GIBCO [Grand Island, NY] #350-1605). IPL-41 Insect cell-culture medium (GIBCO #350-1405). TC-100 Insect cell-culture medium (GIBCO #440-1600). Schneider’s Drosophila medium (GIBCO #350-1720).
1. 2. 3. 4. 5. 6. 7.
2.2.2. Supplements to Basal Medium Yeastolate (GIBCO #670-8200AG). Lactalbumin hydrolysate (GIBCO #670-808OAG). Tryptose phosphate broth (GIBCO #670-8060AG). Lipid concentrate (GIBCO #680-19OOAG). Pluronic polyol F-68 (GIBCO #670-4040AG). Antibiotic concentrates (GIBCO #600-5240AG). Animal serum (i.e., fetal bovine serum),heat-inactivated (GIBCO #2W6140).
1.
2.2.3. Additional Solutions 1. Phosphate-buffered saline solution (GIBCO #3 1O-4190AG). 2. Trypan blue solution (GIBCO #630-525OPE).
Weiss et al.
2.3. Insect Cell Lines We routinely use the Spodopteru frugiperda (S’) cell line (ATCC #CRL 17 11) which has a doubling time of 18-24 h once adapted to any of the following popular media formulations, respectively: 1. IPL41 containing 2.6 g/L of tryptose phosphate plus 5-10% FBS or IPL-41 containing 4 g/L of yeastolate and 5-10% FBS. 2. TC-100 supplemented with 3.3 g yeastolate, and 3.3 g lactalbumin hydrolysate/L, and 10% FBS. 3. Grace’s insect medium supplemented with 3.3 g yeastolate and 3.3 g lactalbumin hydrolysate/L, and 10% FBS. 2.4. Cell Culture
Equipment 1. Cell-culture “T’‘-Flasks, 25 and 75 cm2 (Falcon brand sterile disposable polystyrene: 25 cm2, #3013; 75 cm2, #3024). 2. 100 mL-1000 mL spinner flasks (Coming brand [Coming, NY], Series 26500 Stirring Vessel Systems). 3. Stirring platform capable of constant operation at 70-80 rpm at 28OC + 0.5OC (Thermolyne Model 45700). 4. Orbital shaker with clamps fitted for 100-500 mL Erlenmeyer flasks (Labline Model 3520); disposable 125-, 250-, and 500-mL Erlenmeyer flasks (Corning: 125 mL, #25600-125; 250 mL, #25600-250; 500 mL, #25600-500).
2.5. Equipment
and Reagents
for Cryopreservation
1. Grade DMSO (Sigma [St. Louis, MO] #D2650). 2. Cell cryopreservation vials (Corning: 5 nil,, #66021-970). 3. Automated freezer (Cryomed Model 1010).
3. Methods Cultures are incubated at 28 + 0.5”C (see Notes l-3). A monolayer will reach confluency in 4-7 d with a medium change on day 4. The confluency timing will depend on the cell inocula. Suspension cultures reach maximum densities in 5-6 d. In monolayer culture, the cell population consists of cells that loosely attach to the substrate with some floating in the culture medium. The majority of established insect cell lines are not anchorage-dependent
and may be transferred between monolayer
and spinner/shaker culture repeatedly without noticeable perturbation of normal viability, morphology, or growth rate. However, as cultures may be passage-number-dependent, fresh cultures should be established from frozen seed stocks every 3 mo. Antibiotics are not specifically recom-
Cell Culture
in Serum-Containing
Media
69
mended; however, many are commonly employed. A final concentration of 50 pg/mL gentamicin and 2.5 pg/mL amphotericin B (also called Fungizone) is popular in many laboratories. of Insect Cell Monolayers The following procedure is adaptable to a variety of cell lines in either plastic flasks or dishes. 3.1. Cultivation
1. With a 10 mL pipet, aspirate medium and floating cells from a confluent monolayer and discard. 2. Add 4 mL of complete medium to a 25 cm2 flask (12 mL to a 75 cm2 flask). 3. Resuspend cells by pipeting the medium across the monolayer with a Pasteur pipet (or equivalent device). 4. Observe cell monolayer using an inverted microscope to ensure complete cell detachment from the surface of the flask. 5. Perform viable cell count on harvested cells (e.g., using trypan blue exclusion). 6. Inoculate cells at 2 x lo5 viable cells/ml in respective vessel. 7. Incubate cultures with caps loosened to allow oxygen/air exchange at 28 f 0S”C. 8. On day 4 postplanting, aspirate the spent medium from one side of the monolayer and refeed the culture with fresh medium gently added to the side of the flask. 9. Passagecells as they approach confluency (day 4-7) (see Note 4). Figure 1 shows Sf!3 cells grown in suspension and placed on hemo-
cytometer for counting before planting in monolayer cultures. The cells were grown in IPL-41 supplemented with 5% FBS using the suspension method. A lOO-mL culture in 250 mL. Erlenmeyer shake flask was incubated at 28°C on an orbital shaker with speed set at 135 t-pm. The cells grown in suspension are easily switched to growth in monolayer culture. Figure 2 shows Sf9 cells grown in monolayer culture for 24 h postplanting in IPL-41 with TPB + 5% FBS. The cell inoculaconsisted of 2 x lo5 cells/mL. of Insect Cells in Spinner Culture Beginning with a line qualified as described in Section 3.1. and preferably employing a medium formulation containing additional shear protection agents (such as Pluronic Polyol F-68 at 0.05-o. 1% final 3.2. Cultivation
concentration), the following protocol is a generic outline of the basic technique. Large scale and bioreactor applications of suspension culture are a science in itself and are reviewed elsewhere (3,28,29,30).
Weiss et al.
Fig. 1. Photomicrograph of Sf9 cells grown in monolayer culture for 2 d postplanting in IPL-41 with TPB + 5% FBS. The cell inocula consisted of 2 x lo5 cells/ml.
Fig. 2. Photomicrograph of Sf9 cells grown in suspension that were placed on hemocytometer for counting before planting in monolayer cultures. The cells were grown in IPL-41 supplemented with 5% FBS using suspension method. A 100 mL culture in 250 mL Erlenmeyer shake flask was incubated at 28 + OSOC on an orbital shaker rotating at 135 rpm.
Cell Culture
in Serum-Containing
Media
71
1. Recalibrate the graduation marks on commercial spinner flasks using a graduated cylinder or volumetric flask as a reference. 2. Ensure that impeller mechanisms rotate freely and do not contact vessel walls or base since insect cells are sensitive to physical shearing, Prior to autoclaving, adjust the spinner mechanism so paddles clear sides and bottom of the vessel. 3. Four to six confluent 75 cm2 monolayer flasks are required to initiate a 100 mL culture (4-5 flasks for the spinner culture and 1 flask to be used as a backup). 4. Dislodge cells from the base of the flasks as described in Section 3.1. 5. Pool the cell suspension and perform a viable cell count. 6. Dilute the cell suspension to approx 3 x l@ viable cells/ml in complete growth medium. 7. For culture volumes of 75-100 mL, use a 100~mL spinner vessel. For volumes of 150-200 mL, use a 250~mL vessel, and for 2.0 L, use a 3.0-L spinner (see Note 5). 8. Stock cultures should be maintained in a 150-r& culture employing a 250~mL spinner vessel. The top of the paddles will be slightly aibove the medium allowing for additional aeration to the cultures. 9. Atmospheric oxygen/air equilibration is accomplished by loosening the side arm caps on the vessels (about l/4 of a ton) (see Note 6). 10. Incubate spinner vessels at 28 f 0.5”C at a constant stirring rate of 75 rpm (see Note 7). 11. Subculture spinner cultures to approx 2-3 x lo5 cells/ml twice weekly in well-cleaned, sterile vessels, 12. Once every 2 wk, spinner cultures may be centrifuged (1OOg) for 5 min and resuspended in fresh medium to reduce accumulation of ccl1 debris and toxic waste byproducts. (see Note 8).
3.3. Cultivation
of Insect
Cells in Shaker
Culture
Beginning with a cell line cultured as described in Section 3.2. and employing a growth medium supplemented with a shear-protection agent, such as Pluronic Polyol F-68, this protocol provides a generic outline of the basic shaker technique. The orbital shaker apparatus must have a capacity of up to 140 rpm and should be maintained in a 28 f 0.5” C nonhumidified, nongas-regulated environment. The standard flask employed is the 250 mL disposable sterile Erlenmeyer (see Note 9), especially when working with reduced serum. Aeration is accomplished by loosening the cap approx l/4 turn (within the intermediate closure position). In 1his condition, there is no oxygen limitation to the cells when one uses 100 mL culture volume in a 250 mL flask (see Note 6); and therefore, they
72
Weiss et al.
0 7
IPL-41 IPL-41
+ 5% FE.3 + 10% FBS
0123456
7
m4E
8
9
10
11
(DAYS)
Fig. 3. Graph shows growth kinetics of Sf9 cells in media supplemented with FBS using shaker culture system. One hundred milliliters of cell suspension were planted at 2 x lo5 cells/ml in 250-r& shaker flask on an orbital shaker platform rotating at 125-135 rpm. Cultures were incubated at 28 f 05°C with loosened caps on the flasks to permit gas exchange. Samples were taken daily for viable cell counts. This was a batch culture without refeeding or subculturing after planting of the cultures. As shown in this figure, the maximum cell yields were reached on days 6-8 postplanting with respective media. proliferate at maximal rates. However, the oxygen limitation increases when the volume of cell suspension and vessel volume increase. 1. Inoculate a 250 mL Erlenmeyer flask with 100 nL of complete medium containing 2 x lo5 viable cells/ml. 2. Incubate at 28 f 0.5”C with open caps and set the orbital shaker at a speed of 80-90 rpm (see Note 7). 3, Subculture to approx 2 x lo5 cells/n& twice weekly. 4. Every 3 wk, cultures may be gently centrifuged 1OOgfor 5 min and pellets resuspended in fresh medium to reduce accumulation of cell debris and toxic waste byproducts. Figure 3 shows growth kinetics of Sf!3 cells in media supplemented with FBS using a shaker culture system. A 100~mL cell suspension was
Cell Culture in Serum-Containing
Media
73
20 0
l
LOOSENED CAP TIGHTENED CAP
0
-o-o\
10
5
0
4
5
TIME
(DAYS)
6
Fig. 4. Depicts kinetics of Drosophila mehnogaster (D.mel.2) cell grown in a 100~mLshakerculture. The growth media consistedof Schneider’sj‘ormulation (23,24) supplemented with 2.5% heat-inactivated FIW and 0.05% Pluronic
F-68. When the cultures areprovided oxygen/air supply by loosening1he caps, the maximum cell density reached nearly 2 x lo7 cells/ml 011 day 6 postplanting. If the culture is grown devoid of free oxygen/air supply (tightened caps), the maximum cell density reached is about 7.0 x lo6 cells/mL.
planted at 2 x lo5 cells/ml in 250~mL shaker flask on an orbital shaker platform rotating at 125-135 rpm. Cultures were incubated at 28 k 0.5”C
with flasks having loosened caps to permit gas exchange. Samples were taken daily for viable cell counts. This was a batch culture without any refeeding or subculturing carried out after planting of the cultures. As shown in this figure, the maximum cell yields were reached on days 6-8 postplanting with respective mend (see Note 8). Figure 4 depicts kinetics of Drosophila melanogaster (D.mel.2) cell growth in a 100-n& shaker culture. The growth media consisted of Schneider’s formulation (23,24) supplemented with 2.5% heat-inactivated FBS and 0.05% Pluronic F-68. When the cultures were provided #oxygen/ air supply by loosening caps, the maximum cell yield reached nearly 2 x lo7 cells/ml on day 6 postplanting. If the culture was grown devoid of
74
Weiss et al. Table 1 Metabolic Profile of Spent Media from Drosophila
Days in culture
Cell count 7.00 2.45 7.92 1.39 2.00 2.21 2.30 2.39
x x x x x x x x
105 lo6 lo6 107 107 107 lo7 lo7
D-Galactose, mk+ 1860 1760 1420 680 60 40 0 0
melunogaster (D.mel.2)a
L-Lactate, m&
Ammonium, mk&
300 310 250 170 50 60 70 60
31.0 43.2 61.2 126.0 190.0
Ammonium, (1.72) (2.4) (3;) (7;) (10;)
* Summarizes metabolic profile of spent media from D.mel.2 cells grown under conditions described in Fig. 4. When the cells were planted at 2 x 10s cells/ml, they reached maximum yield of 2.39 x lo7 cells/ml on day 8 postplanting. The glucose was depleted from 1860 to 40 mg/L on day 6 postplanting and was exhausted on day 7 postplanting. Surprisingly, lactate level was not incremental and started to decrease significantly on day 5 postplanting. However, the ammonium accumulation was evident on day 5 postplanting and was increased srxfold on day 8 postplanting when the cell densities reached maximum peak.
free oxygen/air supply (tightened caps), the maximum cell density reached about 7.0 x lo6 cells/mL. The supply of oxygen from ambient air in the shaker culture of D.mel.2 cells resulted in 2.9-fold increase in the cell yield. Table 1 summarizes the metabolic profile of spent media from D.mel.2 cells grown under conditions described in Fig. 4. When the cells were planted at 2 x lo5 cells/ml, they reached maximum cell yield of 2.39 x lo7 cells/mL on day 8 postplanting. The glucose was depleted from 1860 to 40 mg/L on day 6 postplanting and exhausted on day 7 postplanting. Surprisingly, the lactate level was not increased, and it started to decrease significantly on day 6 postplanting. However, ammonium accumulation was evident on day 5 postplanting and was increased by sixfold on day 8 postplanting when the cell densities reached the maximum level. 3.4. Cryopreseruation of Cultured Insect 3.4.1. Freezing Insect Cells
Cells
desired quantity of cells in either spinner or shaker culture, harvesting in midlogarithmic. 2. Determine the viable cell count, and calculate the required volume of cryopreservation medium to yield a final cell density of 0.5-l .O x lo7 cells/ml. 1. Prepare
Cell Culture in Serum-Containing
Media
75
3. Prepare that volume of freeze medium consisting of 7.5% DMSO in a 50% fresh medium and 50% conditioned medium. Hold this preparation at 4OC. 4. Centrifuge cells from culture medium at 1OOgfor 6 min. Resuspend pellet in the determined volume of 4OCcryopreservation medium. 5. Incubate cell suspension at 4°C until well chilled. 6. Dispense aliquots of this suspension to cryovials according to manufacturer’s specifications (i.e., 4.5-5.0 mL cryovial). 7. Achieve cryopreservation in either an automated or manually controlled rate freezing apparatus following standard procedures. Alternatively, one can freeze the cryovials gradually using dry ice (2 h), -70°C freezer (24 h), and finally liquid nitrogen. For ideal cryopreservation, the freeszing rate should be 1OC/min. 8. Frozen cells are stable indefinitely in liquid nitrogen storage.
3.4.2. Recovery of Viable Cells from Frozen CultureGs 1. Recover cultures from frozen storage by rapidly thawing a vial of cells in a water bath at 28 f OSOCfollowed by rapid aseptic transferring of l-heentire contents of the vial into a 250 mL shaker flask containing 100 mL complete growth medium and then incubate culture as outlined in Section 3.3. 2. Maintain culture between 3 x lo5 and 1 x lo6 cells/ml for the first two subcultures after recovery; thereafter returning to the normal maintenance schedule. Additional methodologies on cryopreservation of insect cells have been described in detail in other publications (3,20).
4. Notes 1. The facilities in which one proposes to perform insect cell cull ure must comply with minimal requirements of service, cleanliness, and isolation. Additionally, adequate space must be available for the installation of cellculture-specific incubators, flow hoods, autoclaves, special gases, and bioreactors according to well-established designs (31,32). 2. For serum-supplemented culture, it is necessary to characterize each lot of serum adequately. This includes certification of freedom from bacterial, mycoplasmic, or viral contaminants, as well as functional characterization for the support of cell growth, substrate adhesion, and plating efficiency (3). Since no two lots of serum are identical, each must be exhaustively scrutinized prior to qualification for use, or else be the first suspect of responsibility in the event of inadequate proliferation or production performance. 3. Many cultures of invertebrate cells are more sensitive to exogenous contaminates and toxic byproducts than many of both the lower and higher eukaryotic continuous cell lines. For this reason, novice practitioners are
76
4.
5.
6.
7.
8.
9.
Weiss et al. cautioned to select growth media components of the highest purity and pay particular attention to the quality of the water used in every phase of culture preparation from media formulation to glassware cleaning (3,32). Sf9 cells do not strongly adhere to most substrates in the presence of FBS; therefore, care should be taken to transport cultures gently through the laboratory. Care should also be taken (by maintaining minimal media depth, maximal cap loosening) to provide for good oxygenation of cultures, which is especially critical for virus infections. Although spinner culture is scalable to a degree, there is a physical constraint to particular applications owing to the requirement of adequate gas partitioning in the culture. A general rule is to keep the volume in the spinner vessel below l/3 full and provide more and larger ports for headspace gas exchange as the vessel size increases. Another well-established constant in the culture and applicatton of invertebrate cells is their dependency on high and consistent supplies of oxygen in proliferation and especially production modes (25). In passively oxygenated systems, every effort is made to maximize gas exchange; in actively and controlled oxygenated systems,lO-50% saturation is commonly employed. Shear stresseffects are of ubiquitous concern in suspension culture applications. Sheer forces are generated in suspension culture through a varrety of mechanical events. Normal concentrations (5-10%) of animal sera do provide minimal levels of protection for some standard culture technologies (i.e., some spinner culture applications). However, it is necessary to supplement cultures with additional shear protectants when attempting to employ serum fractions or substrtutes, switch to entirely serum-free formulations at specific points in the culture/production protocol, or change to a culturing method requiring additional shear protection, such as shake flask or aggressively sparged culture. Although a number of such protectants have been described, Pluronic Polyol F-68 IS most widely used (3,18,25,3(I). The ambient pH of the fully supplemented growth medium is another important parameter of both cellular proliferation and production of virus or recombinant proteins. Although many values have been presented for various cell lines in the past, a value of 6.2 for most cultures has emerged to be the best in most culture systems (3,6,20,27). The popular media highlighted here will maintain this pH level in open capped culture systems without COZ supplementation. Although shaker-type culture is scalable to a variety of subsequent vessels, volumes, and flask sizes,each has its own growth characteristics. Therefore, relative flask fill volumes and orbital shaker speeds must be optimized for each configuration.
Cell Culture
in Serum-Containing
Media
77
Acknowledgments The authors are grateful to Terrilyn M. Summers, Senior Secretary (Life Technologies, Inc.) for the preparation of this manuscripl. and to Phillip Grefrath (Life Technologies, Inc.) for conducting experiments with D.mel.2 cells. We also express thanks to Imogene Schneider for providing us with the D.mel.2 cell line. References 1. Grace, T. D. C. (1962) Establishment of four strains of cells from insect tissues grown in vitro. Nature 195,788-789. 2. Lynn, D. E. (1989) Methods for the development of cell lines from insects. J. Tissue Culture Methods 12(l), 23-29.
3. Weiss, S. A. and Vaughn, J. L. (1986) Cell culture methods for large-scale propagation of baculoviruses, in The Biology of Baculoviruses (Granados, 11. R. and Federici, B. A., eds.), CRC, Boca Raton, FL, pp. 64-87. 4. Hink, W. E (1980) Invertebrate systems in vitro, in The 1979 Compilation oflnvertebrate Cell Lines and Culture Media (Kurstak, E., ed.), Elsevier, New York, pp. 553479. 5. Hink, W. F. (1989) Recently established invertebrate cell lines, in InverteBrate Cell System Applications. (Mitsuhashi, E., ed.), CRC, Boca Raton, FL, pp. 269-293. 6. Mitsuhashi, J. (1982) in Advances in Cell Culture (Maramorosch, K., ed.) Academic, New York, pp. 133. 7. Smith, G. E., Summers, M. D., and Fraser, M. J. (1983) Production ‘of human P-interferon in insect cells infected with a baculovirus expression vector. J. Mol. Cell. Biol. 3(12), 2156-2165. 8. Luckow, V. A. (1991) Cloning and expression of heterologous genes in insect cells with baculovirus vectors, in Recombinant DNA Technology and Applications (Prokop, A., Bajpai, R. K., and Ho, C. S., eds.), McGraw-Hill, New York, pp. 97-151. 9. Cunningham, J. D. (1982) in Microbial and Viral Znfecticides (Kurstak, E., ed.), Dekker, New York, pp. 335-386. 10. Entwistle, P. F. and Evans, H. F. (1985) in Comprehensive Insect Physiology Biochemistry and Pharmacology (Kerkut, G. A. and Gilbert, L. I. eds.), Pergamon, Oxford, pp. 347-412. 11. Huber, J. (1986) Practical applications for insect control, in The Biology of Baculoviruses II. (Granados, R. R. and Federici, B. A., eds.), CRC, Boca Raton, FL, pp. 181-202. 12. Evans, H. F. and Entwistle, P. F. (1987) in Epizootiology of Insect Diseases (Fuxa, J. R. and Tanada, Y., eds.), Wiley, New York, pp. 257-322. 13. Knight, P. (1991) Baculovirus vectors for making proteins in insect cells. ASM News 57, #Ill, 567-570.
14. O’Reilly, D. R. and Miller, L. K. (1991) Improvement by deletion of the EGT gene. Biotechnology 2,1086.
of a baculovirus pesticide
Weiss et al. 15. Mitsuhashi, J. (1989) Nutritional requirements of insect cells in vitro, in Znvertebrute Cell System Applications (Mitsuhashi, J., ed.), CRC, Boca Raton, FL, pp. 3-20. 16. Yunker, C. E., Vaughn, J. L., and Cory J. (1967) Science 155,1565. 17. Landureau, J. D. and Steinbach, M. (1970) In vitro cell protective effects by certain antiproteases of human serum. Z. Natulforsch 25b, 23 1. 18. Vaughn, J. L. and Weiss, S. A. (1991) Formulating media for the culture of insect cells. BioPhatm 4(2), 16-19. 19. Hink, W. F. (1990) Established insect cell line from cabbage looper, Trichoplusia ni. Nature 226,466-467.
20. Weiss, S. A., Smith, G. C., Kalter, S. S., and Vaughn J. L. (1981) Improved method for the production of insect cell cultures in large volume. In Vitro 17,495-502. 21. Gardiner, G. R. and Stockale, H. (1975) Two tissue culture media for production of lepidopteran cells and nuclear polyhedrosis viruses. J. Invertebrate Pathol. 25,363-370. 22. Knudson, D. L. and Tinsley, T. W. (1974) Replication of a nuclear polyhedrosis virus in a continuous cell culture of Spodoptera frugiperda: Purification, assay of infectivity, and growth characteristics of the virus. J. Virol. 14(4), 934-944.
23. Schneider, I. (1966) History of larval eye-antenna1 discs and cephalic ganglia of Drosophila cultured in vitro. J. Embryol. Exp. Morphol. 15,271-279. 24. Schneider, I. (1964) Differentiation of larval Drosophila eye-antenna1 discs in vitro. J. Exp. Zool. 156,91-104.
25. Weiss, S. A., Belisle, B. W., DeGiovanni, A., Godwin, G., Kohler, J., and Summers, M. D. (1989) Insect cells as substrates for biologicals. J. Dev Biol. Standard 70,271-279.
26. Hink, W. F., Ralph, D. A., and Joplin, K. H. (1985) Metabolism and characterization of insect cell cultures, in Comprehensive Insect Physiology, Biochemistry and Pharmacology (Karkut, G. A. and Gilbert, L. I., eds.), Pergamon, New York, pp. 547-570. 27. Hink, F. and Bezanson, D. R. (1985) Invertebrate cell culture media and cell lines, in Techniques in Life Sciences, Cell Biology (Kurshk, E., ed.), vol. Cl, Elsevier
Scientific, County Clare, Ireland, pp. CIIl/l-CIII/30. 28. Weiss, S. A., Gorfien, S., Fike, R., DiSorbo, D., and Jayme D. (1990) Biotechnology: The Science and the Business. Ninth Australian Biotechnology Conference (Queensland), pp. 220-23 1. 29. Vaughn, J. L. and Weiss, S. A. (1991) Large-scale propagation of insect cells, in Large-Scale A4ammalian Cell Culture Technology (Lubiniecki, A. S., ea.), Marcel Dekker Inc., New York, pp. 597-617. 30. Murhammer, D. W. and Goochee, C. F. (1990) Structural features of nonionic polyglycol polymer molecules responsible for the protective effect in sparged animal cell bioreactors. Biotechnology 6, 142-148. 31. Weiss, S. A., Smith, G. C., Kalter, S. S., Vaughn, J. L. and Dougherty, E. (1981) Improved replication of Autographa californica NPV in roller bottles: Characterization of the progeny virus. Zntervirology 15,213-222. 32. Freshney, R. I. (1987) in Culture of Animal Cells: A Manual of Basic Techniques, 2nd ed. (Liss, A. R., ed.), New York, pp. O-84,514.
CHAPTER4
Insect Cell Culture
in Serum-Free
Media
Stefan G Weiss, Glenn I? Godwin, Stephen E Gofien, and William G. Whitfbrd 1. Introduction Historically, insect cell culture was considered a method for the production of viral pesticides (1,2). To support basic research and to explore potential replication of entomopathogenic viruses in vitro, there have been over 400 cell lines established to date from various insect species (3). Without exception, the cell lines described in the literature and currently in use have been isolated and established from primary cultures using serum-supplemented insect cell-culture media (4,5; see aZ,roChapter 3 in this volume). The potential importance of insect cell culture was not fully explored until genetic engineering revolutionized baculovirology for foreign gene expression, now known as the baculovirus expression vector system or BEVS (6). Numerous features of this higher eukaryotic expression vector system contribute to its intriguing and rapidly growing popularity. Among these are the safety, practicality, and the simplicity of producing the virus in a variety of configurations; the speed of obtaining and verifying recombinant protein expression; the glycosylation, immunologic, enzymatic, and functional similarities of the expressed product to its authentic native counterpart; and the capability to express a variety of proteins in quantities that may be toxic to mammalian expression systems (3,7,8). The majority of the expressed proteins reported j n the literature were produced in Sf9 cells propagated in insect media supplemented with fetal bovine serum (BBS); only a few of the proteins were produced in serum-free media (4,7,9-l 6). From: Methods in Molecular Brology, Vol. 39: Baculovirus Expresslon Protoccc Edited by: C. 0 Richardson CD1995 Humana Press Inc., Totowa, NJ
79
Weiss et al. Presently, there are about 60 insect cell culture media reported with a majority of these formulations developed to be used with serum supplementation (3,17-20). Prior to 1984, only a limited number of scientific publications were available on serum-free media formulations, and these formulations were targeted toward replication of insect viruses for the production of viral pesticides (17,19,21-24). An evaluation of these formulations for BEVS was described previously, and it was found that they were generally poorly defined, too rich in protein content, contained costly serum supplements, did not support the high cell yields desirable for cost effectiveness, and were cumbersome for downstream processing and product recovery. Again, it should be noted that these formulations were developed with the intent to produce baculoviruses for use in pest control (24,25). The requirements for media to be used in the production of viral pesticides, which is based on low cost and high volume, are different from the requirements of the serum-free-defined media needed for production of recombinant proteins by BEVS. The recombinant proteins produced using BEVS technology by biomedical industries need to be produced in serum-free media with defined components and are based on high cost with low-volume strategy. This strategy also needs to be cost effective. The growing popularity of BEVS provided major impetus for the development of insect serum-free media (SFM) for the production of recombinant proteins, and several SFM have been developed (7,12,13,15,16,26). The need for defined SFM has been well described (26,27). The use of SFM for the manufacturing of medically and agriculturally relevant recombinant proteins eliminates the use of fetal bovine and other animal sera, increases cell and product yields, eliminates exogenous agents, offers a consistent lot-to-lot manufacturing process, and is favored by regulatory agencies (26). The use of SFM in large-scale bioreactors is of paramount importance in the manufacturing process. This has been addressed and elaborated in detail in previous reports (7,13,16,28). The historical and current status, and the general principles for the formulation of media for the cell culture of insect cells, including SFM, were described in detail previously (7,20,26,27). In this chapter, we will attempt to describe the methods and techniques that will enable researchers and biotechnologists, as well as process development and manufacturing engineers, to employ these for a wide variety of applications.
Cell Culture in Serum-Free Media
81
2. Materials 2.1. Material for Monolayer Culture T-flasks (25, 75, and 150 cm2) and plastic roller bottles (Corning [Corning, NY]: 850 cm2, #25245-850). 2.2. Equipment for Shaker Culture Equipment consists of orbital shaker with clamps fitted for 100-500 mL Erlenmeyer flasks (Lab-Line, Model 3520), disposable Erlenmeyer flasks (Corning: 125 rnL, #25600-125; 250 mL, #25600-250; 500 mL, ##25600-500),and an incubator large enough to house at least two orbital shakers (Forma Scientific, Model 3919). 1. 2. 3. 4. 5. 6.
2.3. Cryopreaervation of Serum-Free Cultures Automated freezer (Cryomed, Model 1010). Manual freezer tray. Cell-culture gradeDMSO (Sigma [St. Louis, MO] #D2650). Cell cryopreservationvials (Corning: 5 mL, #66021-970). Fresh complete SFM (i.e., GIBCO [Grand Island, NY] Sf-900 II SFM, #350 0902). Cell-free conditionedSFM (3-4 d conditionedSFM from specifc cell line must be usedfor freezing of the specific correspondingcell line.).
2.4. Media and Solutions 1. Basal insect cell culture medium of choice: a. Sf-900 insect cell culture SFM (GIBCO #350-0900). b. Sf-900 II insect cell culture SPM (GIBCO #350-0902). c. Grace’s insectcell culture medium, supplemented(TNM-FH) (GIBCO #350-1605). d. EX-CELL 400 and EX-CELL 401 (JRH Biosciences,Lenexa, KS). e. Bio-Whittakerinsectcell cultureSFM (Bio-Whittaker,Walkersvihe,MD). 2. Supplementsto basalmedium: a. PBS, qualified (GIBCO #200-6140). b. Pluronic Polyol F-68, 10%(GIBCO #670-4040). 3. Trypan blue stain, 0.4% (GIBCO #630-5250). 3. Methods It is recommended that insect cells be adapted to suspension culture first and then to growth in SFM. When adapting stationary cultures to suspension culture, it is important to proceed slowly, since a drop in viability and increased clumping will be observed during the first three
Weiss et al. to five passages, The following protocol will allow the adaptation of most invertebrate cell lines to suspension culture and reduce/eliminate cell clumping over a short period of time. There are two approaches to be considered when adapting invertebrate cells to SFM: (1) direct planting of cells from medium containing serum to SFM; and (2) sequential adaptation or “weaning.” It is critical that cell viability be at least 90%, and the growth rate be in rnidlogarithmic phase prior to initiating adaptation. 3.1. Suspension Culture 1. Six to ten confluent T-75 cm2 monolayer flasks are required to initiate a 100 n-L suspension culture. 2. Dislodge cells from the bottom of the flasks as described in Chapter 3, Section 3.1., if using serum-containing cultures, or Section 3.4.2., if using cells already adapted to SFM culture. 3. Pool the cell suspension and perform a viable cell count (e.g., trypan blue exclusion method). 4. Dilute the cell suspension to approx 5 x 105 viable cells/ml in complete growth medium. 5. Incubate spinner vessels at 28 f 0.5OCat a constant stirring rate of 50 r-pm for spinner cultures and 100 rpm for shake flask cultures. 6. Subculture when the viable cell count reaches l-2 x lo6 cells/ml (3-7 d postplanting). Increase stirring speed 5 rpm for either spinner or shake flask cultures. If cell viability drops below 75%, decrease stirring speed 5 rpm until culture viability recovers (>80%). 7. Repeat step 6 until you have reached a constant stirring speed of 75 rpm with spinner cultures or 130 rpm with shake flask cultures. At this point, reduce the seeding density to 3 x lo5 cells/ml when you subculture. 8. If large clumps of cells (>lO cells/clump) persist, let the spinner or shaker flask culture sit 2 to 3 min prior to subculturing. This will allow the large clumps to settle to the bottom of the flask. Pull the sample for counting and for seeding new cultures from the upper third of the suspension culture. 9. It may be necessaryto repeat step 8 two to three times until all large clumps are eliminated and overall clumping in the culture is reduced. 10. Cryopreserve a large quantity of cells adapted to suspension culture for future use (refer to Section 3.7.). 3.2. Direct Adaptation to Serum-Free Media For adaptation, we recommend the use of either spinner or shaker flasks. The use of these was described in detail in Chapter 3 of this book, and they are further elaborated in Sections 3.5 and 3.6 in this chapter.
Cell Culture
in Serum-Free
83
Media
1, Cells growing in medium containing 5 to 10% PBS are transferred directly into SFM prewarmed to 28 f OS°C. 2. When the cell density reaches l-3 x lo6 cells/ml (4-7 d postplanting), subculture to a density of 3 x lo5 cells/ml. 3. When the cells are completely adapted to serum-free culture, they should reach maximum densities and have population doubling times that are comparable to the growth observed in serum-containing medium. 4. Stock cultures of SPM-adapted cells should be subcultured once to twice weekly when the viable cell count reaches l-3 x lo6 cells/ml with at least 80% viability. Note: If suboptimal performance is achieved using the direct adaptation method, use the sequential adaptation (weaning) method described in the next section of this chapter.
3.3. Sequential (Weaning) Adaptation to Serum Free Media 1. Subculture cells grown in serum-containing medium into a 1: 1 ratio of SPM and the original serum-supplemented media. 2. Incubate cultures until viable cell count exceeds 6 x lo5 cells/ml (about one population doubling). Subculture by mixing equal volumes of culture and fresh SPM (1:l). 3. Continue to subdivide the culture in this manner until the calculated serum concentration falls below O.l%, cell viability is approx 80%, and a viable cell count exceeding 6 x 10s cells/ml is achieved. 4. Subculture when the viable cell count reaches l-2 x lo6 cells/ml, which is approx 4-7 d postplanting. 5. After several passages,the viable cell counts of most insect lines should exceed 2-4 x lo6 cells/ml with a viability exceeding 85% after approx 4-7 d of culture. At this stage, the culture is considered to be adapted to SPM, and a large quantity of cells should be cryopreserved for future use.
3.4. Monolayer
Culture
The following procedure is adaptable to a variety of cell lines in either plastic flasks or dishes: 1. With a 10 mL pipet, aspirate medium and floating cells from a confluent monolayer, and discard. 2. Add 4 mL of fresh complete SFJMto a T-25 cm2 flask (12 mL to a T-75 cm2 flasks, and 30 mL to a T-150 cm2 flask). 3. Resuspend cells by pipeting the medium forcefully across the monolayer with a Pasteur pipet (or equivalent device).
Weiss et al.
84
Fig. 1. Photomicrograph of Sf9 cells grown in monolayer culture in Sf-900 II SFM, 24 h postplanting. The cells were planted at 3 x 105 cells/ml from suspension culture. The cells usually spread and adhere to the plastic substrate (polystyrene cell culture grade) within 1 h and attain full confluency about 5-6 d postplanting. 4. Observe cell monolayer using an inverted microscope to ensure complete cell detachment from the surface of the flask (see Note 1). 5. Perform viable cell count on harvested cells using trypan blue exclusion method. 6. Inoculate cells at 3 x 105cells/ml into appropriate T-vessel. 7. Incubate cultures at 28 z!z0S”C. 8. With slower growing cell lines, it may be necessary to feed the flasks on day 3 postplanting. This is done by aspirating spent medium from one side of the monolayer and refeeding the culture with fresh SPM, which is gently added to the side of the culture flask. 9. Subculture the flasks when the monolayer (cell sheet) reaches 80-100% confluency, which is approx 4-7 d postplanting. Figure 1 illustrates postplanting.
Sf9 cells grown in monolayer
3.5. Spinner
culture 24 h
Culture
1. Recalibrate the gradation marks on commercial spinner flasks using a graduated cylinder or volumetric flask as a reference.
Cell Culture in Serum-Free Media
85
2. Ensure that impeller mechanisms rotate freely and do not contact vessel walls or base due to insect cell sensitivity to physical shearing. 3. Four to six confluent 75cm2 monolayer flasks are required to initiate a lOOmL culture (4-5 flasks for the spinner culture and one to be used asa backup). Dilute the cell suspension to approx 3 x lo5 cells/ml in complete SFM. 4. Dislodge cells from the base of the flasks as described in Section 3.1, 5. Pool the cell suspension, and perform a viable cell count. 6. Dilute the cell suspension to approx 3 x lo5 viable cells/ml in complete growth medium. 7. For culture volumes of 75-100 mL, use a 100~mL spinner vessel. For volumes of 150-200 mL, use a 250 mL vessel, and for volumes of 2.0 L, use a 3.0-L spinner. 8. Stock cultures should be maintained in a 150 mL culture employing a 250 mL spinner vessel. The top of the paddles should be slightly above the medium allowing for additional aeration to the cultures. 9. Atmospheric oxygen/air equilibration is accomplished by loosening the side arm caps on the vessels (about 114of a turn). 10. Incubate spinner vessels at 28 f 0.5”C at a constant stirring rate of 75 rpm. 11, Subculture spinner cultures to approx 2-3 x lo5 cells/ml twice weekly in well-cleaned, sterile vessels. 12. Once every 2 wk, spinner cultures may be centrifuged (1OOg) for 5 min, and resuspended in fresh medium to reduce accumulation of cell debris and toxic waste byproducts (see Note 2).
3.6. Shaker Culture It should be noted that growing cells in shaker culture in lOO-mL volume with loosened caps is an ideal method because oxygen is not ratelimited under these conditions. It is recommended that the SFM formulatiqn contain 0.05-O. 1% Pluronic Polyol F-58 or a polyol equivalent in performance that prevents shearing. 1. Orbital shaker apparatus must have a capacity for 50-500-n& Erlenmeyer flasks and a shakmg speed of up to 140 rpm. 2. The standard flask employed is the 250~mL disposable sterile Erlenmeyer for a 100-r& volume. 3. The orbital shaker flask assembly should be maintained in a 28 f 0.5OC nonhumidified, ambient air-regulated environment. 4. Oxygenation/aeration is accomplished by loosening the cap approx l/4 turn (within the intermediate closure position). In this condition, there is no oxygen limitation to the cells, and therefore, they proliferate to maximum rates. 5. Inoculate a 250-r& Erlenmeyer flask with 100 mL of complete SFM containing 3 x lo5 viable cells/ml.
86
Weiss et al.
Fig. 2. Photomicrograph of Sf9 cells grown in suspension culture using 100~mL shaker flasks and planted in Sf-900 II SFM, on day 3 postplanting. The cells were planted at 3 x lo5 cells/ml in a 250-mL plastic Erlenmeyer flask, and placed on an orbital shaker platform set at 135 rpm, and incubated at 28 rfi 0.5”C. The cap of the flask was loosened to provide adequate aeration (see Section 3.6.). 6. Set the orbital shaker at 135 rpm for cultures maintained and adapted to suspension culture in SFM. 7. Incubate the culture in SFM until it reaches l-3 x lo6 viable cells/r& Then split the shake flask cultures to approx 3 x 16 viable cells/ml. For consistent optimal cell growth, the culture should be in midlog phase of growth when subcultured. 8. Once every 3 wk, cell suspension from shaker cultures may be gently centrifuged (1OOg) for 5 min, and the cell pellet resuspended in fresh SFM to reduce accumulation of cell debris and toxic waste byproducts (see Note 3). Figure 2 shows Sf9 cells grown in Sf-900 II SFM using shaker culture. 3.7. Cryopreservation ofserum-Free Cultures 3.7.1. Freezing Method for Serum-Free Cultures 1. Grow desired quantity of cells in suspension using either spinner or shaker culture, harvesting in midlog phase of growth at a viability of >90%.
Cell Culture in Serum-Free Media
87
2. Determine the viable cell count using trypan blue exclusion dye method, and calculate the required volume of cryopreservation medium required to yield a final cell density of 1.O-2.0 x lo7 cells/ml. 3. Prepare the required volume of cryopreservationmedium consisting of 7.5% DMSO in 50% fresh SFM and 50% conditioned medium (SIerile filtered). Chill the newly prepared freeze medium and hold at 4°C. 4. Centrifuge cells from suspension or monolayer culture medium at 1OOgfor 5 min. Resuspend cell pellet in the determined volume of chilled cryopreservation medium. Ideally, the cell suspension in cryopreservation media should be stirred with a small magnetic bar at very low speed to maintain a homogenous suspension. 5. Dispense aliquots of this suspension while mixing into cryoviah according to manufacturer’s specifications (i.e., 4.5 mL to a 5.0-n& cryovial). 6. Refrigerate cryovials at 4°C in a refrigerator for 30-45 min. Just before freezing, gently mix the cell suspension in the cryovials to achieve a homogenous cell suspension. 7. Achieve cryopreservation in an automated or manually controlled rate freezing apparatus following standard procedures. For ide,al cryopreservation, the freezing rate should be 1“C/min. One can gradually freeze cells manually using a -20°C freezer, dry ice, a -7OOC freezer, and finally liquid nitrogen. 8. Frozen cells are stable indefinitely in liquid nitrogen storage.
3.7.2. Recovery of Cells from Liquid Nitrogen 1. Recover cultures from frozen storage by rapidly thawing a vial of cells in a 28 f. 0.5”C water bath. Transfer the entire contents of the vial into a 250-n& shaker flask containing 100 mL complete SFM, and mcubate culture as per Section 3.6., steps l-7. 2. Maintain culture between 3 x 105and 1 x lo6 cells/ml for the first two subcultures after recovery, thereafter returning to the normal maintenance schedule. 3. For additional methodologies on cryopreservation of insect cells, refs. (I) and (20) in this chapter should be reviewed (see Note 5).
3.8. Comparison of Growth in Different Serum-Free Media Identical replicate suspension cultures of Sf9 cells in Grace’s Supplemented Medium and 10% heat-inactivated FBS were adaptedto six commercially available SFM formulations using the direct adaptation protocol described in Section 3.2. of this chapter. After ten consecutive passagesin each medium, maximum cell growth was determined. Figure 3 ,surnmarizes cell growth in various SFM formulations that are compared to a
88
Weiss et al.
:
90 ao-
0 .z-
70-
G
60
>
50-
iz
40
-
30
-
20
-
10
-
-
o-
Fig. 3. Growth of Sf9 cells in various SFM formulations using shakerculture method. Maximum cell densitieson day 7 postplanting. serum-supplemented control. The 100~mL shake flask SFM cultures attained maximum densities on day 7 postplanting of 5.3-10.4 x lo6 cells/ml vs 4.3 x lo6 cells/ml in the serum-supplemented control. The best serum-free growth was obtained in SF-900 II SFM (GIBCO/BRL Life Technologies, Inc., Gaithersburg, MD). Suspensioncultures of Tn-368 and Ld652Y cell lines were adaptedto Sf900 Il SFM using the sequential(weaning) adaptationprocedure outlined in Section 3.3. of this chapter. Cell growth has been monitored for >25 consecutive serum-free passages.Figures 4 and 5 show serum-free growth in Sf-900 II SFM of Tn-368 and Ld652Y, respectively. We generally observe significantly higher maximum cell densities with these invertebrate cell lines in Sf-900 II SFM than in serum-supplemented control cultures.
Cell Culture
in Serum-Free
,o
_
Media
89
0 GRACES + 10% FBS l Gl6CO Sf-900 II SFM
0
1
2
3
DAYS
4
5
6
7
8
POST-PLANTING
Fig. 4. Growth kinetics of Tn-368 cells in SFM vs serum-containing medium in suspension using shaker culture methods.
50%
e-0-a /
10
0
GRACES
0
GIBCO
+
10%
FBS
Sf-900
II SFM
01
’
’
’
’
’
’
’
’
’
0
1
2
3
4
5
6
7
8
9
DAYS
10
POST-PLANTING
Fig. 5. Growth kinetics of Lymantria dispar (Ld652Y) cells in SFM vs serum-containing medium in suspension using shaker culture methods.
90
Weiss et al. 600000
-
700000
-
600000
-
500000
-
400000
-
300000
-
200000
-
100000
O-
Fig. 6. Production of rp-Galactosidase in Sf9 suspension cells grown and infected in various S2PM and serum-supplemented control culture. The infection with rAcNPV and production of recombinant protein was carried out in lOOniL shaker flask cultures. Recombinant P-Galactosidase values are from day 4 postinfection. Sf9 cells grown in SFM for 10 passages using lOO-mL shaker cultures were infected with rAcNPV expressing recombinant P-Galactosidase (@Gal), Clone VL-941, at a multiplicity of infection (MOI) of 5.0 when the density of the cultures reached 2.5-3 x lo6 cells/ml. Figure 6 documents maximum $-Gal expression on day 4 postinfection. The range in various SFM was 185,875-549,925 U/mL, compared to 276,225 U/mL in the serum-supplemented control. However, maximum serum-free $-Gal expression was observed with Sf-900 II SFM.
91
Cell Culture in Serum-Free Media Table 1 Replication of Autographa californica Nuclear Polyhedrosis Virus (AcMNPV) in Sf9 and Tn-368 Cells Grown in Grace’s Supplemented Media + 10% PBS or Sf-900 II SPM Cell line Sf9
Tn-368
Growth media“
Culture densityb
Viral MO1
CountC
Grace’s + 10% PBS Sf-900 II SFM Grace’s + Sf-900 II SlW
3.0 x 106
0.5
1.7 x 107
1.0 x 10s
2.9 x 106
0.5
3.5 x 107
2.0 x 10s
7.5 x 105 7.5 x 10s
5.0 5.0
1.5 x 107 5.6 x lo7
3.0 x 107 1.0x 10s
Titefl
‘%race’s insect cell culture medium, supplemented (TNM-FH) + 10% FHS. bDay 0 postinfection. “Occluded virus count.On day 6 postinfection for St-9and day 5 postinfection for Tn-368 cultures. dNonoccluded virus titer. On day 3 postinfection for both cultures asdetermined by plaque assay. the growth and virus infection were performed using cells from lOO-mL shaker cultures.
Table 1 represents occluded and cell-free, nonoccluded virus titers of Sf9 and Tn-368 cultures, respectively, infected with wild-type ,4cNPV. The serum-free cultures show significantly higher virus yields with both cell lines when compared to the yields produced in the serum-supplemented cultures. 3.9. Serum-Free Cell Culture Fundamentals Maintaining a constant, nonrate-limiting in vitro environment is essential for serum-free growth. The low protein or protein-free nature of most SPM means that the cells are even more sensitive to biophys ~a.l conditions, Poor water quality, bioburden (i.e., endotoxin), or improper cleaning or processing of equipment (i.e., spinner flasks) can be extremely toxic to serum-free cultures resulting in suboptimal cell growth or even death. Because stationary and suspension cultures are maintained in nongasregulated incubators, airborne contaminants (i.e., fumes from nearby chemical hoods or floor cleaning agents) can also causeserious pl*oblems. Two common problems involve improper storage of SPM. Invertebrate SPM formulations contain B vitamins, lipids, and other components that are light-sensitive and/or vulnerable to oxidation if exposed to air. We recommend that all SPM be stored away from light, preferably behind black plastic or in closed boxes. If the same bottle of SFM is to be used over a
92
Weiss et al.
long period (>2 wk), aseptically overlay the media with N2 gas and tightly cap the bottle after each use to prevent oxidation of medium components. Another common problem is the supplementation of SFM with surfactants, such as Pluronic Polyol F-68. Many researchers have demonstrated improved cell growth in suspension culture by supplementing serum-containing media with 0.1% Pluronic Polyol F-68. They then continue the practice after switching to SFM. However, commercially available SFM are complete (unless otherwise indicated) and contain surfactant to protect cultures from shear stress. Adding more surfactant, in effect doubling the concentration, is toxic to many insect cell lines. Most of the commercial media formulations that we evaluated are complete and ready to use. The optimal pH and osmolarity for use with lipidopteran cell lines are 6.0-6.4, and 350-375 mosM, respectively. All commercial SFM fall within these recommended values with the exception of the JBH EX-CELL 400 formulation, which had an osmolarity of 395400 mosM. The high sucroseconcentration (16.5 g/L) is the probable reason for the elevated osmolarity. It is of paramount importance to monitor pH and osmolarity of any serum-free or serum-supplemented media, especially when making liquid media from powder. By adhering to acceptable ranges,consistent growth and fewer unexplained problems will occur. We have observed that the majority of insect cell lines in SFM attach more tightly to the substrate and, therefore, are more difficult to dissociate when subculturing. With Sf9 or Sf21 cells, one must scrape or vigorously shake the culture to remove the cells. This greatly decreases the viability of the culture and increases the chances of contamination. It is recommended that cells lines be adapted to suspension culture first and then to SFM since the 5-20% FBS in standard serum-supplemented culture media provides additional protection for the cells against shear stress. We have successfully adapted to suspension culture the Sf9, Sf21, Trichoplusia ni (Tn-368) cabbage looper, Lymantriu dispar (Ld652Y) Gypsy moth, Heliothis zeu (H. zea), and Orgyiu leucostigmu (white-marked tussock moth) cell lines. 3.10. Infection of Serum-Free Cultures with Wild-Type or Recombinant Baculoviruses The key to a successful serum-free infection of insect cells with either wild-type or recombinant baculovirus is that the culture not be rate-limited by nutritional or biophysical factors (e.g., pH, dissolved O2 tempera-
Cell Culture in Serum-Free Media
93
ture). In addition, the culture should be infected in the midlogarithmic phase of growth with an established MOI. It has been reported that the standard serum-supplemented medium utilized for virus infection is rate limited if the cells are infected at densities >2 x lo6 cell&L (29). The same nutritional SFM (26).
limitations
are observed with most commercial
When producing a nonoccluded virus stock, recombinant or wild-type, it is recommended that the culture be infected at a cell density of 2 x lo6 cell/n-L with an MO1 of 0. l-l .O (7,13). With Sf-900 II SFM, successful recombinant infections have been observed with higher cell densities (>4 x lo6 cells/ml) without a significant drop in specific productivity (16,26). With BEVS, the rDNA product being expressed may be a secretedor nonsecretedprotein. Maximum expression of secretedproteins is usually observed 48-72 h postinfection, and nonsecreted proteins 7296 h postinfection. It is important to determine the expression kinetics of each rDNA product, since many proteins (secreted or nonsecreted) are rapidly degraded by cellular proteases released from cells. This will enable one to establish an optimal time for harvesting the culture.
4. Notes 1. Insect cells generally attach very well to substrate in SFM and require extra effort for detachment. It may be necessary to shake the flask vrgorously two or three times to dislodge the cells. The caps of flask should be tightened before shaking vigorously. 2. Although spinner culture is scalable to a degree, there is a physical constraint with particular applications owing to the requirement of adequate gas partitioning in the culture. A general rule is to keep the volume in the spinner vessel below l/3 full, and provide more and larger ports for headspace gas exchange as the vessel size increases. 3. Although shaker-type culture is scalable to a variety of subsequent vessels, volumes, and flask sizes, each has its own growth characteristics. Therefore, relative flask fill volumes and orbital shaker speeds musl be optimized for each configuration. 4. Viability and recovery of cryopreserved cells should be checked 24 h after storing vials in liquid nitrogen by following the recovery procedures described in the following section. 5. For safety, when removing cryovials from liquid nitrogen storage, always use a face shield to prevent possible injury that may be causecl by vials occasionally exploding because of the rapid temperature shift.
94
Weiss et al.
Acknowledgments The authors are grateful to Terrilyn M. Summers, Senior Secretary (Life Technologies, Inc.), for preparation of this manuscript and to MaryLynn Tilkins (Life Technologies, Inc.) for her invaluable technical assistance. We thank Max Summers (Texas A&M) for providing recombinant AcNPV and recombinant-galactosidase and Dr. E. Dougherty (USDA) for the Ld652Y and Tn-368 cells. References 1. Weiss, S. A. and Vaughn, J. L. (1986) Cell culture methods for large-scale propagation of baculoviruses, in The Biology of Baculoviruses (Granados, R. R. and Federici, B. A., eds.), CRC, Boca Raton, FL, pp. 64-87. 2. Weiss, S. A., Smith, G. C., Kalter, S. S., Vaughn, J. L., and Dougherty, E. (1981) Improved replication of Autographu californica NPV in roller bottles: Characterization of the progeny virus. Zntervirology 15,213-222. 3. Agatos, S. N., Jeong, Y. H., and Venhat, K. (1990) Growth kinetics of free immobilized insect cell cultures. Ann. NYAcad. of Sci. 589,372-398. 4. Luckow, V. A. and Summers, M. D. (1989) High level expression of nonfused foreign genes with Autographa californica nuclear polyhedrosis virus expression vectors. J. Viral. 170,3 l-39. 5. Inlow, D., Shauger, A., and Maiorella, B. (1989) Insect cell culture and baculovirus propagation in protein-free medium. J. Tissue Culture Methods 12(l), 13-16. 6. Smith, G. E., Summers, M. D., and Fraser, M. J. (1983) Production of hurnan P-interferon in insect cell infected with a baculovirus expression vector. J. Mol. Cell. Biol. 3(12), 2156-2165. 7. Weiss, S. A., Belisle, B. W., DeGiovanni, A., Godwin, G., Kohler, J., and Summers, M. D. (1989) Insect cells as substrates for biologicals. J. Dev. Biol. Standard 70,271-279.
8. Luckow, V. A. (1991) Cloning and expression of heterologous genes in insect cells with baculovirus vectors, in Recombinant DNA Technology and Applications (Prokop, A., Bajpai, R. K., and Ho, C. S., eds.), McGraw-Hill, New York, pp. 97-151. 9. Dolin, R., Graham, B. S., Greenberg, S. B., Tacket, C. O., Belshe, R. B., Midthun, K., Clements, M. L., Gorse, G. J., Horgan, B. W., Atrnar, R. L., Katzon, D. T., Bonnez, W., Fernie, B. F., Montefiori, D. C., Stablein, D. M., Smith, G. E., and Koff, W. C. (1991) The safety and immunogenicity of Human Immunodeficiency Virus Type I (HIV-I) recombinant gp160 candidate vaccine in humans. Ann. ofZnt. Med. 114(2), 119-127. 10. Luckow, V. A. and Summers, M. D. (1988) Trends in the development of baculovirus expression vectors. J. Biotechnol. 6,47-55. 11. Co&ran, M. A., Ericson, B. L., Knell, J. D., and Smith, G. E. (1987) Use of baculovirus recombinants as a general method for the production of subunit vaccines, in Vaccines 87 (Chanock, R. M. et al., eds.), Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, pp. 384-388. 12. Mairoella, B., Inlow, D., Shauger, A., and Harano, D. (1988) Large-scale insect cell-culture for recombinant protein production. J. Biotechnol. 6, 1406-1410.
Cell Culture
in Serum-Free
Media
95
13. Weiss, S. A., Gortien, S., Fike, R., DiSorbo, D., and Jayme, D. (1990) Large-scale production of proteins using serum-free insect cell culture, in Biotechnology: The Science and the Business. Proceedings for Ninth Australian Biotechnology Conference, Queensland, pp. 220-23 1. 14. Weiss, S. A., DeGiovanni, A., and Godwin, G. (1988) Use of insect cells in biotechnology. In Vitro Cellular and Developmental Biology 24(3), P 53A. 15. Weiss, S. A., Grefrath, P., Whitford, W. G., Pfohl, J., Fike R. M., and Jayme, D. W. (1990) Growth of Insect Cells in a Serum-Free Medium and Production of Recombinant Proteins Using Various Bioreactors. In Vitro Cell. Devel. Biol. 26(3), 30A.
16. Weiss, S. A., DiSorbo, D. M,, Whitford, W. G., and Godwin, G. (1991) Improved production of recombinant proteins in high density insect cell culture. In Vitro Cell. Devel. Biol. 27#3,42A.
17. Mitsuhashi, J. (1982) Media for insect cell culture, in Advances in Cell Culture (Maramorosch, K., ed.), Academic, New York, pp. 133-199. 18. Inlow, D., Shauger, A., and Maiorella, B. (1989) Insect cell culture and baculovirus propagation in protein-free medium. J Issue Culture Methods 2(l), 13-16. 19. Hink, W. F., Ralph, D. A., and Joplin, K. H. (1985) Metabolism and characterization of insect cell cultures, in Comprehensive Insect Physiology, Biochemistry and Phurmacology (Karkut, G. A. and Gilbert, L. I., eds.), Pergamon, New York, pp. 547-570. 20. Weiss, S. A., Smith, G. C., Kalter, S. S., and Vaughn, J. L. (1981) Improved method for the production of insect cell cultures in large volume. In Vitro 17,495-502. 21. Weiss, S. A., Peplow, D., Smith, G. C., and Goodwin, R. H. (1984) Replication of Heliothis zeu baculovirus in insect cells grown in serum-free media (SFM). In Vitro Cell. and Devel. Biol. 20,27 1. 22. Wilkie, G. E. I., Stockdale, H., and Pitt, S. V. (1980) Chemically-defined media for the production of insect cells and viruses in vitro. J Dev. Biol. Stand. 46,29-37. 23. Roder, A. (1982) Development of a serum-free medium for cultivation of insect cells. Naturwissen-schafen 69,92,93. 24. Miltenburger, H. G. (1983) Investigation on cultivation of insect cell lines in serum-free media. European Conference on Hormonally Defined Media 1,31-43. 25. Weiss, S. A., Godwin, G. P., Gorflen, S. F., and Whitford, W. G. (1992) Viral Pesticides: In vitro process development. Proceedings for 10th Annual Australian Biotechnology Association Conference (Melbourne). 26. Weiss, S. A., Godwin, G. P., Got-hen, S. F., and Whitford, W. G. (1992) Serum-free media, in Insect Cell Culture Engineering (Goosen, M. F. A., Daugulis, A., and Faulkner, P., eds.), Marcel Dekker, New York, pp. 179-194. 27. Vaughn, J. L. and Weiss, S. A. (1991) Formulating media for the culture of insect cells. BioPharm 4#2, 16-19. 28. Vaughn, J. L. and Weiss, S. A. (1991) Large-scale propagation of insect cells, in Large-Scale Mammalian Cell Culture Technology (Lubiniecki, A. S., ed.), Marcel Dekker, New York, pp. 597-617. 29. Wood, H. A., Johnston, L. B., and Burand, J. P. (1982) Inhibition of Autographa califomica nuclear polyhedrosis virus replication in high-density Trichoplusia ni cell cultures. J. Virol. 119,245-254.
CHAPTER5
Transfection Techniques for Producing Recombinant Baculoviruses Karen
Munkenbeck
D-otter
and H. AZan
Wood
1. Introduction The development of transfection procedures that optimize the efficiency of recombination events is central to the isolation of recombinant baculoviruses for foreign gene expression. A variety of transfection methods, which were first developed for mammalian cell lines, have been adapted for use with insect tissue-culture cells. Conditions required for the maximum transfection efficiency of any vertebrate or invertebrate cell line vary considerably. This chapter will focus on the methods and conditions that have been employed with lepidopteran tissue-culture cells. In general, these conditions have been developed for efficient transfection of Bombyx mori (Bm-N), Spodopteru frugiperdu (IPLB-SF-21AE), and SF9 (a subclone of IPLB-SF-21-AE) cell lines. However, we have also successfully used these procedures for transfection of Lymantria dispar (IPLB-LD-652Y) and Trichoplusia ni (TN-368) tissue-culture cells. The medium in which lepidopteran cells are grown, e.g., TNMFH, TC-100 or serum-free media, may influence the transfection efficiency. For the isolation of baculovirus expression vectors, cotransfection is generally performed using viral DNA and a bacterial plasmid DNA that functions as a transfer vector. The transfer vector contains a foreign gene inserted downstream from a baculovirus gene promoter and a sufficient amount of viral flanking DNA sequences to From: Methods in Molecular Biology, Vol. 39. Bacuiovirus Expression Protocols Edited by: C. D. Richardson (0 1995 Humana Press Inc., Totowa, NJ
97
Trotter and Wood facilitate recombination between the plasmid and viral DNA sequences (allelic transplacement). Most research has been conducted using allelic transplacement of the polyhedrin gene with foreign gene inserts. Alternatively, attempts have been made to generate recombinant baculoviruses using transfection with a transfer vector before or after infection with nonoccluded virus particles. This approach generally produces a very low percentage of recombinant virus progeny. The most widely used transfection technique has been the calcium phosphate precipitation method of Graham and Van der Eb (I). In this system, calcium and phosphate ions react to form a precipitate that complexes with DNA and cell surfaces. The exact mechanism of the DNAcalcium phosphate complex transfer into the cells has not been elucidated. Other adsorptive agents, including DEAE-dextran, poly-Llysine, polyornithine, and polybrene, have been used to transfect insect and vertebrate tissue-culture cells (2). The efficiency of transfection with these agents may be improved by the addition of carrier DNA or osmotic shock with dimethylsulfoxide or glycerol. More recently, lipofection and electroporation procedures have been developed which, in many cases, have provided improved transfection efficiencies. Lipofection, first reported by Felgner et al. (3,4), employs a cationic lipid, N-[ 1-(2,3-dioleyloxy)propyl]-N,N,N-trimethylammonium chloride (DOTMA), to form liposomes. The liposome complex sequesters the DNA with 100% efficiency. The positive charge on the lipid results in adsorption to the negatively charged cell membrane, facilitating a fusion event that delivers the DNA into the cell. Electroporation was first used for DNA transfer into tissue-culture cells by Neumann et al. (5). Electroporation takes advantage of the fact that the cell membrane acts as a capacitor which is unable to pass current. When the cell is subjected to a brief high-voltage electrical field, the membrane temporarily breaks down creating nanometer-sized pores through which the DNA passes.Closing of the pores is a natural process that is delayed at 0°C. Optimizing transfection is of prime importance in transient assays for the study of gene expression. For the generation of baculovirus expression vector systems, the primary consideration is optimizing conditions that promote maximum rates of recombination and, therefore, the highest proportion of recombinant virus progeny. Accordingly, it should be recognized that the higher transfection rate by electroporation as
Transfection
Techniques
99
compared to the calcium phosphate procedure is not the prime consideration for the isolation of baculovirus expression vectors. 2. Materials 2.1. Insect
Tissue
Culture Cells The most commonly used cell lines are Spodoptera frugiperda
cell lines IPLBSF21AE (X21) (6) and its subclone IPLB-SF9 (SF9) (7), which are available from American Type Culture Collection. Other insect cell lines have been used to produce recombinant baculoviruses, including Bombyx mori (8), Heliothis zea (9), Trichoplusia ni (TN-368), Manduca sexta, Malacosoma disstria (IO), and Lymantria dispar (IPLB-LD-652Y) (II), Cell cultures should be maintained in log phase since many insect cell cultures undergo extensive clumping when they are allowed to reach stationary phase. At high cell densities, insect tissue-culture cells exhibit a cell-to-cell contact inhibition of viral and host DNA synthesis (12). Highdensity cell cultures produce low amounts of virus progeny, and few polyhedra are visible. Accordingly, care should be taken to avoid high cell densities. In the procedures listed below, all cell numbers are viable cell counts. Viable cell numbers are determined by mixing a 50-100 pL aliquot of cells with an equal volume of trypan blue (0.5% w/v). Viable cells do not stain blue. The concentration and percent viability of cells are determined using a hemocytometer. 2.2. Electroporation Apparatus and Materials 1. HBS buffer (HEPES buffered saline): 20 mM HEPES, pH 7.05, 1 mil4 NaHP04, 5 mA4 KCl, 140 mM NaCl, and 10 mM glucose. Filter-sterilize and store at 4°C. 2. 1-2 pg viral DNA and 5 p.g transfer vector DNA. 3. Sterile electroporation cuvets (GIBCO/BRL cat. # 11601-010). 4. Electroporation Device (GIBCO/BRL Cell PoratorTM). 5. Log phase tissue-culture cells. 2.3. Reagents for Calcium Phopsphate Transfections 1. HBS buffer: 20 rnM HEPES, 1 rnM NaHP04, 5 rnM KCl, 140 mM NaCl, and 10 r&f glucose, pH 7.05. The pH of this buffer is critical when using the CaC12procedure. Adjust the pH of the buffer, let it sit overnight, and then recheck the pH. Repeat until the pH is stable. Then filter-sterilize and check the pH again. Store this buffer at 4OC and use at room temperature.
100
Trotter and Wood
2. 2.5M CaC12filter-sterilized. 3. l-2 pg viral DNA and 2-5 pg transfervector DNA. 4. Log-phasetissue-culturecell. 1. 2. 3. 4.
2.4, Reagents for Transfections with Lipofectin l-2 pg viral DNA and 2-5 pg transfervector DNA. LipofectinTM(GIBCO/BRL catalogue# 18292),storedat 4°C (seeNote 1). Tissue-culturemedia--complete (with FBS) and incomplete (serum-free) medium. Log-phasetissue-culturecells.
3. Methods 3.1. Preparation of Viral DNA One of the most important factors in determining the success of transfection is the quality of the DNA. Ethanol precipitation of viral DNA significantly reduces the efficiency of transfection and should be avoided. Baculovirus genomes are typically 100-130 kb in size and should be isolated using procedures that minimize shearing. Because transfection with the supercoiled form of viral DNA is more efficient than with the nicked circle or linear forms, we have included an optional procedure for cesium chloride/ethidium bromide gradient purification of the supercoiled form. However, using viral DNA which had been linearized by a restriction enzyme cut within the polyhedrin gene,Kitts et al. (13) achieved polyhedrin allelic transplacement frequencies approaching 30%. 3.1.1. Nonoccluded Viral DNA An efficient method for isolation of viral DNA from nonoccluded virions from the cell culture supernatant is as follows: 1. Centrifugethe supernatantat 1OOOg for 10 min to pellet cells andcell debris. 2. Pellet the virions at 54,000g(20,000rpm in BeckmanSW 28 rotor) for 1 h at 18OC.Resuspendthe virus pellet in a small volume of 10 mM TMO.1 mM EDTA, pH 7.2. 3. Add proteinaseK (20 mg/mL stock) and SDS (10% stock) to a final concentration of 2 mg/mL and O.l%, respectively. Incubate the mixture at 55°C for 1-16 h. 4. Extract the DNA twice with an equal volume of phenol/chloroform (1: 1). It is important to use high-quality nonoxidized phenol. The phasesshould be thoroughly mixed by gently inverting the tube. Vortexing should be avoided becauseof shearforces. Separatethe phasesby centrifuging at 12,000gfor 5 min. Gently pipet the aqueousphaseinto a fresh tube. Use
Transfection
101
Techniques
wide-bore pipets, andpipet slowly, sincethe viral DNA can be shearedby pipeting through narrow apertures. 5. Extract the aqueousphasewith chloroform/isoamyl alcohol (24:l). 6. Dialyze the aqueousphasecontainingthe DNA against1000vol of 10 rnM Tris, 1 mM EDTA,
pH 8.0, for 18-24 h with three changes of buffer.
3.1.2. Occluded Viral DNA
Alternatively, viral DNA can be isolated from virions occluded in polyhedra. This is not recommended with tissue-culture-derived polyhedra because of the costs of obtaining sufficient numbers of polyhedra. 1. Pellet lo*-log polyhedraat 10,OOOg for 10 mm. 2. Resuspendthe polyhedra in 5-10 rnL of 50 mM Na&JOs, and incubated for 15 rnin (until clearing occurs)at room temperature. 3. Centrifuge the mixture at 10,OOOg for 10 min to remove undissolvedpolyhedra and contaminants. 4. Pellect the releasedvirions, andextract the viral DNA asdescribedin Section 3.1.1., step 2. 3.1.3. Purification
of Supercoiled
Viral DNA.
Bring the DNA preparation to a final concentration of 0.25 mg/mL ethidium bromide. Layer the sample onto a 50% (wt/wt) CsCl solution and centrifuge 24 h at 100,OOOg.The upper band (1.54 g/mL) contains linear and relaxed circular DNA, and the lower band (1.58 g/n&) contains the supercoiled DNA. Remove the lower band using a large-bore needle. Extract the ethidium bromide with butanol. Dialyze the DNA solution against 1000 vol of 10 mM Tris/l mM EDTA, pH 8.0, for 18-24 h with three changes of buffer. 3.2. Vector DNA Preparation The vector plasmid DNA can be prepared by any of the standard plasmid extraction procedures and should be purified on a CsCl/ethidum bromide gradient. When using the lipofection method, miniprep DNA can be used. The transfection and recombination efficiencies are reduced using miniprep DNAs, but the low proportion of recombinants expressing a screening marker, such as P-galactosidase,can be identified and purified. 3.3. DNA Storage The plasmid and viral DNA preparations can be stored in sterile tubes at 4°C for several months. For long-term storage, the DNA can be frozen at -20°C in a nonfrost-free freezer or at -80°C. Repeated freezing and thawing should be avoided since this creates nicks in the DNA.
102
1. 2.
3.
4. 5. 6. 7. 8.
Trotter and Wood 3.4. Electroporation Suspend the cells and centrifuge at 5OOgfor 5 min to pellet. Resuspend in HBS buffer to an approximate concentration of 2 x lo6 cells/ml. Transfer 2 x lo6 viable cells (see Section 2.1.) in 1 mL to the electroporation cuvet, immediately add the viral and vector DNA (~50 PL), and mix by inversion. Incubate the cells on ice for 5 min, and then mix again by inverting the cuvet several times. Place the cuvets in the electroporation device, and apply the electrical pulse. The optimal field strength and pulse duration vary with cell lines. For SF9 cells, the most effective parameters are a pulse length of 2.8 ms and a field strength of 750 V/cm (14). For IPLB-LD-652Y cells, the pulse length was increased to 7.7 ms at 750 V/cm (Yu and Wood, unpublished data). Following the pulse, the cuvet is placed on ice for 10 min to allow for recovery of the cells and uptake of DNA. Take an aliquot to determine viability using the trypan blue procedure (seeSettion 2.1.). Optimum transfection occurswith a cellular survival rate of 20-50%. Dilute the remaining cells with 1 mL of tissue-culture medium, dispense mto a 35-mm well (e.g., Falcon 3046,6-well plate), and place in the tissueculture incubator for 2 h. Remove the medium, add 2 mL of fresh medium, and incubate the culture. Harvest the supematant 48 h after electroporation. After 2 d, the percentage of recombinant virus progeny in the supematant is usually reduced because wild-type virus titers increase at a faster rate than the recombinant virus.
3.5. Calcium Phosphate Mediated Transfections 1. Seed a 6-well plate (e.g., Falcon 3046) with 1 x lo6 viable cells (see Section 2.1.) per well in 2 mL of tissue-culture medium. Place in an incubator for l-2 h to allow cells to attach to the plate. 2. Prepare viral and transfer vector DNA mixture by bringing the volume of the DNA up to 950 pL with HBS buffer in a 1.5-mL microfuge tube. Create a precipitate by adding 50 pL of CaC12 dropwise while slowly vortexing. This 1s a critical step m the procedure, because the speed of mixing determines the size of the precipitate made. One method to create a very fine precipitate is to set the vortex on the lowest speed. Another alternative is to lift the tube off the vortex bnefly while adding the CaC&. A third method is to bubble filtered nitrogen or air through the solution as the CaC12is added dropwise. 3. Allow the DNA-CaC& mixture to at at room temperature for 30 min to form the precipitate.
Transfection
Techniques
103
4. Remove the culture medium from the cells, and add the DNA-CaC& solution (see Note 2). 5. Place cells in an incubator for 1 h (see Note 3). 6. Add 1 mL of culture medium, and incubate for 4-6 h. There should be a “peppered” look-tiny dark granules on and around the cells. Gently remove the transfection buffer, replace it with 2 mL tissue-culture medium and incubate the culture for 2-3 d. 7. Harvest the supernatant and assayfor recombinant virus (see Note 4). 1.
2.
3. 4.
5. 6.
3.6. Lipofection Seed a 25-cm2 flask with 2.4 x lo6 viable cells (see Section 2.1.) in 5 mL of incomplete tissue-culture medium. Place in an incubator for l-2 h to allow cells to attach to the plate. Alternatively, seed a 25-cm2 flask with 1.2 x lo6 viable cells in 5 mL of complete medium, and incubate overnight. Mix the viral DNA and the transfer vector DNA, and bring the volume to 100 p.L with incomplete medium in a polystyrene tube large enough to hold 3 mL. Dilute 30 uL LipofectinTM reagent with 100 pL of incomplete medium, add to the diluted DNA mixture, and mix gently. Let sit at room temperature for 15 min, and then add 3 mL incomplete medium. (Increased LipofectinTM concentrations can kill cells.) Rinse the cells twice with incomplete medium, discard the medium, and replace with the 3 mL DNA/LipofectinTM mixture. Incubate 3-18 h, and then add an equal volume of complete tissue-culture medium containing 20% fetal bovine serum (normal serum concentration = 10%). Alternatively, remove the medium and replace it with 6 mL of complete medium. At 18 h, there should be tiny granule-like spots on and around the cells giving a peppered look to the culture when examined in an inverted light microscope. Continue incubation for 2-3 d. Harvest the supernatant, and assay for recombinant virus.
The efficiency of transfection may be increased lo-fold by washing the cells in Grace’s medium prior to transfection (15). The lipofection procedure requires less stringent conditions than the calcium phosphate method and no specialized equipment. More importantly, we have found that the lipofection method typically produces progeny virus populations containing 15% recombinant virions. If we plate the transfected cells onto a lawn of healthy cells in a plaque assay, 3WO% of the resulting
plaques contain some recombinant (P-galactosidase-producing) virus. Typically, we have obtained 5- to lo-fold lower levels of recombinants with the calcium phosphate and electroporation methods.
104
Trotter and Wood 3.7. ZnfectionlTransfection
Because of the ease and efficiency of infecting cells with the nonoccluded form of baculoviruses, infection before or after transfection with a transfer vector has been used to avoid the need to isolate highquality viral DNA. This procedure was optimized to isolate recombinant Lymantria dispar baculoviruses. Lymantria dispar cells were inoculated with five plaque-forming units (pfu) per cell, and at 4 h postinfection (pi), the cells were subjected to electroporation with transfer vector DNA (Yu and Wood, unpublished data). Similary, SF21 cells inoculated with 1 pfukell of AcMNPV were subjected to lipofection procedures with transfer vector DNA at 1 h pi. In both cases,the recombination frequency was extremely low (0.04-0.06%), and the expression of P-galactosidase was required to identify the recombinants. Preinfection of SF9 cells has been used with the calcium phosphate method of transfection (16). 4. Notes 1. GIBCO/BRL now sells anadditional lipofection reagent,TransfectACEm, which producestransienttransfectionfrequenciesof >90% with mammalian tissue culture cells (17). 2. A modification of the calcium phosphatemethod is to form the DNACaC12precipitatein the tissue-culturemedium. The DNA is diluted in 0.5 mL HEPES-buffered saline (25 mM HEPES, pH 7.1,140 mM NaCl) containing 125 mk! CaC12.The DNA-CaC12mixture is addeddropwise to tissue-culturecells. The phosphatein the tissue-culturemedium complexes with the DNA-CaCl* to form a precipitate (18). 3. Peakmanet al. (19) reportedthat short-wavelengthUV irradiation of cells increasedthe recombination frequency 2.6-fold when using the calcium phosphatemethod of transfectionof SF9 cells. 4. For calcium phosphatetransfectionof Bombyx mori cells, Maeda (8) used a phosphatebuffer containing 50 mM Hepes (pH 7.1), 0.28M NaCl, 0.7 mM Na2HP0,, and 0.7 mM NaH2P04. References 1. Graham,F. L. andVan der Eb, A. J. (1973) A new technique for the assay of infectivity of human adenovirus 5 DNA. Virology 52,456-467. 2. Walker, V. K. (1989) Gene transfer in insects, in Advances in cell culture, vol 7 (Maramorosch, K. and Sato, G., eds.), Academic, New York, pp. 87-124. 3. Felgner, P. L., Gadek, T. R., Holm, M., Roman, R., Chan, H. W., Wenz, M., Northrop, J. P., Ringold, G. M., and Danielsen, M. (1987) Lipofection: A highly efficient, lipid-mediated DNA-transfection procedure. Proc. Natl. Acad. Sci. USA 84,7413-7417.
Transfection
Techniques
4. Felgner, P. L. and Holm, M. (1989) Cationic liposome-mediated
105 transfection. Fo-
cus 11,21-25.
5. Neumann, E., Schaefer-Ridder, M., Wang, Y., and Hofschneider, P. H. (1982) Gene transfer into mouse myeloma cells by electroporation in high electric fields. EMBO .I. 1,841-845.
6. Vaughn, J. L., Goodwin, R. H., Tompkins, G. J., and McCawley, P. (1977) The establishment of two cell lines from the insect Spodopterafrugiperdn (Lepidoptera: Noctuidae). In Vitro 13,213. 7. Summers, M. D. and Smith, G. E. (1987) A Manual of Methods for Baculovirus Vectors and Insect Cell Culture Procedures. Texas Agric. Expt. Sta. Bulletin 1555. 8. Maeda, S. (1989) Gene transfer vectors of a baculovirus, Bombyx mori nuclear polyhedrosis virus, and their use for expression of foreign genes in insect cells, in Invertebrate Cell System Applications, vol. I (Mitshuhashi, J., ed.), CRC, Boca Raton, FL, p. 167. 9. Corsaro, B. G. and Fraser, M. J. (1988) Transfection of Lepidopteran insect cells with baculovirus DNA. J. Tissue Culture Method 12,7-l 1. 10. Burand, J. P., Summers, M. D., and Smith, G. E. (1980) Transfection with baculovirus DNA. Virology 101,286-290. 11. Yu, Z., Podgaite, J. D., and Wood, H. A. (1992) Genetic engineering of a Lymuntria dispar nuclear polyhedrosis virus for expression of foreign genes. J. Gen. Viral. 73,1509-1514. 12. Wood, H. A., Johnson, L. B., and Burand, J. B. (1982) Inhibition of Autographa californica nuclear polyhedrosis virus replication in high-density Trichoplusia ni cell cultures. Virology 119,245-254. 13. Kitts, P. A., Ayres, M. D., and Possee, R. D. (1990) Linearization of baculovirus DNA enhances the recovery of recombinant virus expression vectors. Nucleic Acids Res. 18,5667.
14. Mann, S. G. and King, L. A. (1989) Efficient transfection of insect cells with baculovirus DNA using electroporation. J. Gen. Viral. 70,3501-3505. 15. Hartig, P. C., Cardon, M. C., and Kawanishi, C. Y. (1991) Generation of recombinant baculovirus via liposome mediated transfection. BioTechniques 11,310-313. 16. Goswami, B. B. and Glazer, R. I. (1991) A simplified method for the production of recombinant baculovirus. BioTechniques 10,626-630. 17. Whitt, M. A., Buonocore, L., Rose, J. K., Ciccarone, V., Chytil, A., and Gebeyehu, G. ( 199 1) TransfectACErM reagent: Transient transfection frequencies >90%. Focus 13,8-12.
18. Guarino, L. A. and Summers, M. D. (1986) Functional mapping of a trans-activating gene required for expression of a baculovirus delayed-early gene. J. Virol. 57, 563-571. 19. Peakman, T., Page, M., and Gewert, D. (1989) Increased recombinational efficiency in insect cells irradiated with short wavelength ultra-violet light. Nucleic Acids Res. 17,5403.
CHAPTER6 Selection Martin
of Recombinant Baculoviruses by Visual Screening J. Page and Brian
C. Rodgers
1. Introduction The baculovirus expression system is now a widely accepted tool for the expression of recombinant proteins with many features to recommend it. Selection of recombinants is rapid compared to mammalian expression systems, whereas the capacity to perform many posttranslational modifications necessary for the expression of authentic, functional proteins is retained. Examples of glycosylation (I), phosphorylation (2), myristilation (3), palmitoylation (41, and carboxy-methylation (5) are among the many modifications reported. In addition, coexpression of related proteins in the baculovirus system appears to provide an ideal medium for the study of protein-protein interactions (6-8) and the assembly of virus particles (9-13). The fundamental tenet on which the baculovirus expression system is based is the utilization of the very strongly expressed polyhedrin promoter of Autographa californica nuclear polyhedrosis virus (AcNPV) to drive high level of expression of foreign proteins. AcNPV replicates in insect cells in a biphasic life cycle. In the first phase (12-24 h postinfection [pi]), virus replicates in the nucleus and buds from the cell membranes to form extracellular virus. Later in infection, high-level expression of the AcNPV late gene polyhedrin occurs, and arrays of mature virus particles are embedded in a protective coat of polyhedrin. In the latter stages of infection, polyhedrin can account for up to 50% of total cell protein, and the complexes it forms are readily visible as large, From: Methods In Molecular B!ology, Vol. 39: Baculovlrus Express/on Protocols Edited by. C. D. Fhchardson (D 1995 Humana Press Inc , Totowa, NJ
107
108
Page and Rodgers
crystalline occlusion bodies within the nucleus of infected cells. When foreign proteins are expressed in the baculovirus system, the gene of interest is placed downstream from the polyhedrin promoter. This results in the inserted gene being driven by the strong polyhedrin promoter and in the absence of polyhedrin expression owing to insertional inactivation. Recombinant virus can therefore be easily identified since it will produce occlusion-negative plaques. Generation of recombinant virus is a two-step process-the construction of recombinant transfer vector, and the production of recombinant baculovirus by cotransfection of insect cells with wild-type AcNPV DNA and recombinant plasmid. During virus replication in the transfected cells, recombinant virus arises by a process of homologous recombination (see Fig. 1). Culture supernatants from transfected cells will typically contain 106-lo7 viral plaque-forming units/ml (PFU/mL) of which 0. l-l % will comprise recombinant virus. By plating dilutions of the virus in monolayers of insect cells and culturing them under agarose, single virus particles will give rise to individual foci of infection (plaques). Recombinant viruses are then identified by visual screening for occlusion-negative plaques. Such recombinants are purified by further rounds of plaque assays and then expanded to produce a virus stock for the generation of recombinant protein. The two-step process for the generation of baculovirus recombinants is necessary because the large size of the AcNPV genome (-128 kDa) precludes direct DNA manipulation. All cloning steps are therefore performed in a baculovirus plasmid (transfer vectors that are now commonly available). They all incorporate three common features: 1. A plasmid origin of replication andantibiotic-resistancemarkersfor selection in E. coli.
2. A convenientrestriction siteor polylinker downstreamfrom the polyhedrin promoter to facilitate cloning procedures. 3. A large tract of native AcNPV sequenceflanking the cloning site to promote homologousrecombinationafter transfection. The actual choice of vector will depend on the requirement for fused or native recombinant protein and the cloning strategy to be used. Much work has gone into defining the sequence elements necessary for optimal expression of foreign genes (14,I5) and a variety of vectors are now available. For native, unfused proteins, a vector that
Recombinant
109
Selection by Visual Screening
TRANSFER VECTofl
-0
-
FOREIGN GENE
c
AECOMBINANT TRANSFER VECTOR
POLYHEORIN GENE
-
HoMocoGous RECOMBlNAllON IN v/v0
PROGENY RELEASED INFECTED
VIRUS FROM CELL 0 l-l% CONTAIN FOREIGN GENE I 0 PETRlDlSn
NEUTRAL
+ 27%3dm RED STAIN
CELL MONOLAYER IINFECTED WITH PRDGENY VIRUS)
Fig. 1. Generation of a recombinant baculovirus for expression of a foreign gene. (1) Construction of the recombinant transfer vector. (2) Extraction of wild-type baculovirus DNA. (3) Cotransfection of recombinant transfer vector and baculovirus DNA. (4) Plaques assay for selection of recombinant baculoviruses. Steps l-4 are detailed in Sections 3.1.-3.4.
110
Page and Rodgers
Fig. 2. Baculovirus transfer vectors for the expression of fused (pAc360) and nonfused (p36C) recombinant proteins. Vector p36C was derivedfrom pAc360 using site-directed mutagenesis to change the initiating methionine codon of polyhedrin from ATG to ATC. This results in the first 11 codons of the polyhedrin gene no longer being translated into protein, but retains this leader region, which may be important for other aspects of expression.
retains the complete untranslated leader sequence of polyhedrin is recommended. For instance, the vector p36C (16) described in Fig. 2 or similar vectors, such as pVTL941 (17) or YM-1 (14), would be appropriate. In other situations, benefit may be derived from ligating the foreign gene inframe with the N-terminal coding sequence of polyhedrin to produce a fusion protein, e.g., using the vector pAc360 shown in Fig. 2. This
vector is particularly useful when functional protein is not essential, but material is required for other uses, e.g., immunological studies. This vector also allows for rapid expression of recombinant proteins direct from h cDNA libraries without prior need for subcloning and sequencing. For
Recombinant
Selection by Visual Screening
111
instance, PCR primers can be designed with homology to the sequences immediately flanking the EcoRl cloning site of hgt 11 and can be used to amplify the inserted gene. Because the reading frame of the cDNA insert in hgtll is known (i.e., all antibody +ve hgtll plaques are inframe ZucZ fusions) restriction sites can be incorporated into the primers so as to place any amplified cDNA insert inframe with polyhedrin in the vector pAc360. Two such primers are shown below 5’ primer BamHI T AAG GAT CCC CCG TCA GTA TCG GCG GAA TIC 3’ primer
BamHI
TAT GGA TCC GTA GCG ACC GGC GCT CAG CTG
Immunopositive h plaques are simply picked as agar plugs and the cDNA insert amplified by 30 cycles of PCR, digested with BamHI and subcloned into pAc360 to produce recombinant protein without prior knowledge of the sequence of the cDNA. The remainder of this chapter is aimed at introducing the practical aspects of the baculovirus expression system to researchers who wish to use the technology, but who have no previous experience of handling insect cells or baculovirus. Although important aspects of each technique will be emphasized, a basic knowledge or recombinant DNA and tissue-culture techniques are assumed. 2. Materials 1. The insect cell line Spodopterafrugiperda Clone 9 (Sf9) may be obtained from the American Type Culture Collection (12301 Parklawn Drive, Rockville, MD 20852) or alternatively from any research group working in this area. 2. TC 100 medium may be obtained from a number of commercial suppliers, e.g., ICN Flow and GIBCO Life Technologies. It is supplemented with 10% fetal calf serum and gentamycin (50 pg/rnL). For scale-up work, serumfree insect cell media are now also available, e.g., SF900 and SF90011 (GIBCO Life Technologies) and EX-CELL 401 (J. R. H. Biosciences). 3. Stirrer platforms and flasks: Techne stirrer platforms and paddle flasks are recommended (Techne, Cambridge, UK). 4. Transfection buffer: Prepare a solution of 25 mM HEPES, pH 7.5,150 n&f
NaCl, 125 mM CaC12.Filter-sterilize (0.22 urn) and store at 4°C. The pH is critical and should be pH 7.5 f 0.05, since transfection efficiency is reduced dramatically at lower pH.
112
Page and Rodgers
5. Seaplaque ultrapure low-melting-point agarose (ICN Biomedicals). Prepare 3% (w/v) solution in distilled water and autoclave to sterilize. 6. Jensen neutral red solution (BDH). 7. Transfer vectors and wild-type virus may be obtained from research groups working in the field or alternatively from Invitrogen, San Diego, CA. 8. Restriction enzymes and other reagents for cloning procedures may be obtained from any commercial suppliers of molecular biology reagents. 9. 2X Proteinase K (PK) Buffer: 20 mM Tris, pH 7.6,2 rnM EDTA, 3% SDS, and 0.2M NaCl.
3. Methods To obtain successful expression of cloned genes using the baculovirus expression system, it is important that the principles underlying each experimental step are clearly understood. For this reason, each of the steps necessary for engineering recombinant baculoviruses are shown schematically in Fig. 1 and described below in detail.
3.1. CZoning of Foreign Genes into an AcNPV Transfer Vector As indicated above, becauseof the limitations of the size of the AcNl?V genome, all cloning steps are carried out using a baculovirus transfer vector (seeFig. 2). The choice of transfer vector will dependon the requirements for the recombinant protein and the cloning strategy employed, but should take into account the considerations outlined above. 1. Digest the vector of choice to completion with the appropriate restriction enzyme, e.g., BumHI for vector pAc360. 2. Phosphatase the digested vector using standard methodology. This step is necessary to reduce the high frequency of vector self-ligation that would otherwise occur. 3. Prepare the foreign gene such that it has BarnHI or compatible (e.g., BgZII) “sticky ends.” If such sites are not present, the ends could be altered by ligation of BumHI linkers or by “blunt ending” with subsequent cloning into blunt-ended vector. A third option now available with the advent of polymerase chain reaction (PCR) is to amplify the gene of interest with
primers tailored to include the appropriaterestnctronsites in the appropriate reading frame. Using this technique, any gene may be tailored precisely into the vector chosen. 4. Ligate the foreign gene into the phosphatased vector using T4 DNA ligase and transform into competent E. coli (e.g., DHSa) (see Note 1).
Recombinant
113
Selection by Visual Screening
5. Select recombinants on ampicillin plates, and prepare miniprep plasmid DNA in order to analyze the presence and orientation of insert by restriction digest (see Note 2). 6. Having identified a clone containing the insert in the correct orientation, the residual miniprep DNA may be used for transfection into insect cells (see Note 3). A culture of E. coli containing the recombinant plasmid should be prepared for long-term storage in glycerol as backup and for reference.
3.2. Preparation of High-MoGWt Infectious from Wild-Type AcNPV
DNA
Having prepared recombinant plasmid with the gene of interest inserted in the correct orientation, it is necessary to introduce the gene into baculovirus by cotransfection with DNA prepared from wild-type AcNPV. To be successful, this step requires the use of good-quality, infectious AcNPV DNA. This can be prepared: 1. Culture 100 mL St9 cells in a spinner flask to a density of 1.5-2.0 x lO%nL. 2. Infect at a multiplicity of infection (MOI) of 1 with AcNPV, and incubate at 28OC for 40-48h. 3. Pellet cells at 15OOgfor 5 min in a bench centrifuge. Discard the cell pellet (see Notes). 4. Pellet extracellular virus (EV) from the supernatant by centrifugation at 100,OOOgfor 45 min at 4°C in a swing-out rotor (e.g., 27,500 rpm in a Beckman SW40). 5. Resuspend the viral pellet in 10 mL phosphate buffered saline (PBS), pH 7.4, and pellet EV as before. 6. Resuspend viral pellet in 2.5 mL distilled water, add 2.5 mL, 2X proteinase K buffer, and then add 200 pg proteinase K. Incubate at 50°C for l-2 h. 7. Add an equal volume of 1: 1 phenol chloroform, and mix the phases gently for 2-5 min. It is important that mixing is done gently to prevent shearing of the DNA. 8. Spin at 15OOgfor 5 min in a bench centrifuge to separate the phases. Remove as much of the phenol phase as possible by inserting a Pasteur pipet through the upper DNA phase. 9. Repeat the above extraction once more with phenol-chloroform and then once with chloroform alone. 10. After chloroform extraction, carefully remove the upper DNA-containing phase to a fresh container using a wide-bore pipet to avoid shearing the DNA. Try to avoid carrying over protein from the interface (see Note 5).
114
Page and Rodgers
11, Add 0.5 mL 3M sodium acetate,pH 5.2, and 10 mL of cold ethanol (-2OOC). Mix by inverting and swirling gently. A cotton-wool-like precipitate of DNA should form at this stage. 12. Centrifuge at 1500g for 5 min to pellet the precipitate, and carefully discard all the supernatant. 13. Add 5 mL 75% ethanol to rinse the pellet, and then discard it. Invert the tube on a tissue paper to remove any remaining ethanol (see Note 4). Add 0.5 mL sterile distilled water and store at 4°C. Allow the high-mol-wt DNA to redissolve overnight, and measure the OD260,2s0of an aliquot. Approximately 200-500 pg of DNA should be present (see Notes 6-8). 3.3. Cotransfection of Wild-Type AcNPV DNA and Recombinant AcNPV Transfer Vector into Insect Cells Intact AcNPV DNA is infectious when transfected into insect cells. By cotransfecting plasmid, homologous recombination is induced during virus replication resulting in the formation of recombinant baculoviruses as 0. l-l % of the population. 1. Seed 2 x lo6 insect cells into a 25-cm2 tissue-culture flask, and leave at room temperature for 15-30 min to attach. 2. Into a sterile 1.5-r& Sarstedt tube, add 1 pg of wild-type AcNPV DNA and 2 p,g of recombinant transfer vector. This may be either CsCl purified DNA or miniprep DNA. Add 0.75 mL of transfection buffer, and mix by gently pipeting up and down. 3. Remove all of the medium from the flask of insect cells, and replace with exactly 0.75 mL fresh TClOO medium. Be sure to cover the entire monolayer with a film of medium. Slowly add the 0.75 mL of DNA solution directly onto the cell monolayer dropwise using a 1-mL pipet. The DNA will form a calcium phosphate DNA precipitate in situ because of the phosphates in the medium. 4. Leave at 27OCfor 4 h, then remove the transfection solution, and carefully wash the cell monolayer twice with 5 mL of culture medium. Finally, add 5 mL fresh medium, and incubate in a 27°C incubator for 3-5 d. 5. At the end of this period, a significant proportion (usually about 10%) of the cells should be showing positive signs of viral infection as evidenced by large granular occlusions of polyhedrin in the nuclei of infected cells. By this stage, the culture medium will contain virus particles at a concentration of about lo7 PFU/mL. Only 0.1-l% of these will be recombinant virus, and these have to be identified by visual inspection following a plaque assay.
Recombinant
Selection
by Visual
Screening
115
3.4. Plaque Assay Recombinant viruses present in the medium harvested following cotransfection can be identified and isolated by plaque assay. This method can also be used to determine the titer of an unknown virus stock solution. The principle of the method is to obtain well-isolated viral plaques in a confluent cell monolayer, which can then be screened for recombinants or simply counted for determination of virus titer. Recombinants are identified by virtue of their occlusion negative phenotype, the polyhedrin gene having been eliminated during homologous recombination in transfected cells. 1, Prepare a series of lo-fold dilutions of the virus stock in culture medium. An appropriate range of dilutions for culture supernatant harvested from cotransfections is 1: 100 to 1:10,000, and for determination of virus titer, 1: 10,000 to 1: 1,ooo,ooo. 2. Seed 35mm plastic Petri dishes with 9 x lo5 insect cells in 2 mL TClOO, and incubate at room temperature for 15-30 min to allow the cells to attach. Seed enough for 10-20 plates/dilution for screening recombinant plaques and two plates/dilution for determination of virus titer. 3. Remove most of the culture medium from the dishes, but leave sufficient medium to cover the monolayer and prevent the cells from drying out. Gently drip 100 l.tL of the diluted virus onto the cells in the center of the dish. Incubate for 1 h at room temperature to allow absorptron of the virus. 4. Prepare a 1% agarose/culture medium overlay solution as follows: Autoclave an appropriate volume of 3% Seaplaque agarose solution in distilled water and cool to 37°C. Warm twice this volume of culture medium to 37OC.Mix the agarose and the medium, and maintain at 37°C. Three percent agarose solution subsequently stored at room temperature can be quickly melted in a microwave for future uses. 5. Aspirate all the inoculum from the Petri dishes using a sterile Pasteur pipet. Add 1.OmL of the overlay to the center of each plate, allowing it to spread evenly over the monolayer. Allow it to set for 20 min at room temperature, and then overlay with 1 mL of culture medium. Incubate undisturbed in a humid environment at 28OC for 3-4 d. Movement of the plates during incubation can cause smearing of the plaques. 6. Stain the monolayer with neutral red. This will stain viable cells a deep pink color, leaving the dead and dying virus-infected cells clearly visible as paler circular areas (plaques) 3-4 mm in diameter. Add 1 mL of a neutral red stock solution, diluted 1 in 10 in sterile PBS, to the liquid overlay. Incubate at 28°C for l-2 h. Tip off the liquid overlay containing stain and
116
Page and Rodgers
Fig. 3. Under low-power (x400) examination on an inverted microscope, recombinant viral plaques (A, occlusion negative) are readily distinguished from wild-type AcNPV-derived plaques (B, occlusion positive). invert the’plates. If left for several hours or overnight at 4”C, the plaques will clear further. Each plaque represents one virus or PFU in the inoculum; therefore, the titer of the original stock can be determined by allowing for the dilutions made. 7. Screening for recombinants is done on the basis of visual identification of their occlusion-negative phenotype. This requires well-isolated plaques at a density of 20-200 plaques/plate. With practice, it is then possible to pick out occlusion negative plaques by eye, and confirm these by microscopic analysis. Wild-type plaques containing occlusion bodies have a characteristic refractile appearance, giving a slight silvery sheen when viewed at an angle against a light source. Plaques lacking this sheen can be marked for closer inspection under 400X magnification with an inverted microscope (see Fig. 3). Check every cell in the plaque carefully for occlusion bodies, and if none are seen, mark the plaque for further plaque purification (see Note 9). 8. Aim to pick four to six negative plaques for each recombinant virus if possible (see Note 10). 3.5. Plaque Purification Occlusion-negative plaques may still contain some wild type virus contamination because of diffusion from neighboring plaques. Thus, to
Recombinant
Selection
117
by Visual Screening
isolate a pure viral stock, recombinants are purified by successive rounds of plaque purification. 1. Using a sterile Pasteur pipet, pick a plug of agarose from directly over the potential recombinant occlusion-negative plaque into 0.5 mL of culture medium. Vortex briefly, and leave at room temperature for 30 min or longer to allow the virus to diffuse from the plus. Keep an aliquot at 4°C for long-term storage. 2. Carry out a plaque assay as in Section 3.4., using the plaque suspension undiluted and lo-fold dilutions of it down to 1 in 1000. 3. Stain and screen as described in the previous section. Unless the original plaque was picked from a very crowded plate, there should be at least lO20% occlusion-negative plaques by this stage, and the difference should be clearly visible. 4. Repeat until the plaques generated are 100% occlusion-negative. Generally, three rounds of plaque purification are carried out to ensure elimination of wild-type virus, but the presence of the foreign gene can be confirmed by Southern blotting or product analysis at an earlier stage if required. 3.6, Verification
of Recombinant
Protein
Production
Not all genuine occlusion-negative plaques will produce recombinant protein. Aberrant crossover events can occur during recombination resulting in recombinant virus carrying neither polyhedrin nor the foreign gene. Although it is rare for more than lO-20% of occlusion-negative viruses to be nonproducers, it is wise to check for recombinant product at an early stage, since this will greatly reduce the amount of plaque purification assays required. This can be done by Western blot or ELISA, or by assays for enzyme activity if this is appropriate. 1. Pick a plaque(s) from a second-round plaque purification assay into 0.5 mL TClOO, and store overnight at 4OC. 2. Seed a T-75 tissue-culture flask with 2 x lo6 insect cells, and infect with 0.1 mL plaque supernatant. 3. Incubate at 28°C for 4-6 d until all cells show signs of infection. 4. Tap flask to dislodge cells and pellet at 15OOgfor 5 min. 5. Gently resuspend the cells in 10 mL ice-cold PBS and pellet. Repeat this step once more to remove remaining trace of FCS. 6. Resuspend the cells in 500 pL ice cold PBS containing 1 mM leupeptin and 1 mM as protease inhibitors. 7. Remove 50 pL, add 50 pL 2X SDS-PAGE buffer, and freeze immediately at -2OOC.
118
Page and Rodgers
8. To the remaining 450 pL, add 50 p.L 10% (v/v) NP40, vortex briefly, and place on ice for 5 min. (see Note 11). 9. Vortex briefly again and spin at 15OOgfor 5 min in a bench centrifuge. This will pellet the nuclei. 10. Remove the cytoplasmic supernatant fraction, add an equal volume of 2X SDS-PAGE buffer, and freeze at -2OOC. 11. Wash the nuclear pellet by resuspending in 1 mL ice-cold PBS containing 1 mM leupeptin and pepstatin. 12. Resuspend pellet in 450 pL PBS plus protease inhibitors, add an equal volume of 2X SDS-PAGE buffer, and freeze at -20°C (see Note 12). 13. Analyze these samples by SDS-PAGE and Western blot with reagents appropriate for the recombinant expected. This will demonstrate whether recombinant protein is being produced and whether the product is of the expected size, Coomassie blue staining of the gel may give some identification of protein yield, but this will be underrepresented since the virus will not be pure and conditions of infection will be suboptimal. 14. To analyze the samples by ELISA, freeze the cytoplasmic fraction directly at step 9 without the addition of SDS-PAGE buffer. Likewise at step 13, resuspend the nuclei in 50 pL PBS and then add 450 l,tL of 8M urea in 25 mM HEPES, pH 7.5, sonicate for 30 s and freeze at -20°C. These samples may now be analyzed by direct coating on microtiter plates followed by ELISA.
3.7. Isolation of Recombinant Viral DNA for Characterization of Southern Blotting Although not a routine requirement for establishing recombinant baculoviruses, analysis of the recombinant virus genome by Southern blotting may be useful in some instances and will provide a useful reference if problems are encountered during use. In one such case, a lo-fold drop in the yield of recombinant HIV gp160 was traced to overgrowth of the virus stock by a deletion mutant that could be clearly identified by Southern blotting. Such occasions are rare, but if required, recombinant virus DNA can be prepared. 1, Seed lo7 Sf9 cells into a T75 flask, and infect at an MO1 of 1 with recombinant baculovirus. 2. At 2-3 d pi when there is good cytopathic effect (cpe), remove the cells by tapping the flask, pellet, and wash two times in PBS. 3. Resuspend the pellet in 250 p.L dHzO and add 250 p,L of 2X PK buffer. 4. Add 5 pL of proteinase K (250 mg/mL stock; 200 pg/mL final), mix by inverting, and incubate at 55°C for 1 h.
Recombinant 5. 6. 7. 8.
9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19.
Selection
by Visual Screening
119
Add 500 uL phenol:chloroform, and mix by inverting rapidly several times. Spin for 1 min in a microfuge to separate phases. Remove lower phenol phase, being careful to avoid taking any DNA. Repeat with 1:l phenol:chloroform and once with chloroform extraction. After the final chloroform extraction, remove the upper aqueous phase (DNA) to a fresh tube using a wide-bore Pasteur or cutoff pipet (see Note 13). Add RNase to 20 ug/mL and incubate at 37°C for 15 min (1 p.L of 10 mg/ mL stock). Extract twice with phenol:chloroform and once with chloroform as before. Transfer DNA to a 7 mL bijou with wide-mouthed pipet. Add l/lOth vol3M NaOAc and 2.5 vol absolute EtOH at -2OOC. Mix gently inverting several times. DNA should come out of solution as a stringy mass.Swirl gently to condense DNA sufficiently to pick it out with a cutoff blue tip. Remove precipitated DNA into a fresh Eppendorf tube using a cutoff widebore Pasteur or pipet. Add 1 mL 70% EtOH. Spin for 30 s to pellet DNA. Decant supernatant (or pipet off), and remove last traces of EtOH with the comer of a tissue. Add 50-100 pL dH,O to redissolve. Leave overnight at 4OC. Do not dry first or DNA will not redissolve. Store DNA in separate aliquots at +4OCand -2OOC.Yield should be 20-50 pg/T75 flask.
3.8. Accessory
Screening
Techniques-Hybridization
In cases where identification of recombinant plaques by visual screening proves difficult, hybridization techniques may be used to identify recombinant viruses. Although it is possible to screen direct plaque lifts of baculovirus cultures, this is a relatively troublesome technique and it is easier and more reliable to combine limiting dilution and dotblot techniques. 1. Titrate the supernatant from your initial transfection to obtain a titer expressed as PlWmL. (Usually this will be in the order of lo7 PFU/mL.) 2. Dilute Sf9 cells to lo5 cells/n& and transfer 100 r&/well to flat-bottomed 96-well tissue culture plates using a multichannel pipet and sterile tips, Leave at room temperature for 15-30 min for cells to adhere. 3. Dilute transfection supematant in medium to give titers of approx 200 PFU/ mL (10 PFU/mL) and 60 PFU/mL (-3 PFWSO pL).
Page and Rodgers 4. Add 50 pL diluted virus/well to the appropriate number of plates. Usually one plate seeded at 10 PFU/well and two or three at 3 PFU/well will be more than adequate to identify recombinant vimses. Incubate at 27OC(see Note 14). 5. When good cpe is apparent (usually 3-5 d), transfer the virus-containing supernatants to a fresh 96-well plate using a multichannel and sterile tips. (There is no need to change tips between wells at this stage.) Store the supernatant at 4OC. 6. Process the cells remaining in the original plates by adding 50 pL 0.5M NaOH to each well, and incubate at room temperature for 30 min. This will disrupt the cells and denature the DNA. 7. Neutralize by the addition of 50 I&,/well 2M ammonium acetate. 8. Place an appropriate nylon or nitrocellulose membrane in a standard dotblot apparatus, and apply suction. 9. Transfer the denatured contents of the assayplate to the dot-blot manifold, pipeting up and down once to ensure complete removal of the cells/DNA. 10. Ensure that the contents are completely sucked through the manifold, remove filter, dry, and process as for standard hybridization procedures. 11. Probe filters with labeled insert DNA using standard hybridization procedures, and identify well giving a positive signal, i.e., those containing recombinant viruses (see Fig. 4). 12. Locate the appropriate positive wells in the master plate stored at 4”C, and plaque the supematants from these positive wells. Occlusion-negative plaques should not be readily visible and can than be processed as for visual screening. 13. In some instances, recombinant plaques may be outgrown by wild virus, and occlusion-negative plaques may still only be present at relatively low frequency. In this case,a secondround of dot hybridization may be necessary. 14. For a second-round hybridization screen, virus from the initial screen is titrated by limiting dilution, ideally to 1 PFU/well, and rescreened as before. Add 100 pL of medium to each well of a 96-well-plate. Add 50 yL of a 1:10 dilution of positive first-round supematant to duplicate wells of row A. Perform serial threefold dilutions by transferring 50 pL down the plates to row H. 15. Dilute insect cells to l@/mL, and add 100 yL cells to each well. Incubate at 27°C for 3 d as before. 16. Store supematants in parallel plate, and process cells for hybridization as before. 17. Identify hybridization-positive wells, and plaque titrate supematant from the lowest positive dilution for visual screening. Occlusion-negative plaques should now comprise >20% of the total.
Recombinant
Selection by Visual Screening Baculov~rus
10 XIee”
P screen
-
dot blot hybridisatmn
121
rcreen
- (10 pfu/well)
- (titration
plaque array pmk inclusion
of lo +ver)
-w?s
Fig. 4. Typical results of first- and second-round dot-blot hybridization screens to identify recombinant virus. 3.9. Amplification of Virus Stocks When recombinant virus has heen plaque purified and shown to produce recombinant product, it is necessary to amplify the virus stock to produce sufficient virus for future use. How this is done will depend on the intended use.
3.9.1. for Small-Scale Studies 1. Seed a T75 flask with 2 x lo5 insect cells in 20 mL TClOO or serum-free medium, e.g., SF900, and infect with 200 j.t,Lof supematant from a pure plaque plug.
122
Page and Rodgers
2. Incubate at 28OCfor 4-6 d until all cells show signs of infection, 3. Pipet supernatants into a sterile centrifuge tube, and pellet any cells at 15OOgfor 5 min. Store the virus containing supernatant at 4°C. 4. Seed one or more T150 flasks with 3 x lo7 cells m 40 mL of medium and infect with 2 rnL of the supernatant from 3. (This will give an MO1 of approx 1.) 5. Incubate at 28OCfor 3 d, and then harvest the virus-containmg supernatant as before. When titrated by plaque assay, such supematants should have titer of 4 x lo7 PFU/mL or greater.
3.9.2. For Larger-Scale Studies Virus stocks are expanded more systematically and at low multiplicity to minimize generation of defective viruses (see Section 3.8. for example). Routinely virus stocks are amplified in serum-free medium (e.g., SF900, GIBO Life Technologies or Ex-Cell401, J.R.H. Biosciences), since cells grow to a higher density in these media, and because serum-free medium is the medium of choice for large-scale production of protein. 1. Grow a small scale 30 mL culture of virus as in Section 3.9.1. 2. Keep this as the “Master Virus Stock.” Freeze 5 x 1 mL aliquots at -70°C for long term storage and the rest at 4°C for further amplification. 3. In a 1 L Techne paddle flask with a stir speed of 80 rpm, grow 500 mL of serum-free cells to a density of l-5-2.0 x lo6 mL. 4. Infect the cells at an MO1 of 0.1 usmg the Master Virus Stock from 2. This will normally require approx 5 mL of vnus stock. 5. Incubate at 28°C for 3 d, and then harvest by pelleting the cells at 1500g for 5 min. Store the supernatant in sterile 20 mL aliquots; some at 4OC for immediate use, and some at -7OOCfor long-term storage. This is designated the “Virus Seed Stock” and is used to generate a “Working Virus Stock.” 6. Grow a large volume of serum-free adapted cells in one or more large Techne paddle flasks with a stir speed of 80 rpm. For example: Flask size Working volume 3L l-l.5 L 1OL 2-3 L Use of large volumes reduces the surface area:volume ratio and results in oxygen limitation of cell growth. 7. When the cells reach a density of 2 f 0.2 x lo%& infect at an MO1 of 0.1 using the seed stock generated in step 5. 8. Incubate for 3 d at 28OC,and harvest the supematant by pelleting the cells at 15OOgfor 5 min. Store the supematant in sterile 100 mL aliquots at 4°C
Recombinant
Selection by Visual Screening
123
(for up to 1 yr) or at -7OOC for longer term storage. This is the Working Virus Stock and should be used for all recombinant protein production to maintain consistency. When the Working Stock is exhausted, a further Working Stock can be generated from the Seed Stock as above. 3.10. SmalGScale Production of Recombinant Protein Having obtained and amplified pure virus stocks, studies with the recombinant protein may begin. A range of options are available for expression of recombinant protein, and the method chosen will depend on the amount of protein required. The choices range from small-scale expression in tissue culture flasks, through medium scale culture in roller bottle or Techne spinner flasks to large scale production in fermenters. Brief notes on expression in tissueculture flasks and Techne stirrers are included below.
3.10.1. Small-Scale Expression in Tissue-Culture Flasks 1. Seed 2 x lo7 cells m one or more T75 flasks or 3 x lo7 cells in T150 flasks. 2. Allow to adhere for 15-30 min at room temperature, remove most of the medium leaving only enough to cover the monolayer, and infect with recombinant virus at an MO1 of l-5 PFU/cell. 3. Allow the virus to absorb to the cells for 1 h at 28”C, remove the remaining medium, and refeed the cells with 20 mL (T75) or 40 mL (T%50) of fresh medium. 4. Incubate cells at 28°C for 40-48 h, and then harvest and process as required (see Note 15).
3.10.2. Medium-Scale
Expression in Techne Stirrers
1. Culture an appropriate volume of cells to 2-3 x lo6 cells/ml in serum-free medium in Techne stirrer vessels using a stir speed of 80 rpm. Suggested working volumes: Flasks size Max working vol Total no of cells 250 mL 500 mL 1L 3L 1OL
100 mL 200 mL 500 mL 1.5 L 2-3 L
3 x 10s 6 x lo* 1.5 x log 4.5 x 109 9 x109
2. Infect at a MO1 of 1using the Working Virus Stock generated in Section 3.9. 3. Incubate at 28°C for 40-48 h using a stir speed of 80 rpm. 4. Harvest cells and process as appropriate (see Notes 16-19).
124
Page and Rodgers 4. Notes
1, A recA-negative strain of E. coli (e.g., DH5a) should be used for cloning purposes since other strains, e.g., TGl can on occasions cause problems with plasmid rearrangement. 2. Both the transformation efficiency and the overall cloning efficiency are frequently observed to be low with baculovirus transfer vectors, presumably because of their large size (-9.6 kb). However, even when colony numbers suggest ineffective cloning, recombinant plasmids will normally be found, Preliminary screening by means of colony lifts and hybridization with labeled insert may be useful in such instances. 3. We have also found that Lipofectin (GIBCO/BRL) can be used to efficiently introduce DNA into insect cells (see Chapter 8). 4. It is important not to dry the AcNPV DNA completely after washing in 75% ethanol or it will be very difficult to redissolve. 5. AcNPV DNA is very large-take care to minimize shearing during DNA processing in order to obtain a good, intact, infectious preparation. 6. If difficulty is encountered preparing good-quality DNA from extracellular virus, total DNA extracted from infected cells may be used (see Section 3.8.). This will comprise approx 25% viral DNA, but EV DNA is to be preferred. 7. Store at 4OC. Do not store at -2OOC. Multiple freeze-thaw cycles will reduce viral titers. 8. It is wise to check the quality of AcNPV DNA by transfecting 1 ltg into Sf9 cells using the method. 9. Visual screening of inclusion-negative plaques is easy with practice. A useful exercise to begin with is to mix wild-type virus and any established recombinant virus to give ratios of 1: 1, 3: 1, and 10: 1 wild-type:recombinant. Plaque out the mixtures at approx 100 plaques/well, and view the plaques from the underside while held up against the light. The difference between occlusion-positive and occlusion-negative plaques should be readily apparent at these ratios and will provide a useful demonstration making the screening for low-frequency inclusion negatives after transfection much easier. 10. The secret of visual screening is the production of good-quality, welldefined plaques. Some of the problems that may be encountered in performing plaque assaysare: a. Very small plaques-usually because of seeding the cells too densely, causing a complete monolayer to form before virus infection can proceed. The same effect can also occur if stock cells used to seed the plates are overgrown (>2.4 x 106/mL).
Recombinant
Selection by Visual Screening
125
b. No plaques/poor monolayer in the center of the well-most likely because of cells drying out in steps 3 or 5. Leave sufficient medium to cover the monolayer in step 3 and proceed quickly after removal of medium in step 5. Do not leave plates in a laminar flow hood during virus absorption since this will promote drying. c. Cells appear “grainy” and “ghost-like” under the microscope-probably because of agar being applied too hot and killing the cells. d. Lumpy agar-incomplete mixing of agar following autoclaving or microwave, agar left to cool for too long before addition, or not allowed to set for long enough at step 5 before finally adding TClOO. 11. The fractionation of cells into cytoplasm and nuclei serves two purposes. First, it is useful to know the subcellular localization of your protein, since many recombinant proteins are partitioned predominantly to one or the other compartment in insect cells, and fractionation into cytoplasm and nuclei can provide a quick and easy first step in a purification protocol. Second, some recombinant proteins are susceptible to proteases present in the cytoplasm of insect cells. Nuclear localized proteins are protected from these proteases, and a more representative picture is sometimes obtained by analyzing nuclei and cytoplasm individually rather than total cells. 12. Some cytoplasmic proteases mentioned above are active even in the presence of SDS and mercaptoethanol present in gel-loading buffer, but are inhibited by leupeptin. 13. This is important-forcing high-mol-wt DNA through a fine tip will shear it, as will vortexing. Attention to these points will result in good yields of high-quality DNA. 14. Attempting to screen much larger numbers of viruses by seeding at much greater PFU/well results in higher backgrounds and impaired positive signals on hybridization. 15. Individual recombinants may vary in their optimal culture requirements. It is wise to optimize MO1 (l-10) and time of harvest (40-96 h) for each individual recombinant. 16. Insect cells have a high oxygen demand, particularly after infection with baculovirus. To a large extent, this can be compensated for by using high stir speeds (80 rpm) to increase oxygen exchange. Use of larger working volumes than those recommended will tend to nullify this effect by reducing the surface area-to-volume ratio. 17. Cells must be in log phase growth at infection. Use of cells that have grown to maximum density and are in the stationery phase will result in a dramatic reduction in yield of recombinant protein.
Page and Rodgers 18. As with small-scale culture, the multiplicity of infection and time harvest should be optimized for each individual recombinant. Using the above conditions, it is however possible to produce >lOO mg of recombinant protein from a 3-L culture. 19. It is strongly recommended that wherever possible separate laminar flow cabinets be reserved for uninfected cell work and for virus infected work in order to avoid contamination of cells with virus. When working with large volumes of high-titer virus, remember each microliter of supernatant will contain approx 105PPU of virus.
References 1. Kuroda, K., Geyes, H., Gayer, R , Doerfler, W., and Klenk, H. D (1990) The oligosaccharides of influenza virus haemagglutinin expressed in insect cells by a baculovirus vector. Virology 174,418-429. 2. Patel, G. and Jones, N. C. (1990) Activation in vitro of RNA polymerases II and III by baculovirus produced ElA protein. Nucleic Acids Res. 18,2909-2915. 3. Luo, L., Li, Y., and Kang, C. Y. (1990) Expression of gag precursor protein and secretion of virus-like gag particles of HIV-2 from recombinant baculovirus infected insect cells. Virology 179,874-890. 4. Page, M. J., Hall, A., Rhodes, S., Skinner, R. H., Murphy, V., Sydenham, M., and Lowe, P. N. (1989) Expression and characterization of the Ha-ras p21 protein produced at high levels in the insect-baculovirus system. J. Biol. Chem. 264,19,14719,154. 5. Lowe, P. N., Sydenham, M., and Page, M. J. (1990) The Ha-ras protein, ~21, is modified by a derivative of mevalonate and methyl-esterified when expressed in the insect baculovirus system. Oncogene 5, 1045-1048. 6. Stow, N. (1992) Herpes simplex type 1 origin-dependent DNA replication in insect cells using recombinant baculovirus. J. Gen. Viral. 73,3 13-321, 7. Xi, S. Z. and Banks, L. M. (1991) Baculovirus expression of human papillomavirus type 16 capsid proteins: detection of Ll-L2 protein complexes. J. Gen. Virol. 72, 2981-2988. 8. Ertl, P. F. and Powell, K. L. (1992) Physical and functional interaction of human cytomegalovirus DNA polymerase and its accessory protein (ICP36) expressed in insect cells. J. Virol. (in press). 9. Brown, C. S., Van Lent, J. W. M., Vlak, J. M., and Spa, W. J. M. (1991) Assembly of empty capsids by using baculovirus recombinants expressing human parvovirus B 19 structural proteins. J. Virol. 65,2702-2706. 10. Montross, L., Watkins, S., Moreland, R. B., Mamon, H., Caspar, D. L. D., and Garce, R. L (1991) Nuclear assembly of polyomavirus capsids in insect cells expressing the major capsid protein VPl. J. Virol. 65,49914998. 11. French, T. J., Marshall, J. J. A., and Roy, P. (1990) Assembly of double-shelled, virus-like partrcles of bluetongue vuus by simultaneous expression of four structural proteins. J. Virol. 64,5695-5700.
Recombinant
Selection
by Visual Screening
127
12. Loudon, P. T. and Roy, P. (1991) Assembly of five bluetongue virus proteins expressed by recombinant baculoviruses: inclusion of the largest protein VP1 in the core and virus-like particles. Virology 180,798-801. 13. Le Blois, H., Fayard, B., Vrakawa, T., and Roy, P. (1991) Synthesis and characterization of chimeric particles between epizootic hemorrhagic disease virus and bluetongue virus: functional domains are conserved on the VP3 protein. J. Virol. 65, 4821-4831. 14. Matsuura, Y., Possee, R. D., Overton, H. A., and Bishop, D. H. L. (1987) Baculovirus expression vectors: The requirements for high level expression of proteins, including glycoproteins. J. Gen. Virol. 69, 1233-1250. 15. Luckow, V. A. and Summers, M. D. (1988) Signals important for high-level expression of foreign genes in Autographa californica Nuclear Polyhedrosis Virus expression vectors. Virology 167,56-7 1. 16. Page, M. J. (1989) p36C: an improved baculovirus expression vector for producing high levels of mature proteins. Nucleic Acids Rex 17,454. 17. Luckow, V. A. and Summers, M. D. (1989) High level expression of non-fused foreign genes with Autographa californicu nuclear polyhedrosis virus expression vectors. Virology 170,31-39.
CHAPTER7 Production of Recombinant Baculoviruses Using Linearized Viral DNA Paul A. Kitts 1. Introduction The classical method for making a recombinant baculovirus relies on homologous recombination in insect cells to replace a segment of the bacwlovirus genome with the corresponding segment from a transfer vector that has been modified to include a foreign gene. The frequency of this recombination is low; most of the input viral DNA molecules do not recombine with a plasmid DNA. Typically, only O. l-2% of the progeny viruses are recombinant; hence, large numbers of plaques must be screened in order to find a few recombinant viruses. If linearized viral DNA is used in the cotransfection step, instead of the native viral DNA, which is circular, the proportion of progeny viruses that are recombinant increases to about 25% (I). This can greatly reduce the work involved in screening for recombinant viruses. The use of linearized viral DNA is based on the theory that only circular viral DNA can function as a substrate for replication; it seemsunlikely that a virus whose native DNA is circular would encode the specialized machinery needed to replicate linear DNA. Linearized viral DNA would thus be unable to replicate (Fig. 1, upper panel). One way of restoring circularity to the linear DNA is by recombination with a transfer vector carrying sequenceshomologous to those flanking the break. Recombination on both sides of the break would generate a circular viral DNA, From Methods in Molecular Biology, Vol. 39: Baculovirus Expression Protocols Edited by: C. D. Richardson 0 1995 Humana Press Inc., Totowa, NJ
129
Kitts
130 Linear
Viral
Progeny
DNA
Viral
DNAs
8 cl +
Linear
Vlrsl
DNA
kn
Transfer
Vector
Recombinant
Vlrsl
DNA
Progeny
Viral
DNAs
Fig. 1. Rescueof linear viral DNA by recombination with a transfer vector. (Reproducedfrom Kitts, P. A., Ayres, M. D., and Possee,R. D., Nucleic Acids Res. l&5667-5672 [ 19901by permission of Oxford University Press.) which would be competent for replication (Fig. 1, lower panel). In this process, any foreign DNA that had been inserted between the two segments of viral sequences in the transfer vector would be transferred to the viral genome (Fig. 1, lower panel). Thus, rescue of the linearized viral DNA by a recombinant transfer vector would produce the desired recombinant baculovirus. The results presented by Kitts et al. (I) provide good support for this model. They found that linearized Autographa cdifornica nuclear polyhedrosis virus (AcMNPV) DNA had a 15 to 150-fold lower infectivity than circular DNA. Consequently, in cotransfections using linearized viral DNA, the number of progeny viruses that are derived from unrecombined viral DNA molecules was greatly reduced. The yield of recombinant viruses was reduced to a much smaller extent, leading to a lo-fold higher fraction of recombinants among the progeny viruses. The model predicts that viral DNA linearized at any location could be rescuedby a transfer vector that contains viral sequences homologous to both sides of the break; this has been confirmed for the two locations commonly used to make expression vectors, the polyhedrin and p10 loci (I). AcMNPV DNAs that can be linearized at a specific point have been made by introducing a recognition site for Bsu361 endonuclease, an
Recombinant
Viruses from Linear
Viral DNA
131
AcRP23.IacZ
AcRPE-SC
AcUWl-IacZ
\_,i
Fig. 2. AcMNPV derivatives with a unique restriction site. The site at which Bsu361 endonuclease cuts the viral DNAs (arrows) is shown relative to the polyhedrin and p10 genes (open boxes). The E. coli ZucZ gene is shown as a shaded box, and the linker sequences that replace the N-terminus of the polyhedrin gene in AcRP6-SC as a hatched box. Filled triangles represent the polyhedrin and p10 promoters. enzyme that does not normally cleave the viral DNA (I), at a unique location in the AcMNPV genome. This has been done by incorporating
either an oligonucleotide containing the Bsu361 recognition site or the Escherichia coli ZacZ gene, which naturally contains a Bsu361 site, into the viral genome. Three AcMNPV derivatives with a unique Bsu361 site are shown in Fig. 2: AcRP6-SC (I), AcRP234zcZ (2), and AcUWl-ZucZ (3). Viruses that have a unique SseI restriction site in the polyhedrin locus upstream of the transcriptional start site and coding region of polyhedrin protein have also been constructed (see Chapter 8; Invitrogen, San Diego, CA) The critical factor in choosing which linearized viral DNA to use is that the transfer vector must contain viral sequences from both sides of
132
Kitts Table 1 Appropriate Linearized Viral DNA and Transfer Vector Combinations Transfer vector
Based on the polyhedrin locus, e.g., pAcYM1, pEV55 ,pVL941, pAcCL29, pAcMP1, pAcUW31a Based on the polyhedrin locus and containing a lacZ reporter cassette, e.g., pJVNhe1, pAcDZ1, pETL, pPIOa Based on the p10 locus, e.g., pAcUW1, ~AcAS~~
Linearized viral DNA AcRP23-lacZ AcRP6-SC AcUWl-lacZ
aFor further examples of the different classesof transfervector and references,see Chapter 2.
the cut in the viral DNA so that the linearized viral DNA can be rescued by homologous recombination (Fig. 1). The decision is therefore dictated by the transfer vector being employed (Table 1). If the transfer vector is based on a fragment of AcMNPV from the polyhedrin locus, then one of the viral DNAs that can be linearized at the polyhedrin locus, AcRP234acZ or AcRP6-SC (Fig. 2), should be used. Of these two viruses, AcRP234acZ is preferred because a higher proportion of recombinants can be obtained by using a screen for P-galactosidase to eliminate plaques of the parental virus (see below). When the transfer vector contains ZacZ as a reporter gene, however, AcRP6-SC DNA or linearized viral DNA from Invitrogen must be used. Alternatively, AcUWl-ZucZ DNA, which can be linearized at the p10 locus (Fig. 2), should be used if the transfer vector is based on a fragment of AcMNPV from the p10 locus. Note that the same linearized viral DNA can be used with a variety of transfer vectors containing different sequences embedded within a particular segment of the viral genome; this includes dual-expression vectors and vectors with different promoters. In addition to the increased proportion of recombinants resulting from the use of linearized viral DNA, the use of linearized AcRP23-ZacZ or AcUWl-ZucZ DNA has the advantage that recombinant plaques can be distinguished visually from plaques of the parental virus; plaques of the parental virus will produce P-galactosidase and stain blue with X-gal, whereas recombinant viruses, in which the foreign gene from the transfer vector has replaced the ZacZ gene, will not produce J3-galactosidase and will make white plaques. Using the blue/white screen further increases the probability of picking a recombinant plaque, although some
Recombinant
Viruses from Linear
Frequency of Recombination Cotransfection
Viral DNA
Table 2 with Linearized AcRP23-LacZ DNA“
Fraction of total plaques that are recombinant
1
27% 30% 45% 33%
2 3 4
133
Fraction of white plaques that are recombinant 37% 35% 48% 37%
a One hundred nanograms of linearized AcRP23-1acZ were cotransfected with 500 ng of pAcEI-I, a transfer vector containing the AcMNPV polyhedrin gene (5). The progeny viruses were plaque assayed, and the plaques stained with neutral red and X-gal. Plaques were screened for the presence or absence of polyhedra and for blue/white color: plaques with polyhedra are recombinant, plaques of the parental virus are with no polyhedra, and white plaques lackmg polyhedra are nonrecombinant nonparental.
white plaques
are apparently
generated
by recircularization
of the linear
DNA by processes that destroy the integrity of the ZucZ gene (1). One can expect between 30 and 50% of the white plaques to be recombinant (e.g., Table 2). The standard cotransfection methods can easily be modified to incorporate linearized viral DNA. The only novel step is the restriction of the viral DNA with an endonuclease that cuts at a unique location. The resulting linearized viral DNA is mixed with DNA of the modified transfer vector and used to transfect Spodoptera frugiperdu cells. Inside the cells, the linear viral DNA can be rescued by recombination with the homologous regions of the transfer vector, thereby replacing the viral sequencesflanking the cut with the modified sequencesfrom the transfer vector. Viruses produced from the cotransfection are plated out to give individual plaques, and a few putative recombinant plaques are picked for further analysis. The viruses in these plaques are purified, screened for either the expression or presenceof the foreign gene, and then a larger stock is made from a confirmed recombinant virus. The latter steps are the same as those in the procedures that use native viral DNA. Thus, the use of linearized viral DNA adds only one simple step to the standard procedure, but makes screening for recombinant viruses much less tedious, In a further development of this technique, the AcMNPV derivative BacPAK6 has been constructed such that restriction with Bsu361 not only linearizes the viral DNA but also removes part of an essential gene (4).
134
Kitts
This forces recombination between the restricted viral DNA and transfer vectors that carry a copy of the essential sequences. Consequently, 90100% of the viruses obtained from cotransfections using restricted BacPAK6 DNA are recombinant (4). For transfer vectors based on the polyhedrin locus, most of which carry the sequencesnecessary to rescue restricted BacPAK6 DNA, this is the most convenient method of generating recombinant viruses because restricted BacPAK6 viral DNA is commercially available from Clontech (Palo Alto, CA) or PharMingen (San Diego, CA). 2. Materials 2.1. Linearization
of Viral
DNA
1. AcRP23-ZucZ,AcRP6-SC, or AcUWl-ZacZ viral DNA in TE (10 miI4 Tris-HCl, pH 8.0, 1 rnkf EDTA). 2. Sterile water. 3. 10X NEBuffer 3: 1M NaCl, 500 rniV Tris-HCl, 100 mM MgCl*, 10 mM dithiothreitol, pH 7.9, at 25OC (New England Biolabs, Beverly, MA). 4. 100X Acetylated bovine serum albumin (BSA) (10 mg/mL) (New England Biolabs). 5. Bsu361 endonuclease (New England Biolabs) or an isoschizomer, e.g., SauI (Boehringer Mannheim, Indianapolis, IN). 2.2. Cotransfection of Viral and Transfer Vector DNAS 1, 35mm Tissue-culture dishes (Nunc, Naperville, IL). 2. TClOO medium (GIBCO/BRL, Gaithersburg, MD) supplemented with 5% fetal calf serum (FCS), 50 U/n& pencillin, and 50 pg/mL streptomycin. Store at 4°C. 3. Serum-free TClOO medium with antibiotics. Store at 4OC. 4. TClOO medium with 10% FCS and antibiotics. Store at 4°C. 5. Spodoptera frugiperda Sf21 or Sf9 cells. 6. Polystyrene vials or tubes (Falcon, Lincoln Park, NJ).
7. Lipofectin (1 mg/mL) (GIBCO/BRL). Store at 4°C. 8. Sterile water.
9. Transfervector DNA in TE. 10. Linearized viral DNA from Section 3.1. 2.3. Isolation of Recombinant Viruses 1. 35-mm Tissue-culture dishes (Nunc). 2. TClOO medium with 5% FCS and antibiotics. Store at 4°C. 3. Spodoptera frugiperda Sf21 or St?9cells.
Recombinant
Viruses from Linear
Viral DNA
135
4. 3% (w/v) SeaPlaque low-gelling temperatureagarose(PMC Bioproducts,
Rockland, ME) in water, autoclaved. 5. PBS: 140mmNaC1,27mMKC1,8mMNa&PO~,1.5mMKH~P04,pH7.3.
6. 0.33% Neutral red (GIBCO/BRL). 7. 2.5% X-gal (Boehringer Mannheim) in dimethylformamide, hght-sensitive. Store at -20°C. 2.4. Screening for Recombinant Viruses 2.4.1. Analysis of Virus-Infected Cell Proteins
1, 35-mm Tissue-culturedishes(Nunc). 2. 3. 4. 5.
TClOO medium with 5% FCS and antibiotics, Store at 4OC.
Spodopterafrugiperdu Sf21 or Sf9 cells. PBS. 5X dissociation mix: 10 mk! Tris-HCl, pH 6.9, 10% sodium dodecylsulfate, 25% P-mercaptoethanol, 25% glycerol, 0.02% bromophenol blue.
2.4.2. Analysis of Virus-Infected Cell DNA 1. 35-nun Tissue-culture dishes (Nunc). 2. TClOO medium with 5% FCS and antibiotics. Store at 4°C. 3. Spodopterafrugiperda Sf21 or Sf9 cells.
4. TE: 10 rniWTris-HCl, pH 8.0, 1 mM EDTA. 5. Lysis buffer: 50 mM Tris-HCl,
pH 8.0, 10 m&f EDTA, 5% p-mer-
captoethanol,0.4% sodiumdodecylsulfate. 6. 10 mg/mL proteinase K, made fresh.
7. 10 mg/mL RNase A. Storedat -2OOC. 8. Phenol:chloroform (50:50), equilibrated with 100 mM Tris-HCl, pH 8.0.
9. 3M sodium acetate,pH 5.2. 10. 300 and 70% ethanol. 3. Methods 3.1. Linearization of Viral DNA Viral DNA from an appropriate virus (Table 1) is linearized by digestion with Bsu361 endonuclease or an isoschizomer. The DNA is treated with an excess of enzyme in order to drive the digestion to completion and the extent of digestion is monitored on a minigel (see Note 1). Alternatively, linearized viral DNA can be obtained from a commercial supplier (see Note 16). 1. Mix: 1 ug of viral DNA (see Note 2), 5 uL 10X NEB buffer 3, 0.5 pL 100X acetylated BSA, and Hz0 to total of 50 p,L. Add 5 U of Bsu361, and
mix by gentle tapping of the tube (seeNote 3).
136
Kitts
2. Incubate at 37°C for 2 h. 3. Heat to 70°C for 15 min to inactivate the enzyme. 4. Store the digest at 4°C. The linearized DNA can be stored for several months at 4°C before being used in cotransfections. Do not freeze the viral DNA, since this destroys its infectivity. 5. To check that the digest is complete, run 20 ng of the digested DNA and 20 ng of undigested DNA on a 0.5% agarose minigel in 1X TBE + 0.5 pg/ mL ethidium bromide at 12 V/cm for 1h. Under these conditions, circular viral DNA remains in the well, whereas linear DNA migrates as a discrete band (Fig. 3). If undigested viral DNA can be detected in the digest, more enzyme should be added and the digest repeated. 3.2. Cotransfection
of Viral
and
Transfer
Vector
DNAs
Viral DNA and transfer vector DNA must be introduced into Spodopteru frugiperda cells so that recombination between them will incorporate the foreign gene into the viral genome. The transfection protocol presented here uses the cationic lipid known as Lipofectin, because this method is simple, efficient, and economical with the viral DNA. Alternative methods for transfection can also be used (see Notes 4 and 5). 1, Seed 35-mm dishes with 1 x lo6 to 1.5 x lo6 Sf21 or Sf9 cells. Incubate at 28’C for 2-24 h to allow the cells to attach. 2. Dilute Lipofectin to 100 pg/mL with sterile water in a polystyrene container (see Note 6). You will require 50 pL of dilute Lipofectin for each transfection. 3. Mix 100 ng of linear viral DNA (5 p.L) with 500 ng of transfer vector DNA (see Note 7), and make up to 50 pL with sterile water. 4. Add 50 @ of the dilute Lipofectm to the DNA solution, and mix gently. Leave at room temperature for 15 min to allow the formation of Lipofectin-DNA complexes. 5. While the DNA is incubating, remove the medium from the cell monolayers (this is conveniently done with an aspirator). Wash the cells twice with 2 mL serum-free TClOO medium or Grace’s medium without serum (see Note 8). Add 1.5 mL of serum-free medium to each dish, and leave at room temperature. 6. Add the Lipofectin-DNA complexes drop wise to the medium covering the cells while gently swirling the dish. Swirl to mix. Incubate at 28°C for 5-16 h. 7. Add 1.5 mL TClOO/lO% FCS to each dish. Incubate at 28°C until 48 h after addition of Ltpofectin-DNA complexes to the cells. 8. Transfer the medium, which contains viruses produced by the transfected cells, to a sterile container, and store at 4°C (see Note 9).
Recombinant
Viruses from Linear
Viral DNA
137
3.3. Isolation of Recombinant Viruses The medium harvested from the cotransfection (Section 3.2., step 8) will contain a mixture of parental and recombinant viruses. To isolate recombinant viruses, a plaque assay is performed on this medium in order to obtain viral plaques derived from individual viruses. The plaque assay protocol presented below is a variation on the method described in Chap-
ter 6 of this book. 1. Seed 35-mm dishes with 1 x lo6 Sf21 or St9 cells in 1.5-3 mL TC100/5% FCS, and incubate overnight at 28°C. Alternatively, seed at 1.5 x lo6 cells/ dish, and incubate at 28’C for 2-6 h (see Note 10). Nine dishes of cells will be needed for each cotransfection to be assayed, plus three for controls. 2. Dilute the medium harvested from the cotransfections into TC100/5% FCS to give final dilutions of 10-l and 10a2. 3. Remove the medium from the cells. Gently add 100 l..tLof the virus inoculum to the center of the dish, taking care not to displace any cells. Infect three dishes with neat cotransfection medium and three dishes with each of the 10-l and 10e2dilutions (see Note 11). As controls, plate 100 ltL of the dilution medium, and 100 PL of dilutions of the parental virus, and of a recombinant virus, which will give 10-20 plaques/dish. (These will be useful for comparison when screening for recombinant plaques.) Incubate at room temperature for 1 h on a level surface, 4. During this incubation, melt the required volume of 3% SeaPlaque agarose, and cool to 45OC.Prewarm an equal volume of TC100/5% FCS to 37°C. 5. Remove the virus inoculum from the cells. (Tilt the dish and aspirate from the edge; see Note 12.) Add the warm TC100/5% FCS to the agarose and mix. Gently add 1.5 mL of this mixture to each dish. When the agarose overlay has set, add 1 mL TC100/5% FCS to each dish. 6. Incubate at 28°C for 3-5 d (see Note 13). 7. Stain for virus plaques. a. To distinguish between P-galactosidase-positive and P-galactosidasenegative plaques (e.g., if the parental virus is AcRP23-kc2 or AcUWl-lacZ), stain with X-gal and neutral red. Prepare a stain mix of PBS containing a 1 in 4 dilution of 0.33% neutral red and a 1 in 12 dilution of 2.5% X-gal. Add 100 l.tL of stain mix to the medium in each dish. Incubate at 28°C for 5-8 h. Remove the stain, and leave the dishes inverted for 16 h to allow the plaques to clear and the blue color to develop fully (see Note 13). b. Alternatively, when not using the P-galactosidase-positive/negative screen (e.g., if the parental virus is AcRP6-SC), visualize the plaques by staining with neutral red only. Dilute 0.33% neutral red 1 in 30
138
Kitts
with PBS. Add 1 mL of diluted neutral red to the medium in each dish. Incubate at 28OCfor 2 h. Remove the stain, and leave the dishes inverted for 2-6 h for the plaques to clear. 8. Using a sterile Pasteur pipet, pick 10-20 isolated plaques with the desired recombinant phenotype (i.e., white when the parental virus was ZucZ-positive) into 0.5 mL aliquots of TC100/5% FCS. Vortex and store at 4°C. Allow an hour or more for viruses to diffuse out of the agarose plug, and then use the medium from the plaque picks as a source of virus for screening (Section 3.4.) and further purification (Section 3.5.) (see Note 14).
3.4. Screening for Recombinant Viruses Although recombinant viruses derived from AcRP234acZ or AcUWlZacZ can be distinguished from the parental viruses after staining the plaques with X-gal, not all the white plaques are recombinant (see Section 1.). If AcRPG-SC is used as the parental virus, it is only possible to screen visually for recombinant plaques when the transfer vector coexpresses P-galactosidase or polyhedrin and the foreign gene (Chapters 2 and 9). For these reasons, it is necessary to screen the putative recombinant plaques picked in Section 3.3., step 8 to identify those that are genuine. Early screening allows false positives to be identified and removed from the plaque purification steps (Section 3.5.). A variety of screening methods can be used, depending on the probes for the foreign gene that are available. The preferred methods are those that detect synthesis of the foreign protein, e.g., Western blotting, ELISA, or a biochemical assay for the expressed protein. If an antibody is not available, Southern blotting with a nucleic acid probe or polymerase chain reaction can be used to confirm that the foreign gene is present in the viral genome. Detailed protocols for the different screening methods will not be presented here since these follow standard methods (6,7). Virus-infected cell proteins or DNA with which to perform these screenscan be obtained using the following protocols. 3.4.1. Analysis of Virus-Infected Cell Proteins 1. Seed35-mm disheswith 5 x lo5 Sf21 or Sf9 cells in 1.5-3 mL TC100/5% FCS, and incubate at 28°C for l-6 h (see Note 15). 2. Remove the medium from the cells. Gently add 100 I.LL of a plaque pick (Section 3.3., step 8) to the center of the dish. As controls, plate 100 pL of the dilution medium and 100 pL of an appropriate dilution of the parental
virus. Incubate at room temperaturefor 1 h.
Recombinant
Viruses from Linear Viral DNA
139
3. Add 1.5 mL TC100/5% FCS to each dish. 4. Incubate at 28OCfor 3-4 d, until the cells look well infected. 5. Gently scrape the cells into the medium. Pellet the cells in a microfuge at low speed for 1 min. Remove the supernatant, and gently resuspend the cells in 0.5 mL PBS using a pipet tip. Repellet, discard the supernatant, and resuspend the cell pellet in 120 pL PBS. 6. Add 30 pL 5X dissociation mix, and boil for 5 min. An aliquot of the denatured proteins can be run on a standard protein gel (8) and analyzed by Western blotting (6). The remainder can be stored at -2OOC.
3.42. Analysis of Virus-Infected
Cell DNA
l-5. As in Section 3.4.1.) except that the final cell pellet should be resuspended in 250 pL TE. 6. Add: 250 pL lysis buffer, 12.5 pL 10 mg/mL proteinase K, and 2 pL 10 mg/mL RNase. Incubate at 37°C for 30 min. 7. Add 500 pL phenol/chloroform. Mix by inversion for 5 min. Spin in a microfuge for 2 min to separate the phases.Transfer the aqueous layer to a fresh tube, and repeat the extraction twice more. 8. Transfer the aqueous layer to a fresh tube, add 50 pL 3M sodium acetate, and 1 mL of ethanol. Chill at -20°C for 10 min. Pellet the DNA in a microfuge for 5 min at room temperature. Add 500 pL of 75% ethanol, vortex briefly, and then repellet the DNA. Repeat the ethanol wash. Dry the pellet at room temperature for 30 min. 9. Add 50 p.L TE to the pellet, Allow the pellet to soak overnight at 4”C, and then gently resuspend the DNA using a pipet tip. Aliquots of this DNA can be restricted and analyzed by Southern blotting. 3.6. Plaque Purification of Recombinant Viruses In order to obtain a homogeneous virus stock, it is necessary to plaque purify the recombinant viruses. This is done by performing a plaque assay (see Section 3.3. or Chapter 6) on dilutions of the original plaque pick (e.g., neat, 10-l and 10-2) and picking a well-isolated plaque with the correct phenotype. This procedure should be repeated for two or three rounds until you are satisfied that the virus stock is pure. Time can be saved by starting the plaque purification at the same time as the putative recombinant plaques are being screened (Section 3.4.). 3.6. Amplification of Recombinant Viruses Once a pure recombinant plaque pick has been obtained, the final step is to amplify the virus to obtain a high-titer stock for long-term storage and further use.
140
Kitts U
M
D Circular
Linear
Fig. 3. Separation of circular and linear forms of AcMNPV DNA on an agarose minigel. Twenty nanograms of AcRP6-SC DNA that had been digested with Bsu361 (D), mock digested (M), or not treated (U) were run on an agarose minigel. The location of circular and linear forms of the DNA is indicated.
4. Notes 1. In order to obtain a high fraction of recombinant viruses, it is vital that all the viral DNA is linearized; circular viral DNA is much more infectious than linear DNA. Hence, even a small amount of undigested viral DNA in the sample will give rise to a high background of nonrecombinant viruses. 2. The viral DNA used in this procedure must be of good quality, preferably purified on CsCl gradients (9), with the majority of the DNA in the circular form (Fig. 3). 3. AcMNPV DNA is large and thus is easily damaged by shearing. Sheared viral DNA loses its infectivity. Therefore, the viral DNA should be handled with care throughout these procedures; mix solutions containing viral DNA by gentle flicking of the tube rather than by vortexing. 4. Cationic lipids other than Lipofectin are available, e.g., DOTAP (&&ringer Mannheim), and Transfectace (GIBCO/BRL). It is likely that these agents could be substitutedfor Lipofectin with minor modifications to the protocol. 5. If the calcium phosphate transfection method is used, 1 pg linear viral DNA and 5 pg transfer vector DNA should be used in each cotransfection. Only CsCl purified plasmid DNA should be used in this method. 6. Lipofectin-DNA complexes stick to the walls of polypropylene containers, e.g., most microfuge tubes. Polystyrene tubes should, therefore, be
Recombinant
Viruses from Linear Viral DNA
141
used for diluting Lipofectin and preparing Lipofectin-DNA complexes. 7. Miniprep plasmid DNA made by the alkaline lysis method (6) usually works well in transfections mediated by Lipofectin. 8. A component of serum inhibits Lipofectin-mediated transfection (10). Therefore, it is necessary to replace the normal medium with serum-free medium before adding the Lipofectin-DNA complexes to the cells. Serum-free Grace’s media will work equally well as TC-100. 9. Occasionally a cotransfection will be overgrown by microbial contaminants because one of the reagents, usually one of the DNAs, is not sterile. If this happens, it is often possible to recover sufficient clean virus for a plaque assayby passing the contaminated medium through a sterile O-45pm filter to remove the microbial contaminants. 10. In order to get good plaque formation, it is important to use cells that are in the exponential phase of growth and to seed them at the optimal density; if too few cells are used, the plaques will be hard to distinguish, and if the initial cell density is too high, the plaques will be small. 11. When linear viral DNA is used in a cotransfection, the total yield of virus is reduced. The medium from such cotransfections is therefore not diluted as much as that from standard cotransfections using circular viral DNA. 12. Take care to remove all the virus inoculum from the dishes; any liquid remaining on the cells when the agarose overlay is added will lead to smearing of the plaques. 13. When using AcRP23-lucZ or AcUWl-ZucZ as the parental virus, it is important to leave the plaque assaysfor at least 4 d, and preferably to stain on the 5th d. If the plaque assayis stained earlier, plaques of the parental virus that are slow to develop may not have produced enough P-galactosidase to produce a strong blue color and may be mistaken for plaques of a recombinant virus. It is also advisable to use a staining procedure that allows plenty of time for P-galactosidase-positive plaques to develop a blue color. The aim is to minimize the number of false-recombinant plaques that are picked because they appear to be white. 14. A preponderance of blue plaques from cotransfections with linearized AcRP23-ZacZ or AcUWl-ZucZ viral DNAs indicates that a significant amount of the viral DNA was uncut. 15. If Western or Southern analysis will be performed using a dot-blot apparatus, this infection can be scaled down fivefold to use 24-well plates, or 20-fold to use 96-well microtiter plates. 16. Clontech, Invitrogen (San Diego, CA), and PharMingen presently supply a variety of versions of precut, linearized genomic AcNPV DNA that circumvent the necessity of purifiying virus, extracting nucleic acid, and cutting the genome to completion.
Kitts
142 References
1. Kitts, P. A., Ayres, M. D., and Possee, R. D. (1990) Linearization of baculovirus DNA enhances the recovery of recombinant virus expression vectors. Nucleic Acids Res. I&5661-5672.
2. Possee, R. D. and Howard, S. C. (1987) Analysis of the polyhedrin gene promoter of the Autographa califomica nuclear polyhedrosis virus. Nucleic Acids Res. 15, 10,233-10,248. 3. Weyer, U., Knight, S., and Possee, R. D. (1990) Analysis of very late gene expression by Autographa califomica nuclear polyhedrosis virus and the further development of multiple expression vectors. J. Gen. Viral. 71,1525-1534. 4. Kitts, P. A. and Possee, R. P. (1993) A method for producing recombinant baculovirus expression vectors at high frequency. BioTechniques 14,810-817. 5. Possee, R. D. (1986) Cell-surface expression of influenza virus haemagglutinin in insect cells using a baculovirus vector. Virus Res. 5,43-59. 6. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual, 2nd ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. 7. Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D. D., Seidman, J. G., Smith, J. A., and Struhl, K. (1987) Current Protocols in Molecular Biology. Wiley, New York. 8. Laemmli, U. K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227,680-684. 9. Bailey, M. J. and Possee, R. D. (1991) Manipulation of baculovirus vectors, in Methods in Molecular
Biology, vol. 7: Gene Transfer and Expression
Protocols
(Murray, E. J., ed.), Humana, Totowa, NJ, pp. 147-168. 10. Felgner, P. L. (1991) Cationic liposome-mediated transfection with LipofectinTM reagent, in Methods in Molecular Biology, vol. 7: Gene Transfer and Expression Protocols (Murray, E. J., ed.), Humana, Totowa, NJ, pp. 81-89.
CHAPTER8
Rapid
Procedures for the Isolation and PCR Analysis of Recombinant Baculovirus
Adrienne Day, Tamara Wright, Arnold SewaZZ, Marjorie Price-Laface, Niharika Srivastava, and Malcolm Finlayson 1. Introduction The baculovirus expression system has proven extremely useful in the investigation of the function and structure of a wide variety of proteins produced by the expression of a foreign gene (I). Essential to this system is the isolation of a recombinant baculovirus expressing the foreign gene. This process has, until recently, been an onerous and time-consuming task. A large number of manipulations are required before the desired gene product can be produced at high levels. The first step in the process of isolating a recombinant virus requires that the foreign gene be inserted into an appropriate transfer vector. The resultant vector and the wild-type viral DNA are cotransfected into insect cells and yield recombinant virus at a frequency of 0. l-l % via a process of homologous recombination (2,3). The recombinant does not produce occlusion bodies, but instead contains an intact copy of the foreign gene in place of the polyhedrin gene (4). Recombinant viruses are identified visually on the basis of the absence of occlusion bodies, when viewed under a low-power dissecting microscope. This is normally very difficult without previous experience, but is greatly assisted by the use of a transfer vector that coexpresses the P-galactosidase gene (5). The resultFrom: Methods m Molecular Biology, Vol. 39: Baculowrus Expression Protocols Edited by: C. D. Richardson Q 1995 Humana Press Inc., Totowa, NJ
143
144
Day et al.
ant recombinants when plated in the presence of a chromogenic substance will be blue in color. The purification of the recombinant would then usually require two to three rounds of plaque purification. Once a pure recombinant has been obtained, it is then advisable, before assaying recombinant protein expression, to verify that the DNA genome of the recombinant virus contains the gene of interest. This has traditionally been achieved using either Southern blot analysis (6) or DNA dot-blot hybridization (7). The greatest difficulties in this process are: 1. The low frequency of recombination between the transfer vector and the wild-type DNA to result in the production of a recombinant virus. 2. The identification of the recombinantvirus. 3. The purification of the recombinantvirus away from the wild-type virus. Recent developments in our laboratory have enabled us to reduce significantly the time required and the degree of difficulty involved in the production and isolation of recombinant virus. The use of noninfectious linearized viral DNA has increased the recombination frequency to a level of 30% (8) and significantly lowers the wild-type background. The circular genome of the baculovirus was linearized by the introduction of a unique restriction enzyme site into the polyhedrin gene. Recombination repair of the linearized genome will result in the production of a recombinant virus. When the linearized viral DNA and the transfer vector are introduced into the insect cells using cationic liposomes (9), rather than the traditional method of calcium phosphate-mediated transfection, the increase in transfection efficiency is such that the frequency of recombination can be increased up to 60%. As a result of these two improvements, isolated recombinant virus can be obtained on the first round of plaque screening. The subsequent purification of these isolated recombinants has been further simplified by the use of a microtiter plate assay and PCR analysis. Polymerase chain reaction (PCR) technology provides a fast, sensitive, nonradioactive method for amplifying DNA segments of known sequence with two flanking oligonucleotide primers (10). It can therefore be used to detect the presence of a foreign gene in a putative recombinant virus (II). Following plating and identification of recombinant plaques, plugs containing the isolated recombinant virus are transferred
PCR Analysis
of Recombinant
Baculovirus
I45
directly to 12-well microtiter plates containing a small number of insect cells. Infection proceeds rapidly, and DNA can be isolated on day 3 postinfection from the cells for PCR analysis. The PCR amplification and visualization of the PCR product on an agarosegel can be completed in 1 d. Recombinants are identified on the basis of their electrophoretic mobility when compared to marker DNA. In addition, the purity of the recombinant virus stock can be assessedby checking for the amplification of the polyhedrin gene from wild-type virus using primers for the polyhedrin gene. These developments significantly increase the easewith which recombinant virus can be isolated and identified for analysis of recombinant protein expression. 2. Materials 1. Spodopteru frugiperdu (Sf9) insect cells (ATCC, Accession No. CRL
2.
3.
4. 5. 6. 7. 8.
1711; Invitrogen [San Diego, CA] Cat. No. 825-01) maintained at log phase (l-2 x lo6 cells/mL) in TNM-FH complete media. Healthy suspension cultures should have a mean doubling time of 18-22 h and a viability of 298%, as determined by trypan blue exclusion. Grace’s Antherea medium (12), a relatively simple mixture of salts, carbohydrates and amino acids, is available from commercial suppliers (Invitrogen, Whittaker [Walkersville, MD], JRH Biosciences [Kansas City, MO], or GIBCO [Gaithersburg, MD]). Store at 4°C. It is stable for 6 mo. TNM-FH (I3), a more complete medium suitable for routine growth of cells, is prepared by addition of 3.3g/L yeastolate and 3.3 g/L lactalbumin hydrolysate (both available from Difco) to Grace’s medium. For complete growth media, add 10% fetal bovine serum (sterilized and heat-inactivated) and 5OpglmL Gentamycin (GIBCO). All media should be filter-sterilized after preparation, and stored at 4°C will be stable for 1 mo. Cationic liposomes formed from the cationic lipid dimethyldioctadecylammonium bromide and neutral lipid dioleoyl+a-phosphatidylethanolamine are available commercially from Invitrogen or BRL. AcNPV circular DNA was linearized by introduction of an SseI restriction endonulease site between the Nae I site at position 4396 and the start of the polyhedrin gene at position 4718 (Invitrogen, San Diego, CA). Transfer vectors, baculovirus primers, and agarose for plaque assays are all available commercially from Invitrogen and other suppliers. 20% Polyethelyene glycol (PEG) 4000 in 1M NaCl (both analytical-grade from Sigma), stable at room temperature. Proteinase K 280 U/mL (EM Scientific). Store at -2OOC. It is stable for 6 mo. Mussel glycogen 2 mg/mL in water, stable when stored at -2OOC.
146
Day et al. Table 1
AcNPV Transfer Vectors” Vector pAc360 pVL1392 pVL1393 pBlueBac2 pBlueBac3 pBlueBacHis
Features Fused; natural polyhedrin leader; +34; unique BumHl site. Nonfused; natural polyhedrin leader; +35; polylinker; mutated ATG. Nonfused; natural polyhedrin leader; +35; polylinker; mutated ATG Nonfused; natural polyhedrin leader; unique BamHI and NheI site; coexpression of P-galactosidase driven by ETL promoter (14). Nonfused; natural polyhedrin leader; unique NheI, BumHI, PstI, NcoI, and Hi&III sites; coexpression of B-galactosrdase driven by ETL promoter (14). Nonfused; no polyhedrin leader; ATG, (histidine&, EK cleavage site, multiple cloning site cassette in three different reading frames;
coexpression of P-galactosidase driven by ETL promoter (14). aAcNPV transfer vectors for the production of nonfused and polyhedrin-fused proteins. The mseruon site of foreign DNA relative to the polyhedrin translation ATG (+1,+2,+3, and so forth) is noted.
9. 10X PCR Buffer: 100 mM Tris-HCl, pH 8.3,500 mM KCl, 25 mM MgC12, and 0.01% gelatin, Stored at -20°C, the buffer is stable for 6 mo. 10. Thermophilic DNA polymerase (MBR, Milwaukee, WI) or Taq DNA polymerase (Perkin Elmer-Cetus, Palo Alto, CA), 1.SU/pL. Store at -20°C. 11. 1% Agarose (molecular-biology-grade reagent from Sigma, St. Louis, MO) in 1X TAE buffer. 12. 1X TAE buffer; 4 mM Trrs-HCl, pH 7.8, 2 mM sodium acetate, and 0.2 mM EDTA. 13. Loading dye: 0.5% bromophenol blue, 100 mM Tris-HCl, pH 7.5, 10 miV EDTA, 2% SDS (lauryl sulfate), and 30% glycerol. 14. X-gal or Bluo-gel (GIBCO/BRL) stock solution (50 mg/mL) made up in N,N-dimethylformamide.
3. Methods
3.1. Selection
of a Transfer Vector When determining which of the available transfer vectors (see Table 1) to use in the generation of a recombinant baculovirus vector, a number of factors must be taken into consideration: 1. Whether to produce a recombinant protein as a fusion or nonfusion product. 2. Whether to optimize recombinant expression by maintaining an unmodified polyhedrin leader sequence. This may involve manipulation of the 5’ end of the foreign gene.
PCR Analysis
of Recombinant
Baculovirus
147
3. Whether to screenfor the recombinantplaquesvisually with the coexpression of P-galactosidase.For the novice, we recommend, where possible, the use of the transfer vector that coexpressesthe P-galactosidasegenein the generationof the recombinantvirus, since it allows rapid detection of recombinantsby virtue of the blue coloration. The baculovirus expression vectors that we currently use in our laboratories are summarized in Table 1. The production of fused proteins requires that foreign genesbe inserted in frame with the transfer vector initiation codon, ATG. The production of nonfused proteins requires that DNA inserts contain their own translation initiation site. Generally, transfer vectors that contain intact polyhedrin leader sequences (fusion vector pAc360, and nonfusion vectors pBlueBac2, pVL1392, pVL1393) yield higher levels of expression than vectors that contain interrupted leader sequences (such as the pAc700 series and pAc373) (3). However, protein translation may initiate at the mutated ATG [ATT] of the nonfusion vectors, and if the recombinant gene is inserted in frame with the polyhedrin open reading frame, this will result in two recombinant expression products: the nonfused recombinant protein and the recombinant protein fused with polyhedrin (15). Comprehensive studies of the expression of recombinant proteins using different transfer vectors are available (for reviews, see refs. 1 and 3). After selecting a transfer vector that will meet your requirements, clone the foreign DNA into the transfer vector, and prepare at least 10 pg of highly purified plasmid DNA using standard techniques (16). Sf9 cells are sensitive to some contaminants found in crude plasmid preparations that cannot be removed by phenol extraction or ethanol precipitation. At present, the only consistently reliable method we have found for plasmid purification is CsCl-ethidium bromide gradient centrifugation. Impure preparations of plasmid DNA are toxic to the cells, and many cells may lyse shortly after transfection. This results in an apparently lower recombination frequency and increased difficulty in detecting recombinant virus. To test the quality of a plasmid DNA preparation, include in all transfection experiments a control with AcNPV DNA and cells alone. At about 24 h posttransfection, compare these cells with those transfected with AcNPV DNA plus plasmid DNA. Sf9 cell viability should be >98% for transfection experiments. Please note that the quality of the plasmid DNA may be critical to the successfulconstruction of recombinants.
148
Day et al.
3.2. Transfection of Sfs Cells Plasmids containing foreign genes are cotransfected into insect cells with the linearized wild-type AcNPV DNA by the technique of cationic liposome-mediated transfection (see Note 1). 1. Using log-phase Sf9 cells (l-2 x lo6 cells/ml) 2 98% viable, seed 2 x lo6 Sf9 cells in complete TNM-FH in a 60-mm plate. Allow the cells to attach for at least 30 min. 2. For each 60-mm plate to be transfected, set up the following transfection mix. Pipet 1 mL of Grace’s media containing no supplements into a 1.5 mL sterile Eppendorf. Add 1 ug of linear AcNPV DNA, 3 pg of plasmid DNA (the transfer vector containing the gene of interest), and 20 ,clLof the cationic liposome solution. Vortex (see Note 1). 3. Incubate at room temperature for 15 min. 4. During this incubation period, remove the media from the plates and replace with 2 mL of Grace’s media without supplements. Allow the cells to sit until the incubation in step 3 is within 2 min of completion. This should be approx 10 min. 5. Remove all of the Grace’s media from the cells, and add the 1-mL transfection mix dropwise to the 60-mm plate. 6. Incubate plates on a slowly rocking platform at room temperature for 4 h. (Use setting of 2.5 for a Bellco #774020020 side/side rocking platform.) 7. Following this incubation period, add an additional 1 mL of complete TNM-FH media to each 60-mm plate and incubate at 27OCin a humidified environment for 48 h. 8. Forty-eight hours later, harvest the media from the transfected plates, and store at +4OC until required. This media is the virus stock that will be used in the identification of recombinant virus by plaque assay. Replace the media with 3 mL of fresh complete TNM-FH, and incubate the cells at 27°C for a further 48 h. Although extracellular viral particles have been released into the media, cells may not yet show signs of infection. 9. Check the cells on day 4 posttransfection to visually confirm a successful transfection. This is done using an inverted phase microscope at 250400X magnification. Ten to 50% of the cells should contain viral occlusions, which appear as refractive crystals in the nucleus of the insect cell. Other positive signs of virus infection include a 25-50% increase in the diameter of the cells, a marked increase in the size of the cell nuclei relative to the total cell volume (the nuclei may appear to “fill” the cells), and finally cell lysis and debris.
PCR Analysis
of Recombinant
3.3. Plaque
Purification
149
Baculovirus
of Recombinant
Virus
3.3.1. Preplating Preparations 1. Prepare a 2.5% solution (w/v) of baculovirus grade agarose in distilled water. Autoclave to dissolve the agarose and to sterilize the solution. Incubate at 50°C until needed. For example: for 20 plates, add 1.25 g of baculovirus grade agarose/50 mL of Nanopure (8 M&2) water in a 100 mL bottle. Autoclave for 15 min. 2. Prepare TNM-FH complete media. 3. For each 50 mL of baculovirus grade agarose made, aliquot 50 mL of TNM-FH complete media to a sterile 100-n&, bottle. Place at 50°C until needed. Five milliliters of this solution are used per 100~mm plate. 4. Prepare lo-fold dilutions of virus inoculum (media harvested from transfections) in the range of the expected titer allowing approx 1 mL of diluted virus for each plate. It is essential that the viral inoculum be vortexed vigorously prior to the preparation of the dilutions. We routinely plate at dilutions of 10-i and 10” when transfection is performed using cationic liposomes. 5. Seed Sf9 cells at a density of 5 x 106cells/100 mm plate in complete media. Duplicate plates should be used for each virus inoculum to be tested. Rock plates to distribute cells evenly at room temperture on a side/side rocking platform (use a setting of 5 for a Bellco #774020020 side/side rocking platform) for at least 30 min. Examine the cells to confirm that cells have attached. Remove plates from the rocker, and sit at room temperature for at least 30 min before using (see Note 2). 6. For each loo-mm plate, add 5 mL of TNM-FH complete media (plus chromogenic substrate if appropriate) to a sterile 15 mL Falcon tube. Set aside until needed. When plating involves screening of constructs developed from the pBlueBac2 transfer vector, the chromogenic substrate 5-bromo4-chloro-3-indolyl-P-D-galactoside (X-gal) or halogenated indolyl+D-galactosidase (Bluo-gal) is incorporated into this prealiquoted media at a concentration of 150 ltg/mL. For example: Add 150 l.t.Lof a 50 mg/mL Bluo-gal stock solution (made up in NJVdimethylformamide) to 50 mL of TNM-FH complete media (see Note 3). 3.3.2. Plating Dilutions of Virus from Transfections on Sf9 Cells 1. Remove all but 2 mL of media from the cells once they have firmly attached. We generally plate cells in about 5 mL so you would remove 3 mL. Do not do more than eight plates at once.
150
Day et al.
2. Add 1 mL of each viral dilution dropwise to the appropriately labeled plate. Following addition of the viral dilutions, incubate the plates at room temperature on a slowly rocking platform for 1 h. (Use a setting of 2.5 for a Bellco #774020020 side/side rocking platform.) 3. During this incubation period, heat a water bath to a temperature of 46OC, and place in the laminar flow hood. 4. Just prior to the end of the l-h incubation, retrieve the autoclaved agarose and media that had been incubated at 50°C. Combine one bottle of agarose with one bottle of media, mix, and place in the water bath. 5. Place the prealiquoted 5-mL lots of media (with or without chromogenic substrate) in the laminar flow hood. Have ready at least enough lo-mL sterile pipets for each plate. 6. At the completion of the incubation period, begin working with the first set of eight plates infected. Completely remove the media from the plates, then working quickly while maintaining sterile technique, withdraw 5 mL of the agarose media mix and add to a tube containing the 5 mL of prealiquoted media. Mix immediately by inverting. Then, working from the edge, gently pour the mixture onto one of the plates from which media was removed. Continue until all eight plates have been completed. Care must be taken not to move the plates until the agarose has set (see Note 3). 7. Repeat step 6 until all of the plates have been completed. 8. Incubate plates in a humid environment for 5-6 d, or until plaques are well formed. If a 27°C incubator with high (SO-loo%) humidity is not available, seal the plates m a plastic bag or sealed container, and add slightly damp paper towels to provide humidity. 3.4. Visual Screening fir Recombinant Plaques 1. When plaques are distinct (6 d postinfection), examine plates using a dissecting microscope with a magnification of 30-40x. Place the plate upside down on a nonreflective dark surface (e.g., black velvet), and illuminate the plate from the side using an intense light source. Adjust the angle of the light until the plaques can be observed (usually, a 45’ angle or greater is best). Against a black, nonreflective background, the wild-type plaques should look shiny and almost crystal-like, whereas recombinants will be a dull milky-white color. 2. If the pBlueBac transfer vector has been used to generate the recombinant virus, and a chromogenic substrate included in the agarose the recombinants will be clearly distinguishable since they are blue in color.
PCR Analysis
of Recombinant
Baculovirus
151
3. Look at the plates that have been infected with a lOA dilution of virus first. This dilution should result in plaques that are well separated from each other. 4. Scan the plate at a magnification of 3uOx, and circle any plaques you suspect may be recombinant, occlusion body negative, 5. Reexamine each circled plaque under an inverted phase microscope at 200-400x. Viral plaques are observed as a clear area in the cell monolayer that is ringed by infected cells, which are morphologically distinct from the uninfected cells. They are generally larger in diameter, display a marked increase in the size of the nuclei relative to the total cell volume, and show signs of cell lysis. Examine the entire plaque area for the presence or absence of occlusion bodies. To avoid several rounds of screening it is important that only recombinants that are totally free of occlusion bodies be selected. 6. When several putative recombinant plaques have been located, check the circled plaques again using a dissecting scope. Place a tiny dot within each circle, directly over the plaque to be picked. 1. 2. 3. 4. 5. 6. 7.
8.
3.45. Purification of the Recombinant Virus Prewet the wells of a 12-well microtiter plate with 2 mL of complete TNM-FH media. Seed 5 x lo5 Sf9 cells in each well. The total volume of the wells should not exceed 3 mL. Using a sterile Pasteur pipet and bulb carefully remove the agarose over the recombinant plaque. Transfer the agarose plug containing the plaque to one of the wells in the microtiter plate. Repeat steps 3 and 4 until 11 of the 12 wells have been infected. The 12th well acts as a cells-only control. Seal the microtiter plate with pamfihn, or place the plate in ahumidified incubator or a sealed plastic bag with a moist paper towel. Incubate at 27OCfor 3 d. On day 3, screen the wells visually for the presence of occlusion bodies. At this stage, they may not be obvious. However, by day 5, they should be clearly visible (see Note 5). For wells that are occlusion-negative, pipet the media in the microtiter plate well up and down to dislodge the cells, then remove 0.75 mL of media, and place in a sterile Eppendorf tube. This will be used to carry out PCR (I 7) analysis of the putative recombinant virus. The plates should be kept and incubated at 27°C until all of the cells lyse. The media from these wells should be harvested and stored at 4OC. This is the Pl virus stock.
152
Day et al. Table 2 PCR Analysis of Baculovirus Vectors
Fusion vector
PCR product
pAc360
650 bp
AcNPV wild
838 bp
Nonfusion
vector
PCR product
pVL1392, pVL139 pBlueBac2 pBlueBac3 pBlueBacHisA pBlueBacHisB pBlueBacHisC
650 650 672 735 725 733
bp bp bp bp bp bp
3.6. PCR Analysis of Recombinant Virus The forward and reverse PCR primers (Invitrogen) are chemically synthesized oligomers that are complementary to the polyhedrin loci. The polyhedrin gene spans base +l to +738. As shown below, the forward and reverse primers have been designed to flank this region: Forward primer:
S(-44)-TITACTG’ITITCGTAACAG’ITITG-3’
T,,,= 62°C Reverse primer: 5’(+794)-CAACAACGCACAGAATCTAG-3’ T, = 58°C These primers are compatible with all polyhedrin promoter-based baculovirus transfer vectors that contain 5’ and 3’ homologous polyhedrin sequences.When amplified without inserts, the transfer vectors give fragments of the sizes shown in Table 2. Since the size of the foreign gene insert is known, the size of the PCR product expected can easily be determined and wild-type contamination detected. 3.6.1 Purification
of DNA for PCR
1. Take the sample generated in Section 3.5., step 7, and pellet the cell debris by centrifugation at 5OOOg for 3 min. 2. Transfer the supernatant to a fresh Eppendorf tube, and add 0.75 mL of a solution of 20% polyethylene glycol (PEG) in 1M NaCl. Invert twice to
mix, and allow to standat room temperaturefor 30 min. 3. Centrifuge
at 14,000g for 10 min at room temperature.
Remove all media
from the pellet. An additional quick spin may be requiredto remove trace amountsof media. The pellet may not be visible at this point. 4. To the pellet add 100 pL of sterile water. Carefully wash the sides of the tubesto ensurethat all of the viral particles areresuspended. 5. Add 10 nL of proteinase K (280 U/mL),
and incubate at 50°C for 1 h.
PCR Analysis
of Recombinant
Baculovirus
I53
Table 3 PCR Reaction Reagent DNA Sample 10X PCR Buffer 25 mM dNTP’s 100 ng/j.& Forward Primer 100 ng&L Reverse Primer Thermophilic DNA Polymerase Sterile water added to final volume Overlay with 50 a of mineral oil.
Per 50 uL reaction 5cLL 5cLL 1w 1 PL 1 PJ-J 1.5 u 50 PL
Final concentration 1Xa 0.5 mM 2 ng/pL 2 n&L 0.03 u&L
aIn most cases 10X PCR buffer is supplied with the enzyme
6. Extract with an equalvolume of phenol/chloroform(1: 1). Centrifuge in an Eppendorf centrifuge for 5 min. Transfer the upper aqueousphase to a fresh sterile tube. 7. PrecipitatetheDNA by acidiig l/10 vol3MNaOAc, 5 pL of musselglycogen, and2 vol of 100%ethanol(seeNote 5). Incubateat -20°C for at least20 min. 8. Centrifuge at 14,000gfor 15 min at 4°C. Wash pellet with 80% ethanol. Spin at 14,000gfor 5 min to remove all tracesof the ethanol. 9. Resuspendthe pellet in 10 pL of sterile water. The DNA is now ready for analysisby PCR. 3.6.2. PCR Procedure Include 10 ng of DNA in a final volume of 5 PL. Use each of the following DNA samples as controls: 1, Wild-type AcNPV DNA. 2. Transfer vector (no insert). 3. Recombinanttransfer vector (with insert). Analyze 5 PL of each DNA sample generated in section 3.6.1.) step 9. To a sterile microfuge tube, add the reagents shown in Table 3. Amplify samples for 30 cycles in a thermocycler (EricComp Inc. Single Block System) using the following profile: 1 min at 94°C. 2 min at 55OC.. 3 min at 72OC(longer incubation recommendedfor amplification of large inserts). Following the last cycle, a final 72°C extension step for 7 mix-ris performed. Samples are then cooled to 3O”C, and 5 PL are analyzed by agarose gel electrophoresis.
154
Day et al. Table 4 Efficiency of Generating Recombinant Virus
Transfection technique0
Color of viral plaque White Blue
Total no. plaques
% Recombinant virus
2.9%
Titer plated
Circular DNA Calcium phosphate
492 649 579
15 18 13
507 667 592
10-Z
Linear DNA Calcium phosphate
40 28 27
17 15 19
57 43 46
29.8% 34.9% 41.3%
lo-’
Linear DNA Liposomes
55 45 50
94 93 82
149 138 132
63.0% 67.4% 62.1%
IO-4
2.6% 2.2%
a BlueBac2 expression vector was used to generate recombinants in 3 sets of experiments.
3.6.3. Agarose Gel Electrophoresis 1. Prepare 100 mL of 1% agarose in 1X TAE buffer. Microwave until the agarose is completely dissolved. 2. Allow to cool to 50°C before pouring into a clean mini-gel apparatus (9 x 17 cm) to a thickness of approx 5 mm. Place the comb at the negative electrode end with its teeth just above the surface that forms the bottom of the gel. 3. After the agarose is completely solidified, pour the 1X TAE into the apparatus to cover the gel. Remove the comb slowly, and make sure the wells are submerged in running buffer. 4. Remove 5 @ of PCR reaction from beneath the mineral oil layer, and place in a fresh tube. Ensure that none of the oil is carried over. Add 2 l.kL of loading dye. Load samples carefully into the wells. 5. Run gel at approx 80 V until the blue dye has migrated 3/4 of the length of the gel. Stain in ethidium bromide, and visualize on a shielded UV transilluminator. 3.7. Efficiency for Generation of Recombinants The percentage of recombinants obtained when circular or linear viral DNA is cotransfected with pBlueBac1 into Sf9 cells using calcium phosphate are shown in Table 4. Although the total number of plaques obtained with the linear DNA is less than that seen with the circular, there is an increase of between lo- and 20- fold in the percentage recombination achieved. When cationic liposome-mediated transfection is used
PCR Analysis
of Recombinant
Baculovirus
Fig. 1, PCR analysis of recombinant viral DNA produced using the BlueBac2 vector. Lane 1. DNA markers, 1-kb ladder (BRL); Lane 2. Recombinant transfer vector pBlueBac2 with 800-bp insert. Lanes 3-10. Putative plaque pure recombinant virus. Lane 11. Wild-type Viral DNA. Lane 12. DNA markers, 1-kb ladder (BRL). to introduce the linear viral DNA and transfer vector into Sf9 cells, the percentage of recombinants is doubled (see Table 4). Linear viral DNA and the transfer vector pBlueBac2, into which a DNA fragment of 800 bp was inserted, were introduced into Sf9 cells using cationic liposome mediated transfection. Plaque assays indicated that recombination had occurred at a frequency of 53%. Eight isolated recombinants were picked and used to reinfect cells in a 12-well microtiter plate. Three days postinfection DNA was isolated and PCR analysis carried out. The resultant PCR products, which were examined by gel electrophoresis, are shown in Fig. 1. As can be seen in Lane 2, the PCR product produced from the insertion of an SOO-bp insert into the transfer vector pBlueBac2 is 1450-bp. The wild-type viral DNA generates a product of -840 bp (see Lane 11). The recombinant virus used to generate the DNA analyzed in Lanes 3-9, contains only recombinant virus, whereas that in Lane 10 is clearly contaminated with wild-type virus. Thus, following only one round of plaque purification, seven pure recombinant clones have been
156
Day et al.
Fig. 2. PCR analysis of recombinant viral DNA produced with the pVL1393 vector. Lane 1. DNA markers, 1-kb ladder (BRL). Lane 2. Wild-type viral DNA. Lane 3. Sample 1. Recombinant transfer vector pVL1393 with 400-bp insert. Lanes 4-6. Putative recombinant virus generated from sample 1. Lane 7. Sample 2. Recombinant transfer vector pVL1393 with 700-bp insert. Lanes 8-10. Putative recombinant virus generated from sample 2. Lane 11. Sample 3. Recombinant transfer vector pBlueBac2 with 900-bp insert. Lanes 12 and 13. Putative recombinant virus generated from sample 3. Lane 14. Sample 4. Recombinant transfer vector pBlueBac2 with 1300-bp insert. Lanes 15-16. Putative recombinant virus generated from sample 4. Lane 17. Wild-type viral DNA. Lane 18. DNA markers, 1-kb ladder (BRL).
obtained. This process can be performed with the same ease and efficiency when a transfer vector that does not express P-galactosidase is used to generate the recombinant virus. Further, the PCR analysis does not seem to be affected by the size of the foreign DNA inserted into the transfer vector. These results are shown in Fig. 2. The transfection of each construct was carried out as described in Section 3.2. In the case of sample 1, a 400-bp fragment was inserted into pVL1393, and for each of the three putative recombinants picked, a pure PCR product of the correct size (1.1 kb) was produced. Of the three recombinant viral plaques examined (after cotransfection of sample 2 [pVL1393 with a 700-bp
PCR Analysis
of Recombinant
Baculovirus
157
insert] with linear viral DNA), two were plaque pure, whereas one (Lane 8) was contaminated with wild-type virus. Again all clones gave PCR products of the correct size, 1.4 kb. Samples 3 and 4 were generated by insertion of a 900-bp and a 1.3-kb fragment, respectively, into the transfer vector pBlueBac2. The expected PCR products-l .7 kb for sample 3 and 2.1 kb for sample 4-were obtained, and in each case, the recombinant examined was plaque pure. 4. Notes 1, The mechanism by which liposomes containing positively charged lipids mediate transfection of DNA into cells is not fully understood. It is thought that the negatively charged DNA binds to the positively charged surface of the liposomes. Residual positive charge then mediates interaction with the negatively charged sialic acid residues on the cell’s surface. The presence of proteins in the transfection mix inhibits transfection, since the positively charged proteins compete with the liposomes for binding to the DNA. It is essential therefore that the transfection be carried out under protein-free conditions. Increasing the amount of DNA in the transfection reaction may also result in a decrease in the efficiency of transfection, since the DNA masks all of the positive charges on the liposomes, hindering their interaction with the cell membrane. It should be noted that the transfection time, DNA concentration, and amount of liposomes described here have been optimized for Invitrogen cationic liposomes. Different commercially available liposome preparations are formed using different ratios of positive to neutral lipids and will therefore differ in their activity. 2. It is essential that a good monolayer of evenly distributed cells is formed when carrying out this assay.If a good monolayer does not form, it will be very difficult to visualize the plaques. Prewetting the plates with media and rocking the plates while the cells are attaching helps to achieve this even cell distribution. If the plaques produced are very small in size, it indicates that too many cells were plated to form the monolayer. Try seeding fewer cells. If the cells have been stressedwhen the agarose overlay is applied, there will be no cell growth. Alternatively, if the agarose begins to set as the overlay is applied, it will be lumpy, making visualization of the plaques difficult. If a dilution of lo4 does not result in the production of plaques that are well separated then either: a. Virus titers are too low as a result of an inefficient transfection. It is advisable to repeat the plaque assay using lower dilutions or, altematively, to repeat the transfection; or b. The titer is higher than expected, and the plaque assay should be repeated using a higher dilution.
Day et al.
158
We have noted that when using the linear viral DNA, ghost plaques appear at a approx 10%. These plaques are occlusion body negative, but are not blue in color, It is thought that they arise from recircularization of the linear DNA via homologous recombination such that the polyhedrin gene is either deleted or removed. When generating a recombinant virus with a transfer vector that does not coexpress P-galactosidase, the ghost plaques may be selected as recombinants. However, on PCR analysis, they should be distinguishable from true recombinants. 3. If you experience problems with agarose solidifying, work with only one tube at a time, and plpet the mixture slowly onto the plate from the edge rather than pouring it. 4. Wells containing occlusion bodies are not plaque pure and require additional rounds of purification. The media from these wells can be used to perform another plaque assay in order to obtain a pure recombinant virus. Should the 12-well plates become infected with either fungus or bacteria, passage of the media through a 0.2 p disposable syringe filter will remove the contamination. The filtered media may then be used to reinfect fresh cells.
5. The addition of the mussel glycogen carrier is essential to obtaining a good yield of DNA and should not be omitted.
References 1. Luckow, V. A. and Summers, M. D. (1988). Trends in the development of baculovirus expression vectors. Biotechnology 6,47-S. 2. Luckow, V. A. and Summers, M. D. (1989). High level expression of nonfused foreign genes with Autographa califomica nuclear polyhedrosis virus expression vector. Virology 170,3 l-39. 3. Webb, N. R. and Summers, M. D. (1990). Expression of proteins using recombinant baculoviruses. Technique 2, 173-188. 4. Burand, J. P., Summers, M. D., and Smith, G. E. (1980). Transfection with baculovirus DNA. Virology 101,286-290. 5. Vialard, J., Lalumiere, M. Vernet, T., Briedis, T., Briedis, D., Alkhatib, G., Henning, D., Levin, D., and Richardson, C. (1990). Synthesis of the membrane fusion and hemagglutinin proteins of measles virus, using a novel baculovirus vector containing the P-galactosidase gene. J. Virol. 64,37-50. 6. Southern, E. M. (1975). Detection of specific sequences among DNA fragments separated by gel electrophoresis. J. Mol. Biol. 98,503-517. 7. Kafatos, F. C., Jones, C. W., and Efstratiadis, A. (1979). Determination of nucleic acid sequences homologies and relative concentrations by dot hybridization proce-
dures.Nucleic Acid Res. 7, 1541-1552. 8. Kitts, P. A., Ayres, M. D., and Possee, R. D. (1990). Linearization of baculovirus DNA enhances the recovery of recombinant virus expression vectors. Nucleic Acids Res.
18,5667-5672.
PCR Analysis
of Recombinant
Baculovirus
159
9. Hartig, P. C. and Cardon, M. C. (1991). Generation of recombinant baculovirus via liposome mediated transfection. Biotechniques 11,310-313. 10. Innis, M. A., Gelfand, D. H., Sninsky, J. J., and White, T. J. (1990). PCR Protocols. A Guide to Methods and Applications. Academic, New York. 11. Webb, A. C., Bradley, M. K., Phelan, S. A., Wu, J. Q., and Gehrke, L. (1991) Use of the polymerase chain reaction for screening and evaluation of recombinant baculovirus clones. Biotechniques 11,s 12-5 18. 12. Grace, T. D. C. (1962) Establishment of four strains of cells from insects grown in vitro. Nature 195,788-789.
13. Hink,
W. F. (1970) Established
Trichoplusia
insect cell line from the cabbage looper,
ni. Nature 226,466467.
14. Crawford, A. M., and Miller, L. K. (1988) Characterization of an early gene accelerating expression of late genes of the baculovirus Autographa californica nuclear polyhedrosis virus. J. Viral. 62,2773-278 1. 15. Beames, B., Braunagel, S. C., Summers, M. D., and Lanford, R. E. (1991) Translational initiation from an AUU codon of a baculovirus expression vector. Biotechniques
11,378-383.
16. Maniatis, T., Fritsch, E. F., and Sambrook, J. (1989) Molecular Cloning (A Luboratory Manual). Cold Spring Harbor Laboratory, Cold Spring Harbor, New York. 17. Saiki, R. K, Scharf, S., Faloona, F., Mullis, K. B., Horn, G. T., Erlich, H. A., and Arnheim, N. (1985) Enzymatic amplification of P-globin genomic sequences and restriction analysis for diagnosis of sickle cell anemia. Science 230, 1350-1354.
CHAPTER9 Production of Recombinant Baculoviruses Using Rapid Screening Vectors that Contain the Gene for P-Galactosidase Manon LalumiSre and Christopher D. Richardson 1. Introduction Baculovirus/insect cell expression systems have become a popular choice for the production of high levels of recombinant proteins (I-3). However, the identification and selection of the recombinant baculovirus have been difficult and time-consuming. Screening recombinant viruses and purifying virus containing the foreign gene away from contarninating wild-type virus can be laborious. The process can involve numerous rounds of plaque purification by using visual screening or hybridization techniques to detect recombinant plaques. A key complaint from newcomers to the field of baculovirology is that recombinant viral plaques are almost invisible in comparison to those produced by wild-type virus. Our laboratory has worked on the improvement and refinement of expression vectors for introducing recombinant baculovirus into insect cells (4-6). An approach where P-galactosidase gene and foreign gene under the control of two different promoters recombine as a unit with wild-type viral DNA has proven to be very effective with vaccinia virus (7). With this precedent, we produced baculovirus expression vectors that contained the P-galactosidase gene, a restriction site for introducing a foreign gene, the upstream and downstream flanking sequences of the From: Methods in Molecular Biology, Vol. 39: Baculovirus Expression Protocols Edited by: C. D. Richardson Q 1995 Humana Press Inc., Totowa, NJ
161
162
Lalumibre
and Richardson
polyhedrin gene, and a bacterial shuttle plasmid. The polyhedrin mRNA leader sequence of pVL941 (see Chapter 2) was retained in these vectors. Immediate early (IEl), early (ETL), or very late (PlO) promoters directed the synthesis of P-galactosidase, whereas the strong polyhedrin promoter controlled the transcription of the foreign gene. P-galactosidase and the foreign gene recombine with wild-type genomic DNA to yield recombinants that are polyhedrin-negative, produce the foreign gene product, and form blue plaques when Bluo-gal or X-gal is present in the agarose overlay. The IEl, ETL, or PlO promoters modulate the levels of P-galactosidase produced since minimal amounts of the indicator are often desirable when one wishes to purify the foreign recombinant protein. Throughout our work, we required a standard reporter gene that could be assayed quickly and easily in order to assesslevels of gene expression. The firefly luciferase has previously been used as a reporter in a number of eucaryotic systems including baculovirus (8-21). To be slightly different, we decided to use bacterial luciferase as a foreign gene/ reporter in our new vectors. Bacterial luciferases are widely distributed throughout nature (22,23). Luminescent bacteria occur as free-living organisms in the ocean, as saprophytes growing on dead fish or meat, as gut symbionts in the digestive tracts of marine fish, as parasites in crustacea, and as light-organ symbionts in the teleost fish and squid. The bacteria can be classified into three genera, including Vibrio, Photobacterium, and Xenorhabdus. We used a fusion of LuxA and LuxB genes (24) that codes for the heterodimer enzyme of Vibrio harveyi as a reporter to test the efficiency of two improved baculovirus expression vectors, pPl0 and pETL. The very late PlO and the early ETL promoters of strong and medium strength, respectively, were designed to modulate the quantities of P-galactosidase indicator produced in insect cells. The light-emitting reaction in bacteria involves the oxidation of reduced riboflavin phosphate (FMNHJ and a long-chain fatty aldehyde with the emission of blue-green light (FMNH, + RCHO + O2 + FMN + HZ0 + RCOOH + light [490 nm]). The catalytic activity of bacterial luciferase can readily be quantitated in coupled enzyme assays that utilize glucose6-phosphate dehydrogenase and NAD(P)H:FMN oxidoreductase to generate FMNH2. In this chapter, we describe how to produce recombinant baculoviruses with either of two vectors, pPl0 or pETL. A similar vector has recently
Rapid
Screening
163
Vectors
been described by Zuidema et al. (25), except that the heat-shock promoter was used to drive the P-galactosidase gene. A modified version of our vectors designed to produce secreted recombinant proteins using the secretory signal peptides from honey bee melittin or alkaline phosphatase are currently marketed by Stratagene (La Jolla, CA) and are called pMBac or pPBac (26). The potential of bacterial luciferase as a reporter gene in cultured insect cells and larvae is also demonstrated. Levels of foreign gene expression from thesevectors is very high, and the methodology for assaying bacterial luciferase is described. 2. Materials of Viral and Transfer
2.1. Cotransfection Vector DNAS 1. 25-cm2 Tissue-culture flasks. 2. Spudoptera frugiperda insect cells (39). 3. Grace’s insect media supplemented with lactalbumin hydrolysate, TC yeastolate, 10% fetal bovine serum, 50 pg/rnL of gentamicin sulfate, and 2.5 pg/mL amphotericin B (fungizone). 4. Sterile 1.5~mL Eppendorf tubes. 5. 1X TE: 10 mM Tris-HCI pH 8.0, and 1 mM EDTA. 6. Linearized Autographa califomica nuclear polyhedrosis virus (AcNPV) DNA in 1X TE, supplied by Invitrogen. 7. Transfer vector DNA in 1X TE: pETL (also known asBlueBac II) or pP10. Supplied by Invitrogen (San Diego, CA) or C. Richardson. 8. Sterile Pasteur pipets. 9. Sterile 15-mL polypropylene tubes. 10. Transfection buffer: 25 mM HEPES, pH 7.1, 140 mMNaC1, and 125 n&f CaCI,. Autoclave and store at 4°C. 11. PBS-EGTA: 140 n04 NaCl, 27 mM KCl, 8 mJ4 Na2HP04, 1.5 rni$4 KH2P04, pH 7.3, and 100 mM EGTA. Store at room temperature. 12. Z buffer: 60 mM Na2HP04.7, 40 mM Na2H2P04, 10 mM KCl, 1 miW MgS04.7, and 50 mM P-mercaptoethanol. Store at room temperature. 13. 0.1% SDS. 14. Chloroform. 15. 8 mg/mL PNPG (p-nitropehnyl-P-o-galactopyranoside) in water (light-sensitive). Store at -20°C.
2.2. Plaque
Assay
1. Medium harvested from cotransfection. 2. 15-n& polypropylene tubes. 3. loo-mm tissue-culture dishes.
164
LalumiBre
and Richardson
frugiperda insect cells (Sf9). 5. Grace’s insect media as in Section 2.1. 6. 5% SeaPlaque agarose in water (supplied by PMC Bioproducts, Rockland, ME). Autoclave and store at 4OC. 7. 50 mg/mL Bluo-gal in N,N-dimethylformamide (light-sensitive). Store at -20°C in a glass container. 2.3. Plaque Purification of Recombinant Virus 1. Sterile Pasteur pipets. 2. 24-well plates. 3. Spodoptera frugiperda insect cells (Sf9). 4. Grace’s insect media as in Section 2.1. 5. 5% SeaPlaque agarose (supplied by PMC Bioproducts, Rockland, ME) and 50 mg/mL Bluo-gal. 2.4. Identifying Recombinant Virus by Dot Hybridization 1. 24-well plates. 2. Spodoptera frugiperda insect cells (Sf9). 3. Grace’s insect media as in Section 2.1. 4. Sterile 1.5~mL screw-cap tubes. 5. 0.5N NaOH. 6. 1OM NH4 Acetate. 7. 1M NH4 Acetate, and 0.02M NaOH. 8. 4 X SSC: 600 mM NaCl and 60 m&I NaCitrate, pH 7.0. 9. Nitrocellulose membranes. 10. Dot-blot apparatus. 11. Vacuum oven at 80°C. 2.6. Amplification of Recombinant Virus 1, 25-cm2 and 75-cm2 tissue culture flask. 2. Spodoptera frugiperda insect cells (Sf9). 3. Grace’s insect media as in Section 2.1. 2.6. Analysis of Recombinant Proteins by SDS Polyacrylamide Gel Electrophoresis 1. 24-well plates. 2. Spodopterafrugiperda insect cells (Sf9). 3. Grace’s insect media as in Section 2.1. 4. 1X PBS: 140 r&I NaCl, 27 mM KCI, 8 mM Na2HP04, and 1.5 mM KH2P04, pH 7.3. 5. 2X Sample buffer: 100 mII4 DTI’, 2% SDS, 80 mA4Tris-HCl pH 6.8, 10% glycerol, 0.0012% and bromophenol blue. 4. Spodoptera
Rapid 6. 7. 8. 9. 10. 11.
Screening
Vectors
165
Sterile 1.5~mL Eppendorf tubes. l-n& syringes and 26-gage needles. Vertical electrophoresis apparatus and gel solutions. Coomassie blue: 0.3% Coomassie blue, 40% methanol, and 10% acetic acid. Destain solution: 7.5% acetic acid and 5% methanol. LKB Ultrascan XL lase densitometer.
2.7. Assay of Bacterial 1, 2. 3. 4. 5. 6. 7.
8. 9.
Luciferase
Activity
n-Decanal substrate; Triton-X 100; P-mercaptoethanol. Cofactors FMN and NAD (Boeringer Mannheim, Indianapolis, IN). Glucose-6-phosphate (potassium salt). Glucose-6-phosphate dehydrogenase (Leuconostoc mesenteroides) supplied by Boeringer Mannheim. FMN:NADH oxidoreductase (Photobacterium [Vibriolfischeri) supplied by Boeringer Mannheim (see Note 1). 0.5M potassium phosphate buffer (pH 7); 10% (w/v) bovine serum albumin; 10% (w/v) amido black. Coupled enzyme solution: 50 mM potassium phosphate, pH 7,0.2% (w/v) bovine serum albumin, 5 mM P-mercaptoethanol, 0.002% decanal, 5 @4 FMN, 1 mM NAD, 20 mM glucose-6-phosphate, 0.1 U glucose-6-phosphate dehydrogenase, and 0.0015 U NAD(P)H:FMN oxidoreductase. 96-well microtiter plates; 60-mm plastic Petri plates. Kodak X-OMAT AR X-ray film; TMAX P3200 high-speed 35 mm blackand-white film. 3. Methods
3.1. Cotransfection of Viral and Transfer Vector DNAS An EcoRV fragment (3 kb in length) that contained the fused lux Al lux B genes of bacterial luceferase (24) was inserted at the NheI site of pETL and pPl0 by blunt-end ligation. The restriction maps of these expression vectors are presented in Fig. 1. The complete sequences of
these vectors are available on request, and pETL is sold by Invitrogen (San Diego, CA) under the trade name of BlueBac II. Expression plas-
mids and linearized viral DNA are cotransfected into St-9 cells using either calcium phosphate or lipofectin-mediated transfections. Generally, the lipofectin method is 5-10 times more efficient in introducing DNA into the insect cells. Following the incubation period for transfection, one can perform a PNPG @-nitrophenylgalactoside) assay on the cells to see if P-galactosidase is produced. The intensity of the product produced by j3-galactosidase can be a crude indicator of the efficiency of transfection.
/ ECORV4!W PV” ”
49.~9 PVUII a13
Fig. 1. Restriction map of baculovirus expression vectors, pPl0 and pETL, containing the P-galactosidase gene for purposes of screening recombinant virus. Restriction enzyme sites are indicated as the number of nucleotides distant from the Bgm site. The SV40 polyadenylation sites (SV40 PA), the P-galactosidase gene (ZucZ), the ETL promoter (ETL), the polyhedrin promoter (PH), the p10 promoter (PlO), and the bacterial shuttle vector (pSP72) are also shown.
Rapid
1. 2. 3.
4. 5. 6.
1, 2.
3. 4.
5. 6. 7.
Screening
Vectors
167
3.1.1. Calcium Phosphate Transfection Method Seed 2 x lo6 insect cells into a 25-cm2 tissue-culture flask, and leave at 28°C for 1 h to attach. Into sterile 1.5~mL Eppendorf tube, add 1 pg of AcNPV DNA and 2 pg of recombinant transfer vector (see Note 2). Add 0.75 mL of transfection buffer, and mix by pipeting up and down. Remove all of the medium from the flask, and replace with 0.75 mL fresh medium. Slowly add the 0.75 mL of DNA solution directly onto the cell monolayer dropwise using a Pasteur pipet. The DNA will form a calcium phosphate DNA precipitate in situ becauseof the phosphatesin the medium. Incubate at 28OCfor 4-8 h, then remove the transfection solution, replace with 10 mL fresh medium, and continue the incubation at 28OC for 2 d. At the end of this period, collect the cell media. The supematant is stored at 4OCin a sterile 15-mL polypropylene tube until it is used for plaque assays. Replace the media with 10 mL of Grace’s media containing 10% fetal calf serum, observe the cells over 5 d, and note presence of cytopathic effect (see Note 3). Alternatively, the cells can be collected and assayed for B-galactosidase activity as described below. 3.1.2. Liposome-Mediated Transfection of Sp Cells Seed 2 x lo6 Sf9 insect cells into a 25-cm2 tissue-culture flask and leave at 28OC for 1 h so that they may attach to the plastic. For each transfection, prepare the following transfection mix: Pipet 1 mL of Grace’s media (without fetal calf serum) into a sterile Eppendorf tube containing l-2 pg of linear AcNPV (baculovirus) DNA and 3-6 pg of expression vector DNA (see Note 2). Add 20 j.tL of either Cationic Liposome Solution (Invitrogen) or Lipofectin Reagent (GIBCO BRL) to the DNA/Grace’s solution. Vortex vigorously. Incubate at room temperature for 15 min. During this incubation, remove the media from the 25-cm2 tissue-culture flask and replace it with 2 mL of Grace’s media without calf serum. After 10 min, remove this washing media and add 1 mL transfection mix (prepared above) dropwise to the cell monolayer with a sterile pipet. Incubate the tissue-culture flask for 4 h at room temperature with occasional rocking (once every 15 min) or by gentle rocking on a mechanical agitator. Following this incubation, add an additional 1 mL of complete Grace’s containing 10% fetal calf serum and incubate the transfected cells at 28OC for 2 d. After 2 d, the media from the transfected cells is harvested and stored at 4°C until required for plaque purification. Replace the media taken from
168
Lalumikre
and Richardson
the tissue-culture flask with 3 mL of Grace’s media containing 10% fetal calf serum, and observe them for 5 d, and note cytopathic effects resulting from viral infection (see Note 3), or scrape the cells and assay them for P-galactosidase as described below. These tests can confirm a successful transfection.
3.1.3. Verification of DNA Uptake with PNPG Assay for /3-Galactosidase 1. Wash the cells twice with PBS-EGTA. Transfer to Eppendorf tubes. 2. Add to the cells 0.5 mL of 2 buffer, 2 drops of 0.1% SDS, and 2 drops of chloroform. Vortex and incubate at 28°C for 5 min with the caps off. 3. Add 100 l.tL of PNPG, mix, and incubate at 28OC until a yellow color appears (see Note 4).
3.2. Plaque
Assay
1. Seed 100~mm tissue-culture dishes with 7.5 x lo6 insect cells, and incubate at 28°C for 1 h to allow the cells to attach. Seed enough for 2 plates/ dilution (see Note 5). 2. Prepare a series of lo-fold dilutions in Grace’s insect media using the medium harvested from the cotransfections. An appropriate range is 10-l-10-4. 3. Remove all of the culture medium from the dishes by aspiration with a Pasteur pipet, and gently pour 5 mL of the diluted virus onto the cells on the side of the dish. Incubate for 1 h at 28°C to allow absorption of the virus. 4. During this incubation, melt the required volume of 5% SeaPlaque agarose (see Note 6), and keep at 45OC.Make a 1% solution by diluting 1:4 with Grace’s medium (containing 10% fetal calf serum) prewarmed at 37”C, and add 100 pg/mL of Bluo-gal (see Note 7). 5. Remove the diluted virus inoculum from the cells by aspiration with a Pasteur pipet, and overlay with 10 ml/plate of 1% SeaPlaque agarose solution. Work quickly since the agarose solidifies after about 7 min. Allow the plates to set for 20 min at room temperature. Incubate at 28°C for 31t d. Blue plaques should be visible after this time as shown in Fig. 2.
3.3. Plaque
Purification
The plaques containing Bluo-gal.
of
Recombinant
Virus
the foreign gene stain blue in presence of
1. Using a sterile Pasteur pipet, pick a plug of agarose from directly over the blue plaque, and place into 1 mL of Grace’s medium in a 24-well plate (see Note 8). Virus is allowed to elute overnight at 28OC.
Rapid Screening Vectors
169
Fig. 2. Formation of recombinant baculovirus plaques in the presence of Bluo-gal. Virus expressing bacterial luciferase was generated using either the pETL vector (pETL-Luc) or pPl0 vector (pPlO-Luc). Assays were performed on lo-6 dilutions of stock virus prepared after amplification of two different plaques in Sf9 cells. 2. Perform another plaque assay, at lo-, loo-, and lOOO-fold dilutions using the media containing the eluted virus. Select for blue plaques again. 3. Repeat until the plaques are 100% occlusion body negative. Usually two rounds of plaque assaysare sufficient. 3.4. Identifying Recombinant Viruses by Dot Hybridization
To ascertain that the recombinant virus actually contains the luciferase gene, DNA from Sf9 insect cells that were infected with baculovirus is subjected to dot-blot hybridization analysis. Usually this is performed after the first round of plaque assays. 1. Seed a 24-well plate with 3 x lo5 cells/well, and leave at 28°C for 1 h to attach. 2. After virus has eluted overnight at 28OC,transfer 300 p,L of each eluted virus onto cells. Incubate at 28OCfor 3-4 d. 3. At the end of this period, transfer media (virus stock) to a sterile 1.5-n& Eppendorf tube and freeze at -20°C. 4. Lyse the cells in the wells by adding 200 pL of OSN NaOH and mixing. 5. Neutralize the solution by adding 20 pL of 1OM NH, acetate and mixing. 6. Spot the lysates onto nitrocellulose membranes in a dot-blot apparatus.
I70
Lalumibre
and Richardson
a. Cut a piece of nitrocellulose and Whatman 3MM paper to size, wet the nitrocellulose in hot water, and equilibrate them both in a solution containing 1M NH4 acetate and 0.02M NaOH. b. Set up the dot blot apparatus. The nitrocellulose should be supported from below by the Whatman paper. c. Apply 50-100 pL of cell lysate to the manifold. Apply vacuum pump. Wait up to 15 min. Remove excessliquid with a Pasteurpipet (seeNote 9). d. Remove the nitrocellulose and wash for 2 min in 4X SSC. 7. Air-dry the filter and bake for 2 h at 80°C under vacuum. 8. Hybridize with a probe specific for your gene. 9. Identify positive virus samples that contain your gene. Purify by two rounds of plaque assays.Repeat dot hybridization after the last purification, or check for presence of the foreign gene in the recombinant virus using PCR techniques (see Chapter 8). 3.5. Amplification Once an occlusion-negative
of Recombinant
Virus plaque has been found and plaque puri-
fied, it is necessary to obtain culture medium containing a high titer of the recombinant virus for further experiments. 1. Seed 3 x lo6 insect cells into a 25-cm2 tissue-culture flask, and leave at 28OCfor 1 h to attach. 2. Infect the flask with 500 pL of the virus stock. Incubate at 28°C for 3-4 d. Collect the media and store at 4°C. 3. Of this, 2 mL are used to infect a 75-cm2 tissue-culture flask seeded with 9 x lo6 insect cells. Incubate at 28°C for 3-4 d. Collect the media. Of this, 1 mL is frozen at -70°C for long-term storage; the remainder is kept at 4OC. 3.6. Analysis of Recombinant Proteins by SDS Polyacrylamide Gel Electrophoresis 1. Seed a 24-well microtiter plate with 3 x lo5 insect cells/well. Leave at 28OCfor 1 h to attach. Divide plate in six sections of four wells. Infect each well with 200 pL of amplified recombinant virus except for the first four (first section) (see Note 10). 2. Harvest cells at 0, 12, 24, 48, 72, and 96 h postinfection (1 section/timepoint). Wash cells twice with 1X PBS. Add 50 pL of electrophoresis sample buffer/well, wait 3 min, and pool the four wells into an Eppendorf tube. 3. Shear the DNA 10 times by passing sample through a 26-gage needle attach to a 1-mL syringe. Boil samples for 5 min. An aliquot of the denatured proteins is loaded on an SDS polyacrylamide gel, and stained with Coomassie blue using routine techniques (27). A typical stained gel showing the production of recombinant bacterial luciferase is shown in Pig. 3.
Rapid Screening Vectors
171
B-gal luc
pETL-luc 0
24
49
72
96
0
24
49
72
96
46-
Fig. 3. Total cellular proteins from Sf9 cells infected with wild-type (wt) and recombinant baculovirus (pPlO-luc; pETL-luc) were separated by SDS polyacrylamide gel electrophoresis and stained with Coomassie blue. Cells were harvested and lysed in sample buffer at 0,24,48,72, and 96 h postinfection. /3-galactosidase (P-gal), recombinant luciferase (luc), and wild-type virus polyhedrin (ph) are indicated by arrows. Molecular-weight standards (in kDa) are shown at the left of each gel. 4. Record stained proteins on an LKB Ultrascan XL laser densitometer. The amount of proteins produced is crudely quantitated as a percentage of the total protein applied to the gel. Protein peaks can be cut from the recording
172
Lalumibre
and Richardson
paper, weighed, and compared to the total weight of the paper beneath the recording profile. The amount of recombinant protein is thus a fraction of the amount of total protein applied to the gel.
3.7. Assay of Bacterial
Luciferase
Activity
Bacterial luciferase made an excellent reporter gene with which to assess the utility and efficiency of our vectors (6). The enzyme was cheaper to assay than its counterpart found in fireflies, and a coupled enzyme solution that generated FMNH2 allowed it to be assayed over longer time periods. The recombinant enzyme can be assayed from either Sf9 insect cells or infected cabbage loopers using film detection techniques.
1.
2.
3.
4.
3.7.1. Assay of Luciferase porn Cell Extracts Using X-Ray Film Detection Infected Sf9 cells are harvested at 0, 12,24,48,72, and 96 h post infection. Some of the cells (approx 4 x 105) are suspended in 500 pL of PBS and lysed by making the solution 1% (v/v) T&on-X 100 at room temperature for 5 min. The cell lysates are maintained on ice prior to assaying their activity. Microtiter plate (96-well) assays are performed with a coupled enzyme solution in a total volume of 50 pL containing 50 mM potassium phosphate (pH 7), 0.2% (w/v) bovine serum albumin, 5 mM P-mercaptoethanol, 0.002% n-decanal, 5 pM FMN, 1 m&I NAD, 20 mM glucose-6-phosphate, 0.1 U of glucosed-phosphate dehydrogenase, 0.0015 U of NAD(P)H:FMN oxidoreductase, and 5 pL of extract containing luciferase. Amido black solution (10% w/v) was placed outside the individual plastic wells to minimize light leakage and reflection into adjacent assays. Microtiter plates were exposed to X-ray film in complete darkness (see Note 11) for 2 h prior to development. Levels of luciferase could be quantitated by comparing different dilutions to standard amounts of enzyme isolated from Vibrio harweyi (24) as shown in Fig. 4.
3.7.2. Assay of Luciferase from Whole Infected Caterpillars 1. Trichoplusia ni larvae, commonly known as cabbage loopers (see Chapter 15), are infected by feeding them intracellular virions isolated from lysates of cells infected with recombinant baculovirus ( 10”lo9 plaque-forming units) mixed with a small quantity of insect diet, Lysates are prepared by hypotonic lysis of 10’ cells in sterile distilled water. Virus and cell membranes are sedimented at 100,OOOgfor 60 min and the pellet is resuspended in 1 mL of sterile water. After 34 d, the caterpillars were frozen, placed in
Rapid Screening Vectors
173
Fig. 4. Microtiter assay of bacterial luciferase produced in Sf9 insect cells infected with two different luciferase recombinant baculoviruses (pPlO-Luc and pETL-Luc). Cells were harvested and lysed at 0, 24, 48, 72, and 96 h postinfection. Cell lysates containing recombinant enzyme were placed in chambers of a 96-well microtiter plate containing 50 pL of the coupled enzyme assaymix and exposed to X-ray film for 5 min. 10 mL of luciferase assay cocktail (1% w/v Triton X-100, 50 rnit4 potassium phosphate buffer, pH 7,0.2% w/v, bovine serum albumin, 5 mM P-mercaptoethanol, 0.002% decanal, 5 p&f FMN, 1 nGI4 NAD, 20 mM glucose-6-phosphate, 0.1 U glucose-6-phosphate dehydrogenase, 0.0015 U of NAD(P)H:FMN oxidoreductase). 2. The caterpillars, in plastic Petris (60~mm diameter), were placed on top of X-ray film and were exposed in a light-tight box for l-4 h. Alternatively, luminescent caterpillars were photographed directly in a light tight box using TMAX P3200 black-and-white film (Eastman Kodak, Rochester NY) with a time exposure of 30 min (see Fig. 5). 4. Notes 1. We screened a number of oxidoreductases and diaphorases, and found only the FMN:NADH oxidoreductase from Photobacterium (Vibrio) jishceri supplied by Boeringer Mannheim gave satisfactory and consistent results. 2. Use only CsCl-purified expression vector plasmid, since DNA from minipreparations yields lower transfection efficiencies. 3. You should check for cytopathic effect by looking at the cells under the microscope. These effects can include the accumulation of occlusion
174
LalumiBre
and Richardson
Fig. 5. Assay of luciferase in cabbage loopers (Trichoplusia ni) in the absence (A) and presence (B) of 1% (w/v) Triton X-100. Larvae were infected with intracellular recombinant virus mixed with diet, and incubations were allowed to proceed for up to 4 d. Caterpillars were immersed in a solution that contained the substrate decanal and a coupled enzyme mix that generated the cofactor FMNH,. bodies from the small portion of linear AcNPV genomic DNA that was not completely linearized or cells that are swollen two to three times their normal size. If you do not see any effect, the transfection may have to be redone. However, the PNPG assay is more accurate, and a positive color development indicates that cells are producing P-galactosidase-we recommend this test be done at 4-5 d posttransfection.
Rapid Screening Vectors
175
4. One should obtain an intense yellow color in c20 min if the transfection is satisfactory. 5. It is important that the cells are 75-80% confluent in order to get good plaque formation. 6. Sf9 insect cells are extremely sensitive to impurities in agarose supplied by a number of different distributors. SeaPlaque agarose is ultrapurified and consistently gives us good results. 7. Bluo-gal produces viral plaques that are much darker in color that those made with X-gal. We find Bluo-gal makes picking recombinant plaques much easier. 8. Pick recombinant plaques immediately after the blue color appears, since the contaminating wild-type virus tends to overgrow the recombinant virus rapidly. These plaques are also monitored by phase contrast microscopy for absenceof occlusion bodies. Distinct plaques are visible on assaysfrom the higher dilutions of virus, but are often not evident from virus at higher concentrations since the viral plaques tend to fuse and produce a pale blue color throughout the agarose. 9. Often the nitrocellulose filter plugs up during vacuum filtration. However, even if half of the lysate passesthrough, this should be sufficient to give a signal, Excess lysate should be removed from the well to prevent leakage over the filter and subsequent contamination of other sample spots. 10. As a negative control, one should perform a parallel time-course with wildtype virus. 11. A photographic paper box designed to hold 250 sheets of 21.6 x 27.9 cm paper makes an excellent light-tight chamber.
References 1. Luckow, V. A. and Summers, M. D. (1988) Trends in the development of baculovirus expressionvectors. Biotechnology 6,47-55. 2. King, L. A. andPossee,R. D. (1992) The Baculovirus Expession System: A Luboratory Manual. Chapman& Hall, UK. 3. O’Reilly, D. R., Miller, L., and Luckow, V. A. (1992) Baculovirus Expression Vectors: A Laboratory Guide. W. H. Freeman,New York. 4. Vialard, J., Lalumi&re, M., Vernet, T., Briedis, D., Alkhatib, G., Henning, D., Levin, D., and Richardson,C. D. (1990) Synthesisof the membranefusion and hemagglutinin proteins of measlesvirus, using a novel baculovirus vector containing the P-galactosidasegene.J. Viral. 64,37-50. 5. Richardson,C., Attia, J., Dunn, R., Gupta, S.,O’Connor, M., Semeniuk,D., Tam, J., Hamel, M., Lambert, G., Dennis, M., Jacobs,F., Martin, L, Iorio, C., and Vialard, J. (1992) Engineering glycoproteins for secretionusing the baculovirus expressionsystem,in Baculovirus and Recombinant Protein Production Processes (Vlak, J. M., Schlaeger,E.-J., and Bernard, A. R., eds.)Roche, Basel,pp. 67-74.
176
LalumiBre
and Richardson
6. Richardson, C. D., Banville, M., Lalumiere, M., Vialard, J., and Meighen, E. A. (1992) Bacterial luciferase produced with rapid screening baculovirus vectors is a sensitive reporter for infectron of insect cells and larvae. Zntervirology 34,213-227. 7. Chakrabarti, S., Brechling, K., and Moss, B. (1985) Vaccinia virus expression vector: coexpression of j3-galactosidase provides visual screening of recombinant virus plaques. Mol. Cell. Biol. 5,3403-3409. 8. Gould, S. J. and Subramani, S. (1988) Firefly luciferase as a tool in molecular and cell biology. Anal. Biochem. 175,5-13. 9. Brassier, A. R., Tate, J. E., and Habener, J. F. (1989) Optimized use of the firefly luciferase assayas a reporter gene in mammalian cell lines. Biotechniques 7,1116-l 122. 10. Schneider, M., Ow, D W., and Howell, S. H. (1990) The in vivo pattern of firefly luciferase expression in transgenic plants. Plant Mol. Biol. 14,935-947. 11. Rodriguez, D., Rodriguez, J. R., Rodriguez, J. F., Trauber, D., and Esteban, M. (1989) Highly attenuated vaccinia virus mutants for the generation of safe recombinant viruses. Proc. Natl. Acad. Sci. USA 86, 1287-1291. 12. Hasnain, S E., and Nakhai, B. (1990) Expression of the gene encoding firefly luciferase in insect cells using a baculovirus vector. Gene 91, 135-138. 13. Karp, M., Akerman, K., Lindqvist, C., Kuuisisto, A., Saviranta, P., and Oker-Blom, C. (1992) A sensitive model system for in vivo monitoring of baculovirus gene expression in single infected insect cells. Biotechnology 10, 565-569. 14. Mori, H., Nakazawa, H., Sherai, N., Shibata, N., Sumiota, M., and Matsubara, F. (1992) Foreign gene expression by a baculovirus vector with an expanded host range. J. Gen. Virol. 73,1877-1880. 15. Jha, P. K., Pal, R., Nakhai, B., Sridhar, P., and Hasnain, S. E. (1992) Simultaneous synthesis of enzymatically active luciferase and biologically active p subunit of human chorionic gonadotropin in caterpillars infected with a recombinant baculovirus. FEBS Lett. 310,148-152. 16 Kopylova-Sviradova, T. N., Krauzova, V. I., Timiryasova, T. M , Gorelova, T. V., Shuppe, N. G., and Fodor, I. (1992) Transient expression in a baculovirus system using firefly luciferase as a reporter. Virus Genes 6,379-386. 17. Oker-Blom, C., Jansson, C., Karp, M., Lindqvist, C., Savola, J. M., Vlak, J., and Akerman, K. (1993) Functional analysis of the human alpha 2C-C4 adrenergic receptor in insect cells expressed by a luciferase-based baculovirus vector. Biochim. Biophys. Acta 1176,269-275.
18. Oker-Blom, C., Suomalainen, A. M., Akerman, K., Qi, Z., Lindqvist, C., Kuusisto, A., and Karp, M. (1993) A baculovirus-expressed fusion protein containing the antibody-binding domain of protein A and insect luciferase. Biotechniques 14,808-809. 19. Lahde, M., Raunio, H., Pelkonen, O., Karp, M., and Oker-Blom, C. (1993) Expression of human placental cytochrome P450 aromatase (CYP19) cDNA in insect cells using a luciferase based baculovirus vector, Biochem. Biophys. Res. Corn. 197,1511-1517. 20. Hasnain, S. E., Nakhai, B., Ehtesham, N. Z., Sridar, P., Ranjan, A., Talwar, F. P., and Jha, P. K. (1994) Beta-subunit of human chorionic gonadotropin hormone and firefly luciferase simultaneously synthesrzed in insect cells using a recombinant baculovirus are differentially expressed and transported. DNA Cell Biol. 13,275-282
Rapid Screening Vectors
177
21. Lindqvist, C., Karp, M., Akerman, K., and Oker-Blom, C. (1994) Flow cytometric analysis of bioluminescence emitted by recombinant baculovirus-infected cells. Cytometry 15,207-212. 22. Meighen, E. A. (1991) Molecular biology of bacterial bioluminescense. Microbial. Rev. 55,123-142. 23. Meighen, E. A. (1988) Enzymes and genes from the lux operons of bioluminescent bacteria. Ann. Rev. Microbial. 42, 151-176. 24. Boylan, M., Pelletier, J., and Meighen, E. A. (1989) Fused bacterial luciferase subunits catalyze light emission in eucaryotes and prokaryotes. J. Biol. Chem. 264, 1915-1918. 25. Zuidema, D., Schouten, A., Usmany, M., Maule, D. J., Belsham, G. J., Rossien, J., Klinge-Roode, E. C., Van Lent, J. W. M., and Vlak, J. M. (1990) Expression of cauliflower mosaic virus gene in insect cells using a novel polyhedrin-based baculovirus expression vector. J. Gen. Virol. 71,2201-2209. 26. Mroczkowski, B. S., Huvar, A., Lernhardt, W., Misono, K., Nielson, K., and Scott, B. (1994) Secretion of thermostable DNA polymerase using a novel baculovirus vector. J. Biol. Chem. 269, 13522-13528. 27. Gallagher, S. R. and Smith, J. A. (1989) One-dimensional gel electrophoresis of proteins, in Current Protocols in Molecular Biology (Ausabel, F. M., Brent, R., Kingston, R. E., Moore, D. D., Seidman, J. G., Smith, J. A., and Struhl, K., eds.), Wiley-Interscience, New York, pp. 10. 2. l-10.2.21.
CHAPTER10 Fast Isolation of Recombinant Baculovirus by Antibody Screening Frank
Grosse and Andreas
Manns
1. Introduction A crucial step in the baculovirus expression system technique is the selection of recombinant viruses. In most cases positive clones, i.e., recombinant viruses, are detected via the presenceof foreign DNA, or by exploiting the phenotypic differences between wild-type viral plaques, which are occlusion body positive, and recombinant plaques, which have an occlusion body negative phenotype (14). Once a positive clone has been identified, the recombinant virus must be purified free from any wild-type precursor, A restriction analysis of isolated viral DNA might be necessary to determine whether the foreign gene has been correctly integrated. Having proven the presence of the gene, cells are infected with several independent recombinant viruses, in order to determine which of those gives the best expression of the protein in insect cells. Here we describe an alternative procedure for the selection of recombinant viruses, that is based on a limiting dilution of the virus and the direct detection of the foreign protein in baculovirus-infected Spodoptera frugiperda (Sf9) cells (5). The newly produced foreign protein is detected by a dot-blot screening procedure with antibody probes after cell lysis. Since the E. coli transfer plasmid is not able to direct the synthesis of the foreign protein in insect cells, positive signals must be the result of an infection with successfully recombined baculovirus. Positive recomFrom: Methods in Molecular Biology, Vol. 39: Baculovirus Express/on Protocols Edited by: C. D. Richardson (0 1995 Humana Press Inc., Totowa, NJ
179
180
Grosse and Manns
binants are selected from the highest dilution that gives an unambiguous signal, and again diluted until one or less than one infection/well of insect cells can occur. Recombinant baculoviruses that direct the synthesis of foreign protein and that are free of wild-type virus are subsequently amplified and used as vectors to induce the expression of the protein of interest. The direct detection of the foreign protein simplifies the screening procedure considerably. The detection and isolation of recombinant
virus can be performed within 2-3 wk. Usually, two to three cycles of screening are sufficient to isolate pure recombinants. A further advantage of this technique is the selection of recombinants that induce a high level expression of foreign protein. The only prerequisite for using this technique is the availability of a sensitive antibody against the foreign protein.
2. Materials 1. The insect cell line Sf9 was obtained from G. Fertig (University Darmstadt, Germany). Alternatively, it may be obtained from the American Type Culture Collection (12301 Parklawn Drive, Rockville, MD 20852; ATCC # CRL1711). 2. Cell stocks are maintained at 27OC in TC-100 medium (GIBCO/BRL, Eggenstein, Germany) supplemented with heat-inactivated 5% fetal calf serum (Conco, Wiesbaden, Germany) as log-phase monolayer cultures in plastic flasks of 25cm2 surface area (Nunc, Roskilde, Denmark) and subcultured once a week. For DNA transfection and virus infection, the cells are subcultured a day before to ensure cell growth in the logarithmic phase, as previously recommended (4). Antibiotics, 2.5 p,g/mL amphotericin B and 50 pg/rnL gentamycin (Boehringer, Mannheim, Germany), are used for cells growing in microtiter plates, but can be omitted from cell culture stocks. 3. Microtiter plates with 48 wells can be obtained from Costar, Cambridge, MA. Silent Monitor@ 96-well plates bottomed with an Immunodyne membrane as well as the Silent Monitor Vacuum Manifold can be purchased from Pall GmbH, Dreieich, Germany. 4. Prepare the following buffers: PBS buffer: 10 mM potassium phosphate, pH 7.8, and 130 mh4 NaCl. PBS -I&en: 0.1% (v/v) Tween 20 (Sigma, Deisenhofen, Germany) in PBS. Lysis buffer: 10 mil4 potassium phosphate, pH 7.8, 350 mil4 NaC.1,and 0.5% Lubrol PX (Sigma). 5. Antibodies against HIV-l reverse transcriptase were raised, characterized, and purified by standard methods (6). Blotting-grade horseradish peroxidase-conjugated goat antirabbit antibodies are from Bio-Rad Laboratories
Screening
of Recombinant
Baculovirus
181
(Richmond, CA). Dissolve the primary antibody in 50 mL PBS containing 10% calf serum at a concentration of about 5 pg/mL. The secondary antibody is diluted according to the instructions of the manufacturer in PBS containing 10% calf serum.
3. Methods The method to select for recombinant AcNPV is based on the detection of foreign proteins in cellular lysates of baculovirus-infected insect cells by antibody screening. Lysis of infected cells is induced by 350 mM NaCl containing 0.5% of the detergent Lubrol. The lysed cells are sucked through an Immunodyne membrane, and the membrane-bound proteins are subsequently probed in dot-blot-like procedure with an antibody directed against the virus-expressed foreign protein and a secondary peroxidase-conjugated antibody directed against the first antibody. The recombinant virus is purified to individual protein-expressing clones by repeated serial dilutions. 1. The preparation of the transplacement plasmid and Autograph californica nuclear polyhedrosis virus (AcNPV) wild-type DNA is performed as described elsewhere (4). 2. Five days after transfection, the extracellular virus population is harvested and diluted in steps of 10 into 1 mL TC 100 medium, down to a final dilution of 10”. 3. 50 pL of the virus-containing medium from each dilution is pipeted into the wells of a 48-well microtiter plate (see Note 1). 4. About lo5 Sf9 cells in 400 pL TClOO culture medium supplemented with 5% fetal calf serum are added. 5. The plate is incubated for 7 d at 27°C to allow virus infection and growth. 6. One wk after infection, 100 pL of culture supernatant from each well are transferred to a new plate, which serves as masterplate. The masterplate is kept at 4°C for further usage. 7. The remaining cells are lysed directly by adding 200 pL lysis buffer and rocking for 10-15 min at room temperature. 8. The lysed cells are pipeted into the wells of a Silent Monitor 96-well plate sealed on the bottom with an Immunodyne membrane. 9. The plate is mounted onto a Silent Monitor Vacuum Manifold. 10. The solutions are slowly passed through the membrane by applying a water stream vacuum. 11. For the immunoprocessing, the membrane is carefully peeled from the plate as shown in Fig. 1, For peeling off the membrane, the use of a blunt-ended pair of tweezers might be helpful.
182
Grosse and Manns
Fig. 1. After sample loading, the membrane is peeled from the plate for further processing. 12. The membrane is washed briefly in PBS. 13. To block unspecific protein binding sites, the membrane is incubated for 1 h with 50-100 mL 10% heat-inactivated calf serum in PBS. 14. The membrane is again briefly washed with PBS. 15. Thereafter, the membrane is incubated with primary antibody by gently rocking for 1 h at room temperature. 16. Following three lo-min washes first with PBS, then PBS-Tween, and again with PBS, the secondary peroxidase-conjugated antibody is added to the filter. 17. After 1 h of gentle rocking at room temperature, the filter is washed as described before. 18. The substrate for protein staining is prepared immediately prior to use: 25 mg of 3,3’-diaminobenzidine tetrahydrochloride (DAB) are dissolved in 100 mL PBS. Three milliliters of 1% (w/v) CoCl, are added, and the solution is clarified by passing it through a folded filter. Finally the solution is mixed with 70 pL of 30% H202 and immediately added to the membrane. 19. Signals become visible after a few seconds or minutes. 20. When clearly visible signals are detectable, the reaction is stopped by washing the membrane with PBS. 21. Best signals for evaluation and documentation are obtained with wet membranes. 22. After documentation, the stained membrane is dried between several layers of Whatman 3MM cellulose sheets.
Screening of Recombinant 1
2
3
4
5
Baculovirus 6
7
8
183 9
10
11
12
Fig. 2. Primary immunoscreening for recombinant baculovirns. The figure shows the Immunodyne-membrane probed with polyclonal antibodies directed against HIV-l reverse transcriptase (HIV-RT) followed by peroxidase-conjugated antirabbit immunoglobulin. The initial dilutions of the viral stock ranged from 10v3to 1O4. Wells G1-4 and H1-4 contained uninfected cells. Wells G58 and H5-8 contained cells infected with wild-type AcNPV. Wells Hll and H12 contained 0.15 and 1.5 p.g purified HIV-RT, respectively (taken from ref. 5, with permission of Eaton Publishing). The culture media of cells that give the strongest signal at the highest initial virus dilution (wells E 1 and F12 of Fig. 2) are combined and used for a second screening, as shown in Fig. 3. For the second screening, combined positive virus stocks of the first screening procedure are again serially diluted down to 1O-8-fold. This ensures that a virus concentration of 1 or ~1 plaque forming unit/well is obtained, at least in some of the wells. At dilutions of 10m2to 10-6, the wells still contain wild-type virus, because cells infected with these dilutions are producing polyhedrin occlusion bodies. In contrast, many of the 10m8dilutions do not contain wild-type virus, as judged from the absence of occlusion bodies, but they give a strong signal after cell lysis and dot-blot analysis with anti-HIV-RT antibodies (wells F3 and F6 of Fig. 3). Nevertheless, to be certain of the absence of wild-type virus at the lowest dilution, it is
184
Grosse and Manns 1
2
3
4
5
6
7
8
Fig. 3. Secondary immunoscreening for recombinant baculovirns. Viral stocks from wells El and F12 of Fig. 1 were combined and diluted in steps of ten. Fifty microliters of the virus dilution ranging from 1O-2to lo-8 were used to infect Sf9 cells in a 4%well plate. The presence or absence of occlusion bodies was determined by microscopic examination of the infected cells 120 h postinfection. Immunoscreening was performed as described in the text. Well A7 contained uninfected cells, and well A8 contained cells infected with wildtype AcNPV (taken from ref. 5, with permission of Eaton Publishing). recommended to inspect the infected cells visually and, if necessary, continue with a third cycle of dilutions (see Note 2).
4. Notes 1. To facilitate viral infection, it is recommended to keep the volume of virus and culture medium to a minimum, so that culture medium just covers the cells. For the preparation of viral stocks, a low multiplicity of infection (0. l-l) should be used. For high levels of protein production, a multiplicity of infection of 5-10 is recommended. We have also tried to grow St-9 cells in 96-well plastic plates in order to fit these to the 96-well Silent Monitor plates, without success,however. Probably the volume-to-surface ratio in plates containing more than 4%wells is suboptimal for the growth of St9 insect cells.
Screening of Recombinant
Baculovirus
185
2. The antibody screening technique has been used by us for the identification and isolation of recombinant baculoviruses that direct the overproduction of the HIV-l @gene and parts thereof. In most cases,clones were picked that induce a strong overproduction of the foreign protein, typically comprising 2040% of the total cellular protein, In some cases, e.g., when the protease domain of the @-gene has been deleted, the overproduced protein precipitates as insoluble aggregates, which are strongly reminiscent of inclusion bodies produced by some E. coEi expression vectors. In such cases, it might be useful to select a clone that does not give rise to such a strong overproduction, i.e., a clone with a weaker signal on the Immunodyne membrane.
References 1. Capone, J. (1989) Screening recombinant baculovirus plaques in situ with antibody probes. Gene. Anal. Technol. 6,62-66. 2. Fung, M.-C., Chiu, K. Y. M., Weber, T., Chang, T.-W., and Chang, N. T. (1988) Detection and purification of a recombinant human lymphotropic virus (HHV-6) in the baculovirus expression system by limiting dilution and DNA dot-blot hybridization. J. Virol. Methods 19,33-42. 3. Pen, J., Welling, G. W., and Welling-Wester, S. (1989) An efficient procedure for the isolation of recombinant baculovirus. Nucleic Acids Res. 17,45 l-45 1. 4. Summers, M. D. and Smith, G. E. (1987) A manual of methods for baculovirus vectors and insect cells culture procedures, Texas Agricultural Experiment Station, Bulletin No. 1555. 5. Manns, A. and Grosse, F. (1991) Fast isolation of recombinant baculovirus by antibody screening. BioTechniques 10, 154-158. 6. Harlow, E. and Lane, D. (1988) Antibodies, A Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY.
CHAPTER
11
Continuous Foreign Gene Expression in Transformed Lepidopteran Insect Cells Donald
L. Jarvis
and
Linda
A. Guarino
1. Introduction Protocols for the isolation of recombinant baculoviruses and their use for foreign gene expression in insect cells have been described in detail in the preceding chapters (also see refs. I-5). Generally, this approach involves the replacement of viral DNA sequences encoding the polyhedrin protein with foreign DNA sequencesencoding the protein of interest. The recombinant virus is plaque-isolated, and used to infect cultured insect cells or larvae. The protein of interest will be expressed during the very late phase of infection, when polyhedrin would normally be expressed by the wild-type virus. One major advantage of baculovirus expression vectors is that they can be used to produce exceedingly large amounts of a recombinant protein. This is due to the strength of the polyhedrin promoter, which can induce the synthesis of a very large pool of mRNA for translation into large amounts of protein. A second big advantage of the baculovirus expression system is that insect cells have most of the protein-processing capabilities of higher eukaryotic cells (reviewed in ref. 6). Thus, proteins produced in recombinant baculovirusinfected cells can undergo co- and posttranslational processing, resulting in an end product that is very similar, if not identical, to the native protein. The baculovirus expression vector system has been used to express over 300 different proteins in over 1000 different laboratories throughout the world (ref. 6a and M. D. Summers, personal communication). From:
Methods m Molecular Biology, Vol. 39. Baculovirus Expressron Protocols Edited
by C. D. Richardson
Q 1995
187
Humana
Press Inc., Totowa,
NJ
188
Jarvis and Guarino
This wealth of experience has led to the general observation that nuclear and cytoplasmic proteins are, indeed, expressed at impressive levels by recombinant baculoviruses, commonly around 200 mg/L of infected cells. However, proteins that enter the secretory pathway are often expressed at much lower levels, around 1 mg/L of infected cells (7,8). It must be emphasized that this is a generalization, not a conclusive statement based on a carefully controlled scientific study. Moreover, it must be recognized that some secretory pathway proteins have been expressed at high levels in the baculovirus system (9,lO). Nevertheless, it is fair to state that secretory pathway proteins are often produced at lower levels, as compared to other types of proteins, in the baculovirus expression system. Recombinant baculoviruses lack only the polyhedrin gene, which is nonessential for viral replication in cultured cells. Thus, they are fully equipped to carry out a lytic infection in vitro. If this lytic infection has an adverse effect on the function of the host cell secretory pathway, then this would explain why proteins that undergo processing in this pathway could be difficult to overproduce with the baculovirus expression system. In fact, there is some evidence to support the idea that the function of the host cell secretory pathway is impaired by baculovirus infection. Secretion of human tissue plasminogen activator is less efficient at later times of infection in recombinant baculovirus-infected insect cells (8). Similarly, processing of the influenza virus hemagglutinin protein is slower in baculovirus-infected cells, as compared to influenza virus-infected MDCK cells (II). The possibility that baculoviruses might functionally impair the host cell secretory pathway has prompted us to develop a new approach for baculovirus-mediated foreign gene expression (12). A schematic overview of this approach is shown in Fig. 1. Generally, it involves the construction of plasrnids that can express foreign proteins in the absence of viral infection. This was accomplished by using the transcriptional promoter from the IEl gene of Autogrupha californica multicapsid nuclear polyhedrosis virus (AcMNPV; 13,14). This promoter is transcriptionally active during the immediate early phase of infection and can induce gene expression in uninfected insect cells, as shown by both transient (13) and continuous expression assays (12). By contrast,
the polyhedrin promoter is only expressed during the very late phase of baculovirus infection, when the functional capabilities of the host cell
Gene Expression
in Transformed
Insect Cell Lines
189
Baculovirus-mediated foreign gene expression in stably-transformed insect cells -#
Hmc II(lnsen
dewed codmg sequence here)
Hint II
pIEl-encoded gene X
\ 1. COTRANSFECT:
Hmc II
PIE1 Neo
/ Spodoprerufrugiperda
ceils
2. SELECT Neomykin resistant colonies 2 weeks, 1 mg/ml G418
3. AMPLIFY & ASSAY: Foreign protein expresslon -DNA (Southern analysis) -RNA (Sl nuclease analysis) -Protein (RIP, western analysis)
Fig. 1. Method for stable transformation of lepidopteraninsect cells using lE1 expressionplasmids. Sf9 cells are cotransfectedwith IEl expressionplasmids encodingthe protein of interest and a selectablemarker (neomycin resistance). Transformants are selected in G418, amplified, and assayedfor the ability to expressthe unselectedgene. secretory pathway might be suboptimal. Expression plasmids that express the gene of interest in uninfected insect cells can be constructed by inserting the gene into a cloned copy of the AcMNPV IEl gene (I3,14), at a Hind site located downstream from the IEl promoter (Figs. 1 and 2). Insect cells are then cotransfected with a mixture of the expression plasmid encoding the gene of interest and PIE 1Neo, another expression plasmid that encodes a bacterial neomycin resistance gene (Fig. 1).
190
Jarvis and Guarino B~I AccKlaI NheI SmaI \ I! PVUII I lxmam
/, Hind-I \ ,lO~ NdeI
IEl translation 601-2344 IEl transcription 551-2377
Fig. 2. Restriction map of pAcIE1. Key restriction sites, transcriptional initiation and termination sites, and translational initiation and termination sites are indicated. The thm line with the arrow represents the IEl transcript, and the black, thick line represents the II31 protein. The plasmid contains an ampicillin resistancemarker and origin of replication, as representedby the gray, thick lines.
Subsequently, stably transformed clones capable of expressing the gene of interest constitutively, in the absenceof viral infection, can be isolated by using standard cotransfection and neomycin selection techniques developed in mammalian cell systems (Fig. 1; 15,16). The work on stable transformation of lepidopteran and other insect cells was recently reviewed (I 7). Specific protocols for the production, isolation, and screening of transformed lepidopteran insect cells are described in detail below.
Gene Expression
in Transformed
Insect Cell Lines
191
2. Materials 1. The Sf9 subclone of Spodopterufrugiperdu cells (IPLB-X21-AE; IS) is available from the American Type Culture Collection (1230 1 Parklawn Dr., Rockville, MD 20852). We routinely maintain Sf9 cells in TNM-FH medium supplemented with 10% (v/v) heat-inactivated fetal bovine serum (Hazelton Research Products, Inc., Lenexa, KS). We prepare both Grace’s medium and TNM-FH medium from individual chemical powders according to the published recipes (I). However, these media are now available commercially (e.g., GIBCO BRL Life Technologies, Inc., Gaithersburg, MD; Sigma Chemical Co., St. Louis, MO). All cell culture media are stored at 4°C. Adherent cultures are routinely maintained at 28°C at densities between 0.5-2.5 x lo6 tells/25-cm2 flask. Sf9 cells also can be grown as spinner cultures in TNMFH supplemented with serum, as above, and 0.1% (w/v) pluronic F68 (Sigma Chemical Co., St. Louis, MO; 19). However, we have not used spinner cells for transformation experiments. It appears that Sf9 cells grown in the presence of antibiotics cannot be used for transformation experiments. This is based on the observation that Sf9 cells maintained in the presence of 1.25 pg/rnL amphotericin B and 25 pg/mL gentamycin were highly resistant to G418 selection (20). Therefore, for transformation experiments, we only use Sf9 cells that have never been exposed to antibiotics. At this time, we do not know if Sf9 cells that have been exposed to antibiotics are permanently refractile to G418 selection or if G418 sensitivity can be restored by returning the cells to antibiotic-free medium. 2. Plasmid DNAs are routinely extracted from E. coli by the alkaline lysis or boiling procedures, and then purified by equilibrium centrifugation on CsCl density gradients, using the standard methods described in Sambrook et al. (21). Purified plasmid DNA preparations are stored at 4OC. 3. Transfection buffer: 25 mM HEPES, pH 7.1, 140 mM NaCl, and 125 mM CaC12.The pH of the transfection buffer must be adjusted precisely, and it must be filter-sterilized and stored at 4°C. Grace’s medium must be supplemented with 10% fetal bovine serum prior to being used for transfection. 4. G418 can be purchased from GIBCO BRL. G418 is weighed out immediately prior to use and added to TNM-FH at a concentration of 1 mg/mL. The medium is then filter-sterilized and supplemented with 10% (v/v) heatinactivated fetal bovine serum, 1.25 pg/mL amphotericin B (Sigma), and 25 pg/rnL gentamycin (Sigma). Although Sf9 cells to be used for transformation experiments should not be maintained in the presence of amphotericin B and gentamycin, these antibiotics may be added at the time of selection to prevent contamination. 5. Cloning cylinders can be purchased from Bellco Glass, Inc., Vineland, NJ (Cat. #2090).
192
Jarvis and Guarino 3. Methods 3.1. Insertion of Foreign Genes into IEl Expression PZasmids
The first task for the investigator who wants to produce a transformed Sf9 cell clone is to insert the foreign gene of interest into an IEl expression plasmid. The simplest IEl expression plasmid is pAcIE1, which consists of a copy of the AcMNPV IEl gene cloned into pUC8, as noted above (13). A detailed restriction map of this plasmid is shown in Fig. 2. The AcMNPV lE 1 sequencehas been published (14) and can be extracted from the Genbank database (accession #M16820). Because of the locations of the four HincII
sites, one can simultaneously
excise most of the
IEl coding sequence while maintaining a functional IEl promoter. Thus, a blunt-ended foreign gene fragment can be cloned into H&II-digested pAcIE1 to produce a plasmid capable of expressing that gene in uninfected Sf9 cells. Obviously, this requires a functional translational initiation site in the inserted DNA fragment. Expression plasmids constructed by using this or a very similar cloning strategy have been used successfully to produce transformed insect cell clones that express E. coli P-galactosidase, human tissue plasminogen activator, the AcMNPV polyhedrin protein, and the simian virus 40 large and small tumor antigens (12; D. L. Jarvis, unpublished). Several second-generation IEl expression plasmids, which we call immediate early transfer vectors (IETV), are shown in Fig. 3. These vectors are still in the developmental stages, so we do not yet know how well they will work. However, they are designed to have several advantages over the original IEl expression plasmids. Each will contain two different AcMNPV immediate early promoters, one derived from the IE 1 gene and one derived from another immediate early gene, called IEN (22,23). IETV-1 uses the IEl promoter to express the foreign gene of interest and the IEN promoter to express P-galactosidase. This vector is designed to facilitate screening of transformed Sf9 cell clones after transfection and antibiotic selection, since it will be possible to detect positive clones by using a simple color assay for P-galactosidase expression (21). IETV-2 and 3 incorporate the selected and unselected markers into a single plasmid. These vectors are designed to simplify the cotransfection procedure and, possibly, to improve the efficiency of cotransfection and, thereby, facilitate screening of transformants. The difference between
Gene Expression
I
BACULOVIRUS
in Transformed
Insect Cell Lines
IMMEDIATE-EARLY
Nar I,
TRANSFER VECTORS
193 I
Spe I, Jko RI, Pst I, Sac II, Kpn 1
pE?iq NE0 Upstream4
I ITEW
~D.ZETV-4 1 DHFR
~E.IETV-5
1 DHFR
Upstream+ Polyl(
I IIELJyy
Fig. 3. New, immediate,early expressionplasmids.The right halvesof IETV 2,3,4, and 5, beyond the hashmarks,ate identical to the right half of IETV, which is shown in detail. Each contains the IEl promoter followed by a multiple cloning site and an intact copy of the polyhedrin gene with downstream fknlcing sequences.The differences in the left halves of the various IETV plasmids
are discussed in the text.
194
Jarvis and Guarino
IETV-2 and IETV-3 is that the neomycin resistance gene is expressed under either IEN or IEl control, respectively. Finally, IETV-4 and 5 express the dihydrofolate reductase (DHFR) gene under IEN or IEl control. These vectors are designed for gene amplification in methotrexate, a procedure that was established in mammalian expression systems (24). These plasmids will be used to try to increase the levels of gene expression that can be obtained in transformed Sf9 cells, by increasing the number of integrated copies of the expression plasmid. Although it is not relevant to the use of these vectors for insect cell transformation, it might be noted that the IETV plasmids all contain an intact polyhedrin gene, as well as polyhedrin flanking sequences. The flanking sequences are included in order to target these vectors to the polyhedrin locus. It should be possible to use these vectors to produce recombinant baculoviruses, by using conventional methods (l), that will express a foreign protein under IEl control from the beginning of the infection. Since all of the vectors contain a copy of the polyhedrin coding sequence, these recombinants will be occlusion-positive, which will enhance their infectivity in the natural environment and their potential application as biological insecticides. Finally, it should be noted that the IETV plasmids all contain the complete 5’ untranslated region from the IEl gene. In the original constructs, the gene of interest was inserted at the HincII site at position -39 and our early work suggested that expression levels were higher if the 5’ untranslated region was intact (12). The following steps can be used to clone a foreign gene into pAcIE1 (Figs. 1 and 2), using standard recombinant procedures, as described in Sambrook et al. (21). Analogous procedures can be used to clone a foreign gene into any of the IETV plasmids, at any of the unique sites shown in Fig. 3. 1. Digest pAcIE1 with HincII. 2. Dephosphorylatethe blunt endswith calf intestinal phosphataseor bacterial alkaline phosphatase. 3. Phenol extract twice, andrecoverthe vectorDNA by ethanolprecipitation. 4. Preparethe geneof interestas a blunt-endedDNA fragment, and ligate in molar excesswith the vector fragment. 5. If you have not regenerateda HincII site, heat-inactivatethe ligase, and digest the ligation mixture with H&II. This will reduce the number of parentaltransformants. 6. Transform competentE. coli cells, and plate on ampicillin.
Gene Expression
in Transformed
Insect Cell Lines
195
7. Pick isolated colonies, prepare DNA minipreps, and screen for recombinants that contain the fragment of interest in the proper orientation. 8. Maxiprep the DNA from a recombinant, and purify by equilibrium centrifugation on a CsCl density gradient. 9. Harvest the DNA, quantitate, and verify the identity of the recombinant by restriction enzyme analysis.
3.2. Transfection of Transformed
and Isolation Sfs CZones
Once the expression plasmid has been constructed, it can be used to isolate and identify stably transformed Sf9 cell clones. The following description of the experimental protocol is based on the general approach outlined in Fig. 1. Sf9 cells are cotransfected with an expression plasrnid
plus pIElNeo, transformed cells are selected in G418, and individual transformants are screened to identify those that can express the gene of interest. It should be emphasized that excellent aseptic technique is required throughout this procedure. The selection and amplification of transformants requires long-term culture (about 1 mo) and, especially when the G418 is removed, even minor contaminants will outgrow the transformed cell cultures. 1. Start with healthy Sf9 cells grown in the absence of antibiotics. 2. Seed 2 million Sf9 cells/T25 flask in a total volume of 5 mL antibiotic-free TNM-FH. Seed one flask for an IElNeo-minus control and one flask for each expressionplasmid that you wish to usefor Sf9 cell transformation. 3. Allow the cells to attach for 1 h at 28°C. 4. Prepare the plasmid DNAs for transfection: a. Mix 1 yg pIElNeo with 2-20 ltg of the appropriate nonselected expression plasmid in a sterile microcentrifuge tube. Use the same amount of the nonselected expression plasmid DNA alone, without adding pIElNeo, as a control (see Note 1). b. Ethanol-precipitate or heat the DNAs at 65°C for 15 min. Either method will sterilize the DNA and help to prevent bacterial contamination of the transfected cultures. c. Aseptically add 0.75 mL of transfection buffer to each tube and use a pipetman to redissolve the DNA pellets. 5. Remove the medium from the cells, gently rinse with complete Grace’s medium (supplemented with 10% serum and antibiotics), drain completely, and add 0.75 mL complete Grace’s medium. 6. Add the appropriateDNA to each flask.
196
Jarvis and Guarino
7. Incubate for 2 h at 28°C. The transfection mixture should turn milky white. If this does not happen, there is probably something wrong with your transfection buffer (e.g., wrong pH?). 8. Drain the cells and gently wash twice with complete TNM-FH (supplemented with 10% serum and antibiotics). Feed the cells with 5 mL complete TNM-FH, and incubate overnight at 28°C. 9. Gently remove the cells from the surface of the flask by squirting them with medium through a Pasteur pipet. Avoid the formation of bubbles. Bring the cells to a total volume of 30 mL with complete TNM-FH. 10. Set up 60-mm Petri dishes containing 3 mL of complete TNM-FH. We routinely use six plates for each of the appropriate split ratios. 11. Plate the cell suspension at split ratios that are determined by the amount of nonselected DNA used for transfection (see Note 2): a. For 2 pg DNA, seed at 1:60 and 1:75. b. For 5-10 pg DNA, seed at 1:45 and 1:60. c. For 20 pg DNA, seed at 1:30 and 1:45. 12. Seal the plates in a plastic baggie, and incubate overnight at 28OC. The baggie provides a humidified environment that will minimize evaporation of the medium from the plates. 13. Prepare complete TNMFH containing 1 mg/mL G418 (4 ml/plate) and filter-sterilize. 14. Replace the medium in each plate with 4 mL complete TNMFH plus G418. 15. Reseal the plates in a plastic baggie and incubate for 1 wk at 28OC.All of the cells in the control plate, which was transfected in the absence of pIHlNeo, should die after about 3 d of G418 treatment. If the control cells survive after 1 wk of G418 treatment, this indicates that the selection procedure will not work (see Note 3). Many of the cells in the cotransfected samples also will die; however, significant numbers of survivors should be observed. 16. Remove the old medium very gently rinse the plateswith complete TNM-FH, and add fresh complete TNMFH plus G418. Incubate for another wk at 28OC. 17. Remove the old medium, very gently rinse the plates with complete TNMFH, and add fresh complete TNMFH minus G418. Incubate at 28OCuntil large, dense colonies form. These should be at least 2 mm in diameter and look dense under the microscope. Smaller, relatively less dense colonies are unlikely to survive the cloning procedure described below (see Note 4). 18. Drain an individual 60-mm plate, and pick the colonies. Do not try to do more than one plate at a time, or the second one will dry out before you are finished with the first. 19. Use forceps to dip a cloning cylinder into alcohol, then flame it, and allow it to cool. Coat one end of the cylinder by pushing it into a dollop of vacuum grease in a Petri plate. Push the cylinder, greased-side down, around a
Gene Expression
in Transformed
Insect Cell Lines
197
well-isolated colony. Make sure that a seal forms between the cylinder and the plate. 20. Add 100 PL of completeTNM-FH to the cloning cylinder, use a pipetman to dispersethe cells in the colony gently, and transferto a 96well plate. 21, Watch the cells daily, and amplify in a stepwrsefashion by transferringto a 24-well plate, a 6-well plate, and then to a T25 flask. The T25 flask is defined as passage1. 22. When the T25 flask is confluent, split 1:3 into freshT25 flasks. Use two of theseto preparepassagetwo freezer stocks.Use the other flask for screening and further amplification (seeNote 5). 3.3. Screening in Individual
for Foreign Gene Expression Transformed Sfs Clones
Once transformed Sf!3 cell clones have been obtained, they must be screened for expression of the gene of interest. Obviously, the exact method to be used depends on the nature of the recombinant protein and the tools that are available. The most informative approach is to assay for expression at the protein level, which can be accomplished in a variety of ways. One can assay for biological activity, if the protein has activity and an assay is available. For example, potential P-galactosidase (p-gal) expressing transformants were screened by producing freeze-thaw lysates and assaying those lysates directly for p-gal activity in a simple color reaction (12,21). Lacking an assay for biological activity, there are several good ways to detect protein expression if an antibody is available. These include indirect immunofluorescence, radioimmunoprecipitation, and/or Western blotting, which can be carried out by using published procedures (2,8,12,25). If no tools are available for protein analysis, one can screen transformants by Southern blotting to determine if the cells contain the foreign DNA sequences. However, it must be recognized that the presence of stably integrated DNA does not necessarily guarantee expression of the protein, owing to the possibility of DNA rearrangements during integration. RNA expression can be detected by using the relatively simple dot-blot hybridization approach. Again, this would not provide an unequivocal indication of expression at the protein level. A better way to look for RNA expression is S 1 nuclease or primer extension analyses, since they can be used to detect the expression of specific RNA species. However, these are technically more difficult, and even a positive S 1 or primer extension result does not guarantee expression of the protein.
198
Jarvis and Guarino
3.4. Expression of PGalactosidase and Tissue Plasminogen Activator A study on the use of the AcMNPV IEl promoter for expression of foreign gene products in stably transformed Sf9 cell clones has been published (12). The unselected markers were E. coli (p-gal) or human tissue plasminogen activator (t-PA), which were used as models for cytoplasmic and secretory pathway proteins, respectively. About half of the neomycin-resistant clones that were isolated expressed the unselected foreign gene as well. These clones contained stably integrated copies of the expression plasmids, and the foreign DNA sequences were specifically and accurately transcribed from initiation sites within the IEl promoter. p-gal expression was maintained for over 50 passages,suggesting that the transformants were highly stable. However, the level of p-gal expression was about lOOO-fold lower than the level of polyhedrin promoter-mediated P-gal expression in St9 cells infected with a recombinant baculovirus. This was a worst-case example, since p-gal is very highly expressed in baculovirus-infected cells (about 200 mg/L). The results obtained with t-PA, which is relatively poorly expressed and secreted from baculovirus-infected cells (cl mg/L), were more positive. t-PA was expressed and secreted at approximately equivalent levels by transformed or recombinant baculovirus-infected Sf9 cells. Most interestingly, the transformed cells contained no detectable t-PA in the intracellular fraction, Thus, although about two-thirds of the t-PA produced in infected cells was intracellular, virtually all of the t-PA produced in transformed cells was secreted. These observations supported the premise that the host cell secretory pathway is compromised during the later phases of baculovirus infection (8). They also suggested that stably transformed insect cells, expressing a foreign glycoprotein under the control of the IEl promoter in the absence of virus infection, might be a useful alternative for baculovirusmediated foreign glycoprotein production. The advantages of the transformed cell system are that expression can be continuous and permanent, and the efficiency of processing can be higher. However, the levels of expression provided by this system were disappointingly low, and this represents a significant disadvantage of this expression system. Currently, we cannot recommend the transformed insect cell for the expression of any gene product that can be expressed at reasonably high levels using a recombinant baculovirus. However, for proteins that are poorly
Gene Expression
in Transformed
Insect Cell Lines
199
expressed and processed in baculovirus-infected cells, the transformed cell approach is a promising alternative. In a followup study, we found that baculovirus infection of these transformed cells transiently stimulated and then strongly inhibited the production of both tissue plasminogen activator and p-gal (26). This has interesting implications for several possible applications of transformed Sf9 cells, as discussed in ref. 26. 4. Notes 1. We have determined the effects of using different amounts of the nonselected DNA for Sf9 cell transformation. The results showed that, within the range of 2-20 pg, there was no significant difference in the efficiencies of cotransformation or in the levels of expression obtained in the resulting transformants (20). 2. The split ratio used to seed the transfected cells is critical, since it will determine whether well-isolated transformed cell colonies are obtained. Interestingly, we have found that fewer total G418resistant colonies are obtained from cells transfected with larger amounts of the unselected DNA. Therefore, if you use more DNA, you must use a lower split ratio (higher cell density). Since the protocol indicates that the cells in the flask should be suspended in a total of 30 mL (step 9), a split ratio of 1:30 would be obtained by seeding 1 mL of this suspension/plate. For split ratios higher than 1:30, you must seed cl mL or perform dilutions with complete TNMFH as the diluent. We routinely do the appropriate dilutions and seed 1 ml/plate. The best way to ensure that you will obtain well-isolated colonies is to seed multiple plates at several different split ratios, perhaps from 1:20 to 1:lOO. 3. If the control cells survive G418 treatment, check to make sure that these cells were not grown in TNM-FH supplemented with antibiotics. Also check the age of your G418 stock, since it has a finite shelf life. GIBCOBRL provides an expiration date on its product. If you seeefficient killing in your control plates, but excessive cell numbers in your test plates, this means that the cells were seeded too densely. Use a higher split ratio. 4. Generally, it should not be difficult to get good transformed colonies. However, getting these colonies to survive the cloning procedure is the trickiest part of this protocol. The following are key hints that will facilitate the successof this procedure: a. The colonies must be relatively large and dense, because smaller, less dense colonies generally will not survive. Yet, the colonies cannot be allowed to get too large, or they will detach from the plate and will be lost when you remove the growth medium. This is ajudgment call that has to be learned by experience.
200
Jarvis
and Guarino
b. Vacuum grease is used to make the seal between the cloning cylinder and the plate. We simply scoop some vacuum grease into a Petri dish, and surface-sterilize it by placing it under W light in a laminar flow hood. The grease does not seem to create a significant contamination problem. A light coat of grease on the end of the cloning cylinder is sufficient to form a water-proof seal. c. Move quickly when you are picking colonies. We only drain one plate at a time and usually pick only four colonies from one plate. Once cloning cylinders have been placed over all four colonies, we add medium to each one before actually proceeding to pick any colonies. If the colonies dry out, which can happen very quickly, the cells are unlikely to survive the cloning procedure. d. Transformed Sf9 cells are quite fragile. Thus, one should not resuspend the colonies overenthusiastically. We use a Pipetman (P200), and draw the TNMFH in and out two or three times to break the colony loose and partially disperse the cells. If the cells are clumpy when seeded into the 96-well plate, they can be dispersed later with a higher chance of survival. e. One can also be too gentle, however, and fail to remove many of the cells from the plastic. This can be monitored by looking at the plates under the microscope for residual cells after picking a colony. Scraping the colony with the pipet tip seems to be better than repetitive pipeting. f. Be patient during amplification of the cells and allow them to reach normal densities, but only within reason. Our best transformed clones have grown to 80-90% confluency after l-3 days, but some can take a wk or more at each step. It is not unusual to have some colonies which completely stop growing, and we usually discard these and slow-growing colonies, in favor of others. g. Expect about 50% survival of the transformed cell colonies you pick. We routinely pick a dozen colonies and expect to get six survivors. 5. Freezer stocks are prepared by gently removing the cells from the surface of the T25 flask, centrifuging at low speed (e.g., half-speed in an IEC clinical centrifuge) for 1 min, and gently resuspending the freezer stocks in complete TNM-FH plus 10% DMSO. It is best to freeze the cells at the rate of about -l”C/min to maintain cell viability. This can be accomplished by using a relatively inexpensive commercial apparatus that has been designed for this purpose (“Cry0 1°C freezing container,” Nalge Company, Rochester, NY; catalog #5100-0001). Subsequently, the cells should be placed in liquid nitrogen for longterm storage.
Gene Expression
in Transformed
Insect Cell Lines
201
References 1, Summers, M. D. and Smith, G. E. (1987) A manual of methods for baculovirus vectors and insect cell culture procedures. Texas Agricultural Experiment Station Bulletin No. 1555. 2. Webb, N. R. and Summers, M. D. (1990) Expression of proteins using recombinant baculoviruses. Technique 2,173-188. 3. Luckow, V. L. and Summers, M. D. (1988) Trends in the development of baculovirus expression vectors. Bio/Technology 6,47-55. 4. Miller, L. K. (1988) Baculoviruses as gene expression vectors. Ann. Rev. Microbial. 42, 177-199. 5. Luckow, V. L. (1991) Cloning and expression of heterologous genes in insect cells with baculovirus vectors, in RecombinantDNA Technology andApplications (Prokop, A., Bajpai, R.K., and Ho, C. S., eds.), McGraw-Hill, New York, pp. 97-152. 6. Jarvis, D. L. and Summers, M. D. (1992) Baculovirus expression vectors, in Recombinant DNA Vaccines: Rationale and Strategies (Isaacson, R. E., ed.), Marcel Dekker, New York, pp. 265-291. 6a. O’Reilly, D. R., Miller, L. K., and Luckow, V. A. (1992) Baculovirus Expression Vectors. W. H. Freeman, New York. 7. Greenfield, C., Patel, G., Clark, S., Jones, N., and Waterfield, M. D. (1988) Expression of the human EGF receptor with ligand-stimulatable kinase activity in insect cells using a baculovirus vector. EMBO J. 7, 139-146. 8. Jarvis, D. L. and Summers, M. D. (1989) Glycosylation and secretion of human tissue plasminogen activator in recombinant baculovirus-infected insect cells. Mol. Cell. Biol. 9,214-223.
9. Ellis, L., Levitan, A., Cobb, M. H. and Ramos, P. (1988) Efficient expression in insect cells of a soluble, active human insulin receptor protein-tyrosine kinase domain by use of a baculovirus vector. J. Virol. 62, 1634-1639. 10. Lanford, R. E., Luckow, V., Kennedy, R. C., Dreesman, G. R., Notvall, L., and Summers, M. D. (1989) Expression and characterization of hepatitis B virus surface antigen polypeptides in insect cells with a baculovirus expression system. J. Virol. 63, 1549-1557. 11. Kuroda, K., Veit, M., and Klenk, H.-D. (1991) Retarded processing of influenza virus hemagglutinin in insect cells. Virology 180, 159-165. 12. Jarvis, D. L., Fleming, J. G. W., Kovacs, G. R., Summers, M. D., and Guarino, L. A. (1990) Use of early baculovirus promoters for continuous expression and efficient processing of foreign gene products in stably-transformed lepidopteran insect cells. Bio/Technology 8,950-955. 13. Guarino, L. A. and Summers, M. D. (1986) Functional mapping of a trans-activating gene required for expression of a baculovirus delayed-early gene. J. Virol. 57, 563-571. 14. Guarino, L. A. and Summers, M. D. (1987) Nucleotide sequence and temporal expression of a baculovirus regulatory gene. J. Virol. 61,2091-2099. 15. Wigler, M., Perucho, M., Kurtz, D., Dana, S., Pellicer, A., Axel, R., and Silverstein, S. (1980) Transformation of mammalian cells with an amplifiable dominant-acting gene. Proc. Natl. Acad. Sci. USA 77,3567-3570.
202
Jaruis and Guarino
16. Southern, P. J. and Berg, P. (1982) Transformation of mammalian cells to antibiotic resistance with a bacterial gene under control of the SV40 early region promoter. J. Mol. Appl. Gen. 1,327-341. 17. Jarvis, D. L. (1993) Foreign gene expression in insect cells, in Insect Cell Culture Engineering (Goosen, M. F. A, Daugulis, A. J., and Faulkner, P., eds.), Marcel Dekker, New York, pp. 193-217. 18. Vaughn, J. L., Goodwin, R. H., Thompkins, G. J., and McCawley, P. (1977) The establishment of two insect cell lines from the insect Spodopteru frugiperdu (Lepidoptera:Noctuidae). In Vitro 13,2 13-2 17. 19. Murhammer, D. W. and Goochee, C. F. (1988) Scaleup of insect cell cultures: protective effects of pluronic F-68. Bio/Technology 6, 1411-1418. 20. Jarvis, D. L. (1991) Continuous expression of foreign proteins in stably-transformed lepidopteran insect cells, in Proceedings of the 1991 World Congress of Cell and Tissue Culture, Tissue Culture Association, Columbia, MD, pp. 76-92. 21. Sambrook, J., E. F. Fritsch, and T. Maniatis. (1989) Molecular Cloning; A Laboratory Manual, 2nd ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. 22. Carson, D. D., Guarino, L. A., and Summers, M. D. (1988) Functional mapping of an AcNPV immediate early gene which augments expression of the IE-1 tansactivated 39K gene. Virology 162,444-45 1. 23. Carson, D. D., Summers, M. D., and Guarino, L. A. (1991) Molecular analysis of a baculovirus regulatory gene. Virology 182,279-286. 24. Kaufman, R. J. (1990) Selection and coamplification of heterologous genes in mammalian cells, in Methods in Enzymology, vol. 185, (Goeddel, D. V., ed.), Academic, San Diego, pp. 537-566. 25. Volkman, L. E. (1988) Autogrupha c~lifomicu MNPV nucleocapsid assembly: inhibition by cytochalasin D. Virology 163,661-666. 26. Jarvis, D. L. (1993) Effects of baculovirus infection on IEl-mediated foreign gene expression in stably transformed insect cells. J. Virol. 67,2583-2591.
CHAPTER12
Scale-Up of Recombinant Virus and Protein Production in Stirred-Tank Reactors Rosanne L. Ybm, Antoine W. Caron, Bernard Massie, and Amine G Kamen 1. Introduction Stirred-tank fermentors were originally designed to perform microbial cultures (I). When the need arose for adequate,posttranslationally modified proteins, molecular biologists and bioprocess engineers turned to mammalian and insect cell culture using these fermentors for large-scale productions (i.e., New Brunswick Scientific’s CelliGen, New Brunswick, NJ; Alfa-Laval’s Chemap, Mannedorf, Switzerland). Stirred-tank reactors (STR) are well known and characterized (as well as airlift reactors mentioned in Chapter 13) and provide a simple approach to large-scale suspension cultures when monolayer and spinner flask cultures cannot provide ample material. Depending on the quantity required, bioprocesses can vary in complexity from short-term (5-7 d) batch cultures to longer-term fed-batch (7-10 d) or perfusion processes (7 to 2 30 d). Scale up of recombinant protein production using the baculovirus insect cell system involves a variety of fields, such as recombinant baculovirus construction (2), medium development (2-6), metabolic studies (7-9), protein purification and quantification, as well as rational exploitation of available bioreactor hardware (10). Research has been also focusing on the factors affecting recombinant protein production in SEI cells: cell density and physiology (II), serum concentration (12,13), From: Methods in Molecular Biology, Vol. 39: Baculotirus Expression PrOtOCOlS Edited by: C. D. Richardson Q 1995 Humana Press Inc., Totowa, NJ
203
204
Tom et al.
multiplicity of infection (134 and defective interfering virus particles (1#,15). The main goal is to achieve and maintain maximal productivity with each successive increase in process scale. However, the ability to generate material, with the proper biological activity in a sufficiently pure form, probably constitutes the greatest challenge. Furthermore, any scale-up process should first be tailored to the product’s intended use, in terms of quantity and quality required, for example, biopesticide production calls for thousand-liter cultures and extremely cheap medium. On the other hand, crystallization of hormone receptors for structural and functional studies demands milligram quantities at high levels of purity. The baculovirus system’s widespread use for recombinant protein expression is largely owing to its efficiency in producing high levels of recombinant proteins. This potential should encourage the researchers to ensure optimization of their baculovirus recombinant constructs in order to obtain maximal protein yields (15a,15b). When scaling up protein production, differences in expression levels can be attributed to a combination of factors: from bioreactor operating conditions, to medium quality, and even to the stability of the recombinant virus construct and/or the recombinant protein itself. Posttranslational modifications and associated biological activity of the target product are influenced by the insect cell host, and the production medium in which it is cultured during infection. Verification of the recombinant protein’s structure and activity ultimately takes part in the scale-up process itself. The final protein yield and activity depend on successful culture maintenance. Infected insect cells become increasingly fragile and sensitive to the physicochemical environment of the culture (3,16) during the infection phase. One of the primary bioengineering objectives is to oxygenate large-scale, high-density cultures sufficiently at low shearing rates. Insect cells have been found to have comparable specific oxygen uptake rates (q0,) to animal cells (3,7). However, during infection (recombinant protein/virus production stage) q02 values nearly double as compared to uninfected cultures (7,7a,16a,16b). Insect cells can easily attain high densities (>107 cells/ml) and exert high oxygen demands, which can be met by increasing the O2 partial pressure,agitation, and/or sparged flow rates. Since infected insect cells are sensitive to the mechanical stresseswithin the bioreactor (16,17), the power dissipated in increasing the oxygen transfer rate must not be detrimental to the host cells (10).
Scale-Up in Stirred-Tank
Reactors
205
Table 1 Recombinant Proteins Produced with Infections at 2 x lo6 Cells/mL* Protein
Description
Concentration, mg/L
Papain VP6 EGPR-ED G proteins Yeast Kex2 Protease
Secreted Intracellular Secreted Membrane Secreted
l-5 160-180 2 15 15
Reference Tessier et al., 1991 (21) Caron et al., 1990 (II) Chapter 20 in this book Labrecque et al., 1992 (22) Germain et al., 1992 (23)
*Secreted proteins are produced in serum-free media.
A number of improvements in bioreactor design (18), cell protection additives (17,19), and media formulations (3) have greatly facilitated large-scale insect cell culture. The methods presented here pertain to the details of classic stirred-tank bioreactor preparation and associated accessories. A helical ribbon impeller (HRI) prototype bioreactor (7,lO) is featured in this chapter. Accessory hardware, such as the crossflow filter (20), is also presented as an alternative device for cellmedium separations. Culture monitoring and off-line metabolic analysis (20~) are exploited for process optimization. A thorough understanding of cellular kinetics (i.e., nutrient consumption, inhibitor accumulation, and respiration) and optimized media formulations (i.e., low-serum or serum-free) will yield better bioprocess control, improved product yields, and cost-efficient scale-up. Improvements are still being sought because there is not one universal bioreactor bioprocess that works for every recombinant protein expression system. The following methods were based on the production of many different recombinant proteins (intracellular, membrane-associated, secreted) in Sf9 cells cultured from the 3- to 11-L scale, at low and high cell densities. 2. Materials 2.1. Cell Line
and Recombinant
Viruses
1. S’podopterufrugiperda cell line Sf9 (ATCC CRL1711) 2. Recombinant Autographa califomica Nuclear Polyhedrosis
Table 1 for references).
Viruses (see
Tom et al.
206
1, 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13.
2.2. Media and Solutions Grace’s Antheraea insect cell medium powder (24). Protein-free media: IPL/41 (4) or SF90011. Pluronic polyol F-68 10% (w/v). TC yeastolate. TC lactalbumin hydrolysate. Calcium chloride (if not already supplied in Grace’s powder). Sodium bicarbonate. Sodium chloride 15% (w/v). Sodium hydroxide 1ON Phosphoric acid diluted in distilled water (1:7) for adjusting pH in serumbased cultures. Gentamicin. Fetal bovine serum (FBS). TNMH formulated according to Summers and Smith (2). The protein-free medium IPL/41 is constituted as described by Inlow et al. (4), which reports the use of a lipid emulsion. Media and solutions should be prepared with presterilized, deionized water to prevent viral contaminations if not using commercial liquid preparations or if virus-free deionized water is not available. 2.3. Basic
Stirred-Tank Bioreactor Components and Accessories Most stirred-tank bioreactors (STR) operate in a similar fashion, whether they are destined for microbial or eukaryotic cell culture. Differences between bioreactor models for animal cell culture lie in their low-shear and mass-transfer capacities, which are governed by a combination of their aeration and agitation regimes. Fig. 1 shows the setup of a 5-L STR connected to a cell-medium separation unit that works by crossflow filtration (optional accessory). STRs are usually supplied with paddle (Rushton) or marine impellers for culture agitation. Rushton impellers mix fluids by shearing with radial pumping, whereas marine impellers propel fluids in an axial motion. Mixing in the latter case is achieved mainly by fluid circulation with minimum stress. 2.3.1. Basic Components Comprising a 5-L Stirred-Tank Bioreactor 1. Ancillary tubing should be sterilizable silicone for inoculation, harvest, reservoir, and acid or base additions 2. Microbial an vent/filters (0.2 pm) for feed bottles, sampler, and gas inlet and outlet.
Scale-Up
in Stirred-Tank
Reactors
207
1 CONDENSOR
--L!k B S ^---. {,
; /
/
\
’ .I ‘5
S AMPLER
d
EFJ ,*
CELL/MEDIUM SEPARATOR,’
,
/
,
,#’ /
/
If
t’
STIRRED TANK BIOREACTOR
Fig, 1.5-L STR with: -3.84-L working liquid volume, sufficient number of ports for probes, gases and feed/exit lines, parameter controls, and accessory hardware (connected by dotted lines). Inoculation and harvest lines not shown. A: Acid (and/or base) bottle; F: fresh medium; S: spent medium. A crossflow filter device (CFF) is used as a cell/medium separator unit (connected to STR by dashed lines). Inset: custom-made slanted paddle impeller used to improve the oxygen transfer rate at the liquid-gas interface of the culture. 3. Nitrogen, oxygen, and air should be of ultrahigh purity. 4. pH, dissolved oxygen (DO), and temperature probes (if not supplied with bioreactor already). 5. External DO controller (i.e., a solenoid valve system for blending gases) if not supplied with bioreactor.
208
Tom et al.
6. External pH controller to activate/shutoff acid/baseaddition (if not available in bioreactor unit). 7. Peristaltic pump for acid/baseline(s) (1-12 rpm). 8. Linear recorder to monitor DO and/or pH control.
9. Laminar flow hood in proximity of the bioreactor. 10. Useful accessories to have for leak test: gas rotameters (O-500 mL/min and/or O-2 L/min range), Teflon tape, gas manometer, diluted detergent solution (snoop). 2.3.2. The Helical Ribbon Impeller Bioreactor (HRI)
Originally designed for the mixing of viscous solutions, the HRI’s utility has been extended to the culture of shear-sensitive cells (7,10). The HRI resembles a double helix, and yields homogeneous mixing and aeration by bubble entrainment, thus circumventing the need for sparging at cell densities up to at least lo7 cells/ml. The 11-L HRI bioreactor (Fig. 2) was designed to provide efficient mixing with high mass-transfer rates while minimizing mechanical damage to shear-sensitive cells. Its high mass-transfer capacity is attributed to the presence of three surface baffles, which increase the turbulence of the fluid on mixing. Specifications of this STR are described in detail elsewhere (7, IO). A 3.5-L Chemap microbial bioreactor (Alfa-Lava& Mtinedorf), with in situ sterilization, was retrofitted with an HRI and three surface baffles; it has a 2.8-3.0-L working volume. The 3.0- and 11-L HRI reactors yield similar physical environments for Sf!J cultures (10). 2.3.3. Cell-Medium
Separation
Unit
A circular cross-flow (tangential) filtration system (CFF) was designed in our lab (20) to separatecells from spent medium in a perfusion process (Section 3.6.3.). Like hollow-fiber cartridges, the CFF is another type of cell-medium separator. The CFF design is an autoclavable, three-plate, double-membraned system. The 30-cm (diameter) circular, stainless-steel version contains a spiral tube (3/16 in.) providing a concentric path for the culture. The CFF has also been used in batch processes that required a fresh medium replacement step after virus adsorption (Section 3.6.2.). The following lists the additional components required to use the CFF: 1. Filtration membranes (25cm diameter) 2- or 5q.m pore sizes(Nucleopore,
California). 2. Peristaltic pump for culture recycling (Watson Marlowe #6035, Smith and Nephew, UK). 3. Silicone peristaltic pump tubing 15.9-mm id; 3.2 mm thick.
Scale-Up
in Stirred-Tank
L
Reactors
209
III Y
Fig. 2. Diagram of the 11-L helical ribbon impeller bioreactor (I-RI) prototype with three surface baffles. It is a 16-L vessel with an 11-13 L working volume.
2.4. Analytical
Equipment
2.4.1. Off-Line Data Acquisition 1. Hemocytometer and trypan blue. 2. External pH meter. 3. HPLC system for carbohydrate. organic, and amino acid analyses.
210
Tom et al.
4. Enzymatic kits are commercially available that can determine glucose, lactate, pyruvate, or glutamine. 5. Detailed analytical conditions are mentioned in a previous paper (7). 2.4.2. On-Line Data Acquisition These components are optional, and depend on the availability and financial resources of the research facility. 1. Computer with a data acquisition program (temperature, DO, pH, gas flow rates, pressure, and so forth). 2. Mass flow meters, for oxygen and nitrogen flow rate control as a gas blending unit. 3. Mass spectrometer.
3. Methods The following methods show detailed and practical approaches for large-scale, Sf9 cultures in STRs that otherwise may not be documented in reactor manuals or in publications. The flow charts summarize STR preparation and sterilization (Fig. 3), as well as inoculation and culture monitoring procedures (Fig. 4). Regardless of size, STR preparation for autoclaving is similar between models with the exception of those that have in situ sterilization (Fig. 3). 3.1. Inoculum
Preparation
1. Use spinner flasks containing an impeller with a magnet for agitation (incubated in humidified air, 27OC, 100 RPM). See Table 2 and Fig. 5 for typical growth rates. Alternatively shake flasks can be used as long as the culture volume does not exceed 40% of the vessel volume. 2. When preparing inocula for bioreactor cultures, it is highly recommended to have a backup inoculum prepared also. 3. Sf9 inoculum cultures are seeded between 1 and 2 x lo5 cells/ml in TNMFH medium supplemented with 10% (v/v) FBS and 0.1% (w/v) Pluronic F-68. Within 4-5 d, the density should be -2-3 x lo6 cells/ml (mid- to late-exponential phase) and be ready to inoculate (to l-2 x lo5 cells/ml) the bioreactor using a lo- to 2 20-fold dilution (see Notes l- 4). 4. If the cells are destined for serum-free culture, then initial seeding densi-
ties should not be ~3 x lo5 cells/ml. 3.2. Bioreactor
Preparation
and Sterilization
1. All feed/outgoing tubes from the bioreactor must be long enough to permit manipulations (i.e., filling and bottle changes) inside a laminar hood (see Note 5). Nylon cable tie-wraps are ideal for securing all tube connections.
verily ebctrolyte levels In pH and DO probes, cattbmte fmd insert them tnto bbmactor for sterilhatlon Check other sensors le. Temp. pressure, etc. .
a
Check gaskets, seals, septa. fdters. tublngs & o-rings 6.7.14
J m Fermenter filled 6 me&y for stelfuzatbn
Check-fist for autoclaw sterlllzation: -Check aft flners (L tubng -Auloclave for Lz50 mh, 121 ‘C 6 15 PSI. Stetfthatbn dumtbn h bbreactorvobmedependent -Fiernove from the auto&we when tempemture is bwer then 30 “c to avoid non-sterile au sIlclkm. &or -Ptacethesterite.hotreactofna bmharakfbwhodtocodoff.
clean
Check-fiat for in sttu stufftzatk~~ -Folbw menufactureh stenliitlon procedure, otherwwe fotbw these mah steps cbse all ports -Install exhaust alr-gmup wtth pressure hokfvlg vshfe -Set stirrer at 200-300 RPM Set safety hood around bbreador -Set steniiitlon dumtl0n set temperature stertruahon -Start stenlizatkn -After stenlizatbn albw bmreactor to cool dam to temperature setpomt for alnure mrKllltlons I
I
*
or nitmgen into the sterile bmeactor to mantah a posltnre pressure -Attach water condenser at gas outlet
Fig. 3. Flow chart for STR preparation and sterilization; numbers appearing besides each step refer to details in the Section 4. of this chapter. After harvest, all bioreactor parts and pieces should be thoroughly disinfected (i.e., Virocidin-X), cleaned with nonabrasive detergent, and be rinsed with distilled water. Exposure to hypochlorite-based solutions should be avoided.
212
Tom et al.
exithneslnseltd -.
atO%and loo% SET CONTROL PARAMETERS -Temperature setpoInt= 27 “C -Strrflng speed = 50-100 RPM Addlves. F-66, Ab, Anti-A, FBS if not SFM
1
OFF-UNE ANALYSIS
-Viability -nutrient analysis (Glucose, amino aads, carboxvllc aads;.) a - lnhlbitors analysis (ammonia, lactic
I
ON LINE DATA ACQUISITION: -W PH. Temperature -RPM -PO2, PC02 -Gas flow rates -Pressure I
I
-
Fig. 4. Flow chart summarizing broreactor inoculation, culture process, and monitoring. Abbreviation: F-68 = Pluronic F-68; A b= antibiotic; FBS = fetal bovine serum; SFM = serum-free medium; Anti-A = antifoam. Numbers next to each step refer to details in Section 4 of this chapter. 2. All tubes leading out of the bioreactor must be clipped (see Note 6), whereas the pH and DO probes containing their respective electrolytes should be inserted in their designated ports (Fig. 3). The pH probe should already be calibrated (pH 4-7). The DO probe is calibrated after sterilization.
Scale-Up in Stirred-Tank
Reactors
213
Table 2 Sf9 Growth Characteristics* Medium TNMFI-I (2) +lO% FBS EL/41 (4) SF90011
Doubling time, h
Maximum cell (x 106 cells/ml)
17-24
4.5-5
22-28 24-27
5.0-6 8.0-9
*Sf9 growth characteristics were reproducible between spinner and bioreactor cultures.
3. Connect resterilizable microbial air vents/filters (0.2 pm) to reactor sampler, and gas inlet and outlet (see Note 7). 4. Bioreactor ports/openings that are not employed should be sealed off with accessory bolts supplied or by connecting a short piece of silicone tubing that is clipped. 5. Prior to sterilization, the bioreactor can be up to half-full with distilled water (dHzO) to keep the pH and O2 probes wet. 6. If the probes are polarographic, as for in situ sterilizable bioreactors, be sure to polarize/charge them at least 24 h before use. 7. Do a leak test prior to sterilization to verify sources of leaks at inlets and ports. Introduce about 5 psi gas into the reactor, and check for any pressure drop using a gas manometer at the bioreactor outlet (see Note 8). Neglecting this stepwill increase the probability that an unchanged gasket (O-ring), a loose bolt, or a punctured line will be the source of contamination as air gets sucked in during bioreactor cooling or the culture period. 8. Although all reactor tubings are clipped (i.e., feed, exit, and sampling lines closed), the gas inlet and outlet should remain unclipped. The bioreactor should have at least two or three openings to permit vapor circulation between the reactor interior and the autoclave chamber. 9. Put the bioreactor first into the autoclave, followed by the tubes, bottles, and accessorres,if applicable. Minimize tubing crossovers and entanglement. 10. The reactor, half-filled with dH20, should be steam-sterilized at 121°C. Sterilization duration depends on the volume of water present, i.e., 60 min for 2 L. A minimal amount of water should be present in order to keep the probes wet during sterilization. 11. Allow the sterile bioreactor to cool off in the laminar flow hood with its accessories. 12. If the hood cannot accommodate the sterile reactor setup, secure the reactor on its module to cool off. Send air or nitrogen through the gas inlet at
214
Tom et al.
,o.o
+ y
a. b. e. d.
3 t ,+
B&h E&h wlth SFM nplawmonl SFM growth and roplramenl Imfllclont Infoction
.
1
V.”
2
infection
6
4
TIME
6
10
(d)
Fig. 5. Four examples of Sf9 growth and infection curves in STRs; arrows indicate time of infection. (A) Simple 4-L batch (TNMFH + 10% FBS + 0.1% pluronic F-68) with infection at 2 x 106cells/ml (CelliGen bioreactor); (B) 11-L (HRI bioreactor) batch growth (TNMFH + 10% FBS + 0.1% pluronic F-68) with infection at 4.5 x lo6 cells/r& followed by serum-free medium (SFM) replacement (TNMFH + 0.2% pluronic F-68); (C) 2.8-L batch growth and fresh medium replacement, following infection at approx lo7 cells/ml, using serumfree SF90011 medium only (Chemap bioreactor retrofitted with an HRI); (D) Continued cell growth owing to inefficient infection at 2 x lo6 cells/ml in IPL/ 41 (CelliGen). -100-200 mL/min in order to maintain a positive pressure within the bioreactor. This prevents nonsterile air suction into the bioreactor during cooling. 3.3. Bioreactor Inoculation 1. Culture media should be prepared in advance and should have passed a sterility test where an aliquot of the freshly filtered medium was incubated (27OC) for -2 d without any contaminant appearance (Fig. 4).
Scale-Up
in Stirred-Tank
Reactors
215
2. DO calibration can be performed in either water or in a partially filled (half to two-thirds) bioreactor containing basal insect cell medium without pluronic F-68 and serum (see Note 9). Adjust the temperature set-point (sp) to 27OC and the agitation rate to 70-75 rpm (culture operating conditions). 3. If a supplementary gas sparger is available in the bioreactor, it can be used to bubble the gas into the liquid to speed up the DO calibration. 4. If a linear recorder (or a data acquisition system) is attached to the DO control panel of the reactor, on-line printing of the DO curve can facilitate the determination of the O2saturation point. 5. Send nitrogen through the inlet to deoxygenate the liquid to 0%; when the reactor DO indicator reads “0” then set “zero point” to 0. 6. Remove the nitrogen source and send air through the inlet to oxygenate the liquid; after the DO indicator appears to have stabilized, adjust the setpoint value to 100% to complete the DO probe calibration. 7. During medium fill-up and inoculation, remove the air flow until bioreactor is filled to appropriate volume. Gas flow rate can resume immediately once all additives have been added to the culture. 8. Add to the culture medium its appropriate additives (FBS, pluronic F-68, emulsion, antifoam, and so on) and 10-30 pg/mL gentamicin (see Notes 10 and 11). 9. Add the appropriate volume of cells to attain correct density (see Note 12). A small volume of medium can be washed down the inoculation line to pick up any residual cells in the inoculation tube into the bioreactor. 1. 2. 3. 4.
3.4. Culture Control and Monitoring Set the temperature control to 27°C during fill-up/inoculation and connect the cold water line to the air condenser, which will minimize liquid evaporation. Set the gas flow rate to between 200 and 400 mL/min for surface aeration of the culture, or at a rate corresponding to 0.05-O. 1 vol gas/liquid vol/min (vvm) for sparged cultures. Remove the air source from the gas inlet and connect the DO control line to it. Adjust the DO sp value to 30-35% (Fig. 4). Activate the DO controller and within 2-3 h the DO level reading should be near the sp value. Air, N2, and O2 are the only gases required to maintain DO control for insect cell culture (see Note 13), since COZ is not needed. If a mass spectrometer is available, oxygen consumption and carbon dioxide production rates (OUR and CPR) can be monitored by measuring the inlet and outlet gas compositions (7).
216
Tom et al.
5. Sample approx 75-100 mL from the bioreactor and put it in a lOO-mL spinner flask or 250~mL Erlenmeyer flask, aseptically. This constitutes the small-scale culture control for the bioreactor (see Note 14). 6. Count bioreactor and control cultures at least once a day to monitor normal growth rates. 7. Cell supematantsshould be collected, filtered (45 pm), and frozen at -80°C for subsequent analyses of sugars (i.e., glucose), amino acids (i.e., glutamine, alanine), and byproducts (ammonia, lactic acid). 8. Verify and adjust culture pH to 6.2 f 0.1, if necessary (see Note 15), in order to maintain high cell viability (II).
3.5. Culture
Infection
Large volumes (>500 mL) of virus stock at low passage number (-Sth6th after plaque purification) can be produced by simply infecting a bioreactor culture and retaining the supematant at the end of the infection phase (see Notes 16 and 17). 1. For any volume of virus stock, a low multiplicity of infection (MOI) (MO1 = 0.1 to 1 PFWcell) should be used in order to minimize the chances of producing defective interfering particles (DIPS). Recent reports showed an association between high MO1 infection and increased probability of DIP production (14,15). Infect an STR (or spinner/shake flask) culture at -2-3 x lo6 cells/ml and watt 72-96 h postinfection before harvesting (see Note 18). 2. During the course of infection, the culture viability will decline and contribute to increased amounts of cell debris in the medium. Centrifugation at SOO-lOOOg,20-30 min, 4OC,can further separate the cellular debris from the supernatant, which will serve as viral stock. 3. The cell/medium separator can be employed to concentrate (see Note 19) the infected culture and collect the permeate (centrifugation may be required again to remove cell debris if the permeate appears cloudy). 4. Higher-titer virus stocks(2108 PFU/mL) can be obtained by harvesting the supernatant of infected, higher-density cultures (5 x lo6 cells/ml), provided that the medium is replaced for fresh medium, 1 h after infection (see Section 3.6.2.). 5. For recombinant protein production, the MO1 should be higher than that used to produce virus stocks. To ensure simultaneous infection of all the cells in the bioreactor, an MO1 of at least l-5 PFU/cell can be employed. If the virus titer is low (510’ PFWmL), the amount of virus added to the culture should not exceed more than 10% of the total culture volume. Excessive virus volumes could introduce unnecessary amounts of spent medium cellular debris and possibly compromise recombinant protein production.
Scale-Up in Stirred-Tank
Reactors
217
3.6. Production Modes Three production modes used in our labs are described in this section. Batch-replacement and perfusion modes were developed to circumvent limitations in media compositions when batch mode procedures were not sufficient to support growth of high-density cultures (15-6 lo6 cells/ ml,) and maximal specific production therein. Operating procedures increase in complexity as one goes from batch to batch/replacement to perfusion processes. Moreover, each process becomes increasingly difficult to manipulate when higher volumes and production scales are sought. Although 75-L bioreactors are not normally found in labs, multiple runs of simple batch cultures can be used to accumulate the protein quantities desired. Alternative production modes, such as batch/replacement and perfusion permit high-density cultures in lab-scale bioreactors (2-20 L). In order to determine which production mode to use, the operator must consider what the production demand is and what the expression level of the recombinant baculovirus is. Although commercial insect cell media compositions are always being improved, detailed knowledge of nutrient utilization, growth, and production kinetics should be well known, for a given medium, in order to optimize production conditions when using any production mode. In any case, the challenge is to maintain high productivity at high cell densities. 3.6.1. Batch Process Grow the cells to 1.5-2 x lo6 cells/ml (Fig. 5) and then infect them with the virus at an MO1 that has been previously determined (2,13a) to be efficient for recombinant protein production, Growth and infection take place in the same medium. Reduced product yields have been observed in TNMFH + 10% FBS (4) and in IPL/41 (II) at infection densities between 2 x lo6 and 3 x lo6 cells/ml. In SF90011up to 3 x lo6 cells/ rnL can be infected in the same medium. Product yields can be improved at higher densities with medium renewal however. 3.6.2. Batch and Replacement (Sequential Batch Process) 1. Grow the cells to 5-10 x lo6 cells/n& dependingon the medium’s biomassyield, and then infect the cells at an appropriateMO1 (Fig. 5). 2. Allow 30-60 mm for virus adsorption. 3. With a cell/medium separator,the culturecanbe concentratedto -10% of its original volume (seeNotes 18-20).Therecyclingpump speed,for crossflow filtration, can be setto 0.5-l L/mm. A peristaltic pump can be installed in
218
Tom et al.
order to increase the exit rate of spent medium (Fig. 1). The resultant culture concentratecanremain in the bioreactorandbe immediately reconstituted with fresh medium to its initial volume. For absolutely serumfree protein productions, the resultant concentrate can be centrifuged (-2OOg,15 min, 4”C), followed by serum-freemedium reconstitution, 3.6.3. Perfusion
Process
Perfusion allows simultaneous medium feeding and spent medium removal in order to attain and maintain higher biomass concentrations (20) that would not be reached in simple batch cultures. A cell/medium separator device permits cell retention and recycling in a closed loop, whereas spent medium (permeate) exits across the membrane (Fig. 1). 1. Cells can be grown to 2-5 x lo6 cells/ml in batch mode, and then perfusion can be started. 2. Calculate the volume of fresh medium neededto effect the requiredfresh medium feeding rate. 3. Adjust the peristaltic pumps such that the permeatewill be pumped out from the cell/medium separatorat the sameratethat fresh medium is being pumped into the bioreactor. 4. Activate the cell recycling pump such that it pulls the culture out of the bioreactorinto the cell/medium separatordevice from which the cells will return to the bioreactor. 5. Connectthe tubing from the fresh medium bottle to the bioreactor,andthe permeatetubing from the cell/medium separatorunit to the spentmedium bottle, using a double-headedperistaltic pump. Alternatively a level controller can activate the fresh medium feed, allowing the culture volume to remain constant.Make sure that the fresh and spentmedia are flowing in the properdirectionsoncethe pump hasbeenadjustedto the desiredperfusion rate. 3.7. Culture Harvest The moment of harvest depends on the nature of the recombinant protein, which could be intracellular, membrane-associated, or secreted. To ensure maximum protein production and recuperation, the production kinetics should already be known for a given product before culture scaleup is considered. However, in scaling up infection cultures, conditions in the bioreactor may slightly alter the production kinetics. Thus, off-line monitoring of the production during infection can determine if the harvest should be executed 24-36 h earlier or later than originally planned. Harvest conditions will vary depending on the lability of the protein, which
Scale- Up in Stirred-Tank
Reactors
219
will dictate immediate downstream processing (i.e., concentration and purification), or storage at -80°C until further processing (see Notes 21 and 22). 1. The higher the postinfection cell viability, the greater amounts of intracellular protein that can be recovered. Harvesting can be done when the cell viability is 275%, and/or at 3-4 d postinfection, depending on when maximal protein concentrations occur. At cell viabilities below 75%, imminent product loss will occur owing to cell lysis and proteolytic degradation, if the product is very labile. 2. Similarly, membrane-associated proteins can be harvested using the same criteria for intracellular product harvesting. Cell viability will decrease rapidly during infection as recombinant proteins are incorporated into the cell membranes. Hence culture harvest may occur earlier than planned. 3. For secreted proteins, the supernatant can be collected at -75% viability or lower. However, only if the protein is sensitive to proteolytic degradation with increased cell lysis, should higher viability harvesting be executed. 1.
2.
3.
4. 5.
4. Notes In the original TNMFH recipe, pluronic F-68 was not employed. However, it has been shown that Sf!J cells, in spinner and/or bioreactor cultures, have prolonged, maximal viabilities compared to cultures devoid of pluronic F-68. It is the emulsifying agent used in protein-free IPL/41 (3,4). When serum-free cultures are inoculated too low (12.5 x lo5 cells/ml), growth tends to lag by 2 or 3 d probably because of overdilution of endogenous cellular growth factors in the inoculum carryover to the fresh medium (25). By contrast, serum-containing cultures can be seeded at lower densities because of growth factors and lipid content provided by the serum. It is essential to maintain a stringent subculture routine. It is imperative to ensure that the insect cell-culture reagents lots (i.e., FBS, ultrafiltered yeastolate) are screened in order to ensure consistent growth and production behavior. Once the reagents are acceptable, order large quantities to maintain consistencyfrom one experiment to another. Maintain stringent medium preparations in order to prevent haphazard variations in growth (i.e., cell selection) and production characteristics if commercial liquid media are not used. Stainless steel or polycarbonate Quick-Connects (Swagelok Canada Ltd., Niagara Falls, Ontario) are useful to aseptically connect bottles and accessories to the bioreactor, especially if a laminar flow hood is not in the proximity. Bottle filling can be done inside a laminar hood and can subsequently be emptied after the sterile quick-connections have been made, away from the hood.
220
Tom et al.
6. To ensure prolonged, reliable use of ancillary tubing, protective sleeves can be employed at the bioreactor ports. Often, tugging and pulling of the silicone lines will cause internal tears at these metal ports, thus causing unexpected and costly leaks. Use tubing (-l-cm length) 1 size smaller in diameter to make a snug fit at the ports. The designated line will make a tighter fit around the sleeved port, and it will be protected from internal tears caused by naked ports. 7. Change all microbial air vents on a regularly scheduled basis, depending on the manufacturer’s specifications. 8. An alternate leak test can be done by feeding gas at a low flow rate (-100 mL/ min) via a gas rotameter, through the bioreactor inlet while the outlet is clipped. If the bead in the rotameter falls down to 0, then the reactor is leak-proof. If after more than 5 min the bead has not fallen, then the leak source must be divulged according to the leak test method described earlier. 9. DO probe calibration can be done with the medium in the reactor following water removal. However, add the required amount of serum and pluronic F-68 after calibration to prevent foaming. 10. Antibiotics, such as gentamicin, should be used in large-scale cultures since the cost of manpower, medium and final product is substantial. 11. Although serum (10% v/v) provides some shear protection to the cells, it is not enough for large-scale cultures, especially if its concentration is diminished during infection. Protective agents (methyl-cellulose, polyvinyl alcohol, polyethylene glycol [19], or pluronic F-68 [0.1-0.25% w/v]), should be used for large-scale cultures. 12. Avoid underinoculation of the bioreactor (
Scale-Up
in Stirred-Tank
Reactors
221
14. As a diagnostic tool and an internal verification, control cultures should always be taken following bioreactor inoculation. This will indicate if toxins/contaminants are introduced by the bioreactor after inoculation or if the inoculation procedure was not aseptic. Small-scale controls should parallel the bioreactor growth and infection characteristics. 15. pH control is often overlooked, but if pH control is stringently controlled between 6.2 and 6.3, culture viability will be prolonged (ll), thus minimizing premature product degradation owing to cell mortality. By contrast, serum-free cultures do not require as much pH control, since their pHs have a tendency to decrease. 16. Certain guidelines should be emphasized when large-scale recombinant protein/virus production is being considered. Improper preparation, titration, and/or storage of virus stocks can lead to ineffective culture infections (Fig. 5D) and vastly reduced product yields. If cells continue to grow following infection, it is likely because of an inefficient infection and/or the virus titer was over-estimated (Fig. 5D). Virus stocks can be stored at 4OC for as long as 2 yr without significant loss in protein yields (2). Summers and Smith (2) suggest that viral plaque aliquots could be frozen at -8OOC only for maintaining stock reserves beyond the 2-yr limit. Minimize high passage virus utilization Virus sticks should be stored in the dark also (26). 17. Viral stocks can be prepared according to Summers and Smith (2), by infecting a concentrated culture at lo7 cells/ml, followed by resuspension to the appropriate density. However, when larger-than-spinner-sized volumes have to be infected, it can be cumbersome to centrifuge and concentrate large culture volumes for infection at lo7 cells/ml. If the infection density does not exceed 2-3 x lo6 cells/ml (3, I I), the required viral stock volume (510% culture volume) can be added directly to the cells as a batch mode production process (see Section 3.6.1.). 18. Cell-medium separation can take from 20 min (2.8-L culture) to 1 h (4-l 1 L culture), depending on the culture volume/density, presence of medium additives (i.e., PBS, pluronic, and so forth), membrane porosity, and pump speed which controls the transmembrane pressure. 19. If no cell/medium separator is available, change spent medium for fresh medium by centrifugation of the entire culture (-2OOg, 15-30 min, 4°C) and resuspend the resultant pellets in the fresh medium (serum-supplemented or serum-free). If the fresh medium is serum-free, use -0.2-0.25% (w/v) Pluronic F-6 for increased cell protection. 20. The cell/medium separator or centrifugation can be used to harvest the culture by culture concentration to separate the medium from the biomass.
Tom et al.
222
21. Make sure quantification and punficatron procedures work at all scales. Although purification and/or detection methods work for monolayer or spinner-grown cultures, it may not always work in larger-scale cultures where protective agents have been added (i.e., pluronic, antifoam). This point IS particularly important for secreted proteins. 22. When harvesting the recombinant protein from large quantities of cells or supernatant, insure adequate protease inhibitor protection, or freeze at -80°C as soon as possible rf the product is particularly labile. Small-scale samples are treated and analyzed more quickly while larger volumes need more time to be extracted; for labile proteins prolonged extraction and purification time can incur yield losses.
Acknowledgments The authors would like to thank M. T. Debanne and M. O’Connor for their EGFR-ED recombinant; D. C. Tessier and D. Y. Thomas for recombinant papain; D. Germain and T. Vernet for recombinant Kex; and M. Dennis, J. Labrecque, M. Caron, and P. Payette for their recom-
binant G proteins. We also thank Glenn Godwin of Gibco for donating to us his Sf-900DPM-adapted cells. From our own cell culture team, we also thank Charles B6dard for valuable support and opinions on the writing of this chapter.
References 1. Griffiths, B. (1992) Closing the culture gap. Bio/Technology 10,3 l-32. 2. Summers, M. D. and Smith, G. E. (1987) A Manual ofMethodsfor Baculovirus Vectors and Insect Cell Culture Procedures. Texas Agricultural Experiment Station, Bulletin No. 1555. 3. Maiorella, B., Inlow, D., Shauger, A., and Harano, D. (1988) Large-scale insect cell-culture for recombinant protein production. Bio/Technology 6, 1406-1410. 4. Inlow, D., Shauger, A., and Maiorella, B. (1989) Insect cell culture and baculovirus propagation in protein-free medium. .I, Tiss. Cult. Methods 12(l), 13-16. 5 Weiss, S. A., Smith, G., Kalter, S. S., and Vaughn, J. L. (1981) Improved method for the production of insect cell cultures in large volume. In Vitro 17,495-502. 6. Vaughn, J. L. (1971) Design of cell culture media, in Invertebrate Tissue Culture, vol. 1 (Vago, C., ed.), Academic Press, New York, pp. 3-40. 7. Kamen, A. A., Tom, R. L., Caron, A. W., Chavarie, C., Massie, B., and Archambault, J. (1991) Culture of insect cells in a helical ribbon impeller bioreactor. Biotechnol. Bioeng. 38,619-628. 7a. Kamen, A. A. and Tom, R. (1994) Mass spectrometry determination of insect cell respiration rates in culture and productton processes, in Animal Cell Technology Products of Today Prospects for Tomorrow (Spier, R. E., Griffiths, J. B., and Berthold, W., eds.), Butterworth/Heinmann, Toronto, pp. 345-350.
Scale-Up
in Stirred-Tank
Reactors
223
8. Hink, W. F., Ralph, D. A., and Joplin, K. H. (1985) Metabolism and characterization of insect cell cultures, in Comprehensive Insect Physiology, Biochemistry and Pharmacology, vol. 10 (Kerkut, G. A. and Gilbert L. I., eds.), Pergamon Press, Oxford, pp. 547-570. 9. Hink, W. F. (1970) Established insect cell line from the cabbage looper, Trichoplasia
ni. Nature 226,466-467.
10. Kamen, A. A., Chavarie, C., Andre, G., and Archambault, J. (1992) Design parameters and performance of a surface baffled helical ribbon impeller bioreactor for the culture of shear sensitive cells. Chem. Eng. Sci. 47(9-ll), 2375-2380. 11. Caron, A. W., Archambault, J., and Massie, B. (1990) High-level recombinant protein production in bioreactors using a baculovirus-insect cell expression system. Biotechnol. Bioeng. 36, 1133-l 140. 12. Broussard, D. R. and Summers, M. D. (1989) Effects of serum concentration and media composition on the level of polyhedrin and foreign gene expression by baculovirus vectors. J. Invert. Pathol. 54, 144-150. 13. Wang, P., Granados, R. R., and Shuler, M. L. (1992) Studies on serum-free culture of insect cells for virus propagation and recombinant protein production. J. Invert. Pathol. 59,46-53.
13a.Licari, P. and Bailey, J. R. (1991) Factors influencing recombinant protein yields in an insect cell-baculovirus expression system: multiplicity of infection and intracellular protein degradation. Biotechnol. Bioeng. 37,238-246. 14. Wickham, T. J., Davis, T., Granados, R. R., Hammer, D. A., Shuler, M. L., and Wood, H. A. (1991) Baculovirus defective interfering particles are responsible for variations in recombinant protein production as a function of multiplicity of infection. Biotech. Lett. 3(7), 483-488. 15. Kool, M., Voncken, J. W., Van Lier, F. L. J., Tramper, J., and Vlak, J. M. (1991) Detection and analysis of Autographa californica nuclear polyhedrosis virus mutants with defective interfering properties. Virology 183,739-746. lSa.Matsuura, Y., Possee, R. D., Overton, H. A., and Bishop, D. H. L. (1987) Baculovirus expression vectors: the requirements for high level expression of proteins, including glycoproteins. J. Gen. Viral. 68, 1233-1250. 15b.Vialard, J., Lalumibre, M., Vernet, T., Briedis, D., Alkhatib, G., Henmng, D., Levin, D., and Richardson, C. (1990) Synthesis of the membrane fusion and hemagglutinin proteins of measles virus, using a novel baculovirus vector containing the P-galactosidase gene. J. Viral. 64(l), 37-50. 16. Tramper, J., Williams, J. B., and Joustra, D. (1986) Shear sensitivity of insect cells in suspension. Enzyme Microb. Technol. 8,33-36. 16a. Streett, D. A. and Hink, W. F. (1978) Oxygen consumption of Trichoplusia ni (TN-368) insect cell line infected with Autographa califomica nuclear polyhedrosis virus. J. Invert. Pathol. 32, 112-l 13. 16b.Schopf, B., Howaldt, M. W., and Bailey, J. E. (1990) DNA distribution and respiratory activity of Spodoptera frugiperda populations infected with wild-type and recombinant Autographa californica nuclear polyhedrosis virus. J. Biotechnol. 15, 169-186.
224
Tom et al.
17. Murhammer, D. W. and Goochee, C. F. (1988) Scaleup of insect cell cultures: Protective effects of pluronic F-68. Biotechnology 6, 1411-1418. 18. Bliem, R., Konoptsky, K., and Katinger, H. (1991) Industrial animal cell reactor systems: Aspects of selection and evaluation in, Advances in Biochemical Engineering Biotechnology, vol. 44 (Fiechter, A., ed.), Springer-Verlag, New York, pp. 1-26. 19. Papoutsakis, E. (1991) Media additives for protecting freely suspended animal cells against agitation and aeration damage. Tibtech. 9,3 16-324. 20. Caron, A. W., Tom, R. L., Kamen, A. A., and Massie, B. (1994) Baculovirus expression system scaleup by perfusion of high-density Sf-9 cultures. Biofechnol. Bioeng. 43,88 l-89 1. 20a.Bedard, C., Tom, R., and Kamen, A. (1993) Growth, nutrient consumption and end-product accumulation in Sf-9 and BTI-EAA insect cell cultures: insights into growth limitation and metabolism. Biofechnol. Prog. 9,615-624. 21. Tessier, D. C., Thomas D. Y., Laliberte, F., Khi, H. E., and Vemet, T. (1991) Enhanced secretion from insect cells of a foreign protein fused to the honey bee melittin signal peptide. Gene 98, 177-183. 22. Labrecque, J., Caron, M., Torossian K., Plamandon, J., and Dennis, M. (1992) Baculovirus expression of mammalian G protein a subunits. FEBS 304(2,3), 157-162. 23. Germain, D., Vernet, T., Boileau, G., and Thomas, D. Y. (1992) Expression of the Saccharomyces cerevisiae Kex2 endoprotease in insect cells: evidence for a Cterminal autoprocessing event. Eur. J. Biochem. 204, 121-126. 24. Grace, T. D. C. (1962) Establishment of four strains of cells from insect tissues grown in vitro. Nature 195,788,789. 25. Roberts, P. L. (1984) Growth of insect cells in recycled medium and the use of various serum supplements. Biotechnol. L&t. 6(10), 633-638. 26. Jarvis, D. L. and Garcia, Jr., A. (1994) Long-term stability of baculoviruses stored under various conditions. BioTechniques 16(3), 508-513.
CHAPTER13 Expression of Tissue Inhibitor of Metalloproteinases by Recombinant Baculovirus-Infected Insect Cells Cultured in an Airlift Fermentor Laurie K Overton, Indravadan Pate& J. David Becherer, Gyan Chandra, and Thomas A. Kost 1. Introduction The recombinant baculovirus expression system, developed in the laboratories of Summers (I) and Miller (2), has been widely used for the production of heterologous proteins (3,4). In this system, the gene to be expressed is cloned into a plasmid transfer vector, downstream of a strong baculovirus gene promoter, and cotransfected with wildtype virus DNA into insect cells. An in vivo homologous recombination between viral DNA and the specific segment of the shuttle vector containing the promoter and heterologous gene occurs at a low frequency. This recombination leads to the production of a small population of recombinant viruses carrying the foreign gene under the control of the viral polyhedrin promoter. These recombinant viruses can be readily isolated from the wildtype population by plaque purification. The infection of insect cells with the recombinant virus results in the production of the foreign gene product. As described in recent reviews (34, the baculovirus system possesses many advantages for the expression of heterologous proteins. A major advantage is that the insect cell host can be readily cultured in medium From: Methods in Molecular Biology Vol. 39: Baculovirus Expression Protocols Edited by: C. D. Richardson Q 1995 Humana Press Inc., Totowa, NJ
225
226
Overton
et al.
in either anchorage-dependent or suspension culture. Until recently, application of this system has been limited to the small-scale culture of insect cells in tissue-culture flasks or small spinner-culture vessels. Relatively few studies have been reported describing the scale-up of insect cell-culture systems for the production of recombinant proteins. Maiorella et al. (7) have described the growth of the Spodoptera frugiperda cell line (Sf9) in 3-L spinner flasks and a 21-L airlift fermentor. The authors were able to obtain cell densities of 5 x lo6 cells/ml in the airlift fermentor. Infection of the culture with a recombinant baculovirus expressing human-macrophage colony-stimulating factor (M-CSF) resulted in the production of 40 mg/L of recombinant M-CSF. In another study (8), Sf9 cells cultured in a 570~mL airlift fermentor were used to study the production of P-galactosidase following infection with a recombinant virus expressing this gene. Godwin et al. (9) have described the production of human recombinant erythropoietin following the infection of Sf9 cells cultured in a 5-L airlift fermentor. The culture and infection of Sf9 cells in a helical ribbon impeller reactor have recently been described (10). A recombinant VP6 protein was produced at approx 35 l.t,g/106cells. This yield was comparable to that obtained by infecting cells grown in small spinner flasks or Petri dishes. As opposed to batch-type culture processes, insect cells have also been cultured in continuous (11-13) and semicontinuous culture systems (14). Cavegn et al. (15) have recently reported using such a system for producing the extracellular domain of endothelial leukocyte adhesion molecule and the insulin receptor tyrosine kinase domain from recombinant virus-infected Sf9 cells. This chapter is aimed at extending the scale-up experience available for the production of recombinant proteins by describing our experience using a 5-L airlift fermentor for the culture and infection of Sf9 cells. We have successfully used this system to produce recombinant human type 1 tissue inhibitor of metalloproteinases (TIMP). Metalloproteinases are a family of enzymes with proteolytic activities for various portions of the extracellular matrix (I 6,17). TIMP, a ubiquitous glycoprotein with a mol wt of 28.5 kDa (17-19), inhibits metalloproteinases, such as collagenase and several others involved in connective tissue catabolism (20,21). TIMP may play an important role in metastases (17), and in the progress of such diseases as rheumatoid arthritis or periodontal disease (18).
TIMP
Expression
in an Airlift
Fermentor
227
2. Materials 1. The Sf9 cell line can be obtained from the American Type Culture Collection, Rockville, MD. The cells were maintained in Grace’s medium (GIBCO), supplemented with 3.3 pg/mL lactalbumin hydrolysate (Difco), 3.3 pg/mL yeastolate (Difco), 10% fetal bovine serum (Hyclone), 50 l.tg/ mL gentamicin (GIBCO) and 0.1% Pluronic F-68 (GIBCO), in 125mL spinner flasks (Techne). 2. For scale-up, cells are grown in Ex-Cell401 medium (JRH Biosciences) with 50 pg/mL gentamicin. 3. Transfections are done, according to the manufacturer’s instructions, using a mammalian transfection kit (calcium phosphate) from Stratagene. 4. A 5-L airlift fermentor can be obtained from Porton Instrument&H Fermentation (Hayward, CA). The dissolved oxygen and service control module are from Porton also. The dissolved oxygen and pH probes come from Ingold (Wilmington, MA) . 5. The Spin-Trex unit for concentration of medium comes from New Brunswick. 6. Leur-lock fittings are supplied by Value Plastics. Silicone tubing (l/4- and 3/16-m. inner diameters) is used throughout the work except for PharMed tubing (Cole-Palmer, Chicago, IL), which is used on the air lines. 7. Nitroblue tetrazolium and S-bromo-4-chloro-3-indolylphosphate are from Promega (Madison, WI). Megga Block III is obtained from Onasco. Restriction and modifying enzymes are from BRL (Gaithersburg, MD). 8. The shuttle plasmid, pJVPlOZ can be obtained from the laboratory of Chris Richardson (22) or Invitrogen (San Diego, CA). 9. pGEM-4Z is supplied by Promega (Madison, WI) and pAGS3 can be obtained from Jun-ichi Miyazaki (23). 10. Tris-buffered saline/0.05% Tween 20 (TBS-T). 11. Alkaline phosphatase-conjugated goat antirabbit (1:7500 dilution). 12. 100 mM Tris-HCl, pH 9.5’100 mM NaCl, and 5 mM MgC12. 13. Sephacryl5-100 HR column (26 mm x 100 cm) equilibrated in 50 m&Z HEPES, and 50 mM NaCl, pH 7.5. 14. Endoglycosidase F (Boehringer-Mannheim, Mannheim, Germany). 15. 100 mA4 Sodium phosphate, 50 m&4 EDTA, and 0.5% NP-40, pH 6.1. 16. Synthetic thioester substrate Ac-Pro-Leu-Gly-thioester-Leu-Gly-OEt (Glaxo, Research Triangle, NC). 17. 5,5’-dithio-bis[Znitrobenzoic acid] (DTNB) (Pierce Chemical, Rockford, IL). 18. 100 pL of 15 tipurified collagenase (Sigma, St. Louis, MO). 19. 20 n-&YTris-HCl, 200 mM NCl, 5 mA4CaCl,, 10 & Zn2+,and0.01% Brij-35. 20. Millex FG50 and Millipak 60 0.22~pm filters (Millipore, Bedford, MA).
Overton et al.
228
3. Methods There are a number of steps that must be performed prior to scaling up the production of TIMP or any other recombinant protein in a S-L airlift fermentor. This section will describe the cloning of the TIMP gene into the transfer plasmid, isolation and scale-up of recombinant virus, detection of TIMP by immunoblot analysis, production of TIMP in a 5-L airlift fermentor, purification of TIMP from the culture medium, and a biological activity assay for TIMP. 3.1. Construction of Transfer Plasmids 1, It was of interest to determine if a recombinant virus carrying two copies of the TIMP gene would produce more TIMP than a recombinant virus containing only a single copy. The TIMP gene was isolated from a human lung cDNA library cloned in h gtl 1using oligonucleotide probes designed from the published TIMP sequence (19). 2. An EcoRI fragment containing TIMP was isolated and subcloned into the EcoRI site of pGEM-4Z. This plasmid was digested with Hi&III, and the TIMP fragment was isolated and subcloned into the W&III site of pAGS3. 3. For subcloning into the transfer vector, the &z&II TIMP fragment from pAGS3 was blunt-ended with T4 DNA polymerase, NheI linkers were added, and the fragment was cloned into the NheI site in pBR322. The NheI fragment from pBR322 was isolated and cloned into the NheI site in pJVPlOZ (Fig. 1) to create pPHT (Fig. 1B). 4. Another construct was made that contained a second copy of the TIMP gene in place of the ZucZ gene derived from pJVP10. pPHT was digested with XbaI and blunt-ended with T4 DNA polymerase. Sac1 linkers were added and the plasmid was digested with SacI, An ZVheITIMP fragment was isolated and end-filled with T4 DNA polymerase. Sac1 linkers were ligated to the TIMP fragment, and the gene was cloned into the Sac1 site creating pPHPlOT (Fig. 1A). 5. A third construct was engineered with TIMP under the control of only the PlO promoter by digesting pPHPlOT with ZVheIand religating the plasmid to create pPlOT (Fig. 1C). The recombinant viruses derived from these transfer plasmids are referred to as PHT, PHPlOT, and PlOT.
3.2. Isolation
and Scale-Up
of Recombinant
Virus
1. Recombinant virus can be produced and plaque-purified according to the methods of Summers and Smith (2#), and is rigorously described in Chapter 9 of this book. The expression of P-galactosidase by recombinant virusinfected cells results in the production of easily discernible recombinant plaques. The scale-up of the virus stock is described below.
TIMP
Expression
in an Airlift EWE II lzw7
Fermentor
229
, H’“d “1 0
Fig. 1. Three shuttle plasmids were constructed from the parent plasmid pJVP1OZ. A TIMP fragment with NheI ends was constructed as described in Section 3. The TIMP gene was cloned into the NheI cloning site in pJVPlOZ to create pPHT (B). The ZucZ gene was removed from pPHT, a Sac1site was then engineered, and another copy of TIMP was cloned into the Sac1site to create pPHPlOT (A). pPHPlOT was digested with NheI to remove a single copy of TIMP and the plasmid was relegated to obtain pPlOT (C).
230
Overton et al.
2. A recombinant plaque in an agarose plug is isolated and placed into one mL of Ex-Cell401 medium. The tube is vortexed for one min or the virus is allowed to elute from the agarose plug overnight. 3. Flasks (25-cm2) are seeded with 4 x lo6 cells. Following cell attachment, the medium is removed, one mL of virus solution is added, and the flask is incubated on a rocking platform for 1 h at room temperature. The virus inoculum is removed from the cells and replaced with 4 mL of Ex-Cell401 medium. The infected cultures are incubated for 3 or 4 d. 4. The medium is collected and 1 mL of virus inoculum is added again to a 25-cm2 flask. The medium is collected after 3 or 4 d. A 225-cm2 flask is seeded with 5 x lo7 cells, and 4 mL of virus are added to the flask. The virus is absorbed to the cells for 1 h. The inoculum is removed, and 35 mL of Ex-Cell401 medium are added. Virus is harvested 3 or 4 d later and used to infect five 225-cm2 flasks as before. The 175 mL of virus that is collected from these flasks is used as the inoculum for the 5-L airlift fermentor. 5. The virus titer was determined by plaque dilution assay as described in Chapter 9. The concentration of virus is normally approx l-2 X lo* PFWmL.
3.3. Detection
of TIMP
by Immunoblot
A time-course can be performed with the different recombinant viruses to determine which expresses the highest level of TIMP. 1. For the immunoblot analysis, 25-cm2 flasks are seeded with 4 x lo6 cells and infected with one mL of virus stock solution for 1 h at a multiplicity of infection (MOI) of 10. The virus inoculum is removed, and 4 mL of fresh medium are added. The supernatant is collected at 24, 48, 72, and 96 h postinfection. 2. Ten microliters of culture medium from each sample are added to 10 ~,LLof 2X Laemmli sample buffer (25) and boiled for five min. The proteins are separated by electrophoresis through a lo-20% SDS-polyacrylamide gel. 3. Following electrophoresis, the proteins are electroblotted onto a nitrocellulose membrane, and nonspecific binding sites are blocked by incubating the blot for 1 h at room temperature in Megga Block III. The blot is rinsed with Tris-buffered saline/0.05% Tween 20 (TBS-T) and incubated for 1 h at room temperature with a polyclonal anti-TIMP antibody diluted in TBS-T. 4. The blot is rinsed with TBS-T and then incubated with a 1:7500 dilution of alkaline phosphatase-conjugated goat antirabbit antibody for 1 h at room temperature. 5. The blot is rinsed with TBS-T and developed for 5 min in a solution of 100 mM Tris-HCl, pH 9.5, 100 mM NaCl, and 5 mit4 MgCl, containing the substrates nitroblue tetrazolium and 5-bromo-4-chloro-3-indolylphosphate. An immunoblot is shown in Fig. 2.
TIMP
Expression
in an Airlift
Fermentor
231
30 -
21 -
14 Fig. 2. Immunoblot analysis of TIMP from recombinant baculovirus-infected Sf9 cells. Three recombinant viruses (PHT, PlOT, and PHPlOT) were used to infect Sf9 cells cultured in 25-cm2 flasks. Ten-microliter aliquots of medium, collected 24,48,72, and 96 h postinfection, were electrophoresedthrough a lO-20% SDS-polyacrylamide gel and transferred to a nitrocellulose membrane. TIMP was detected by reacting the membrane with a polyclonal anti-TIMP antibody. 6. We find that PHT produces as much secreted TlMP as PHPlOT, and PlOT does not result in recombinant virus stocks with higher yields than PHT. For these reasons, PHT is chosen for scale-up. 3.4. Assembly, Preparation, of the 5-L Airlift
and Sterilization Fermentor
In our laboratory, the scale-up of insect cell culture is performed in a 5-L airlift fermentor operated in batch mode rather than continuous or perfusion production. This is owing to the lytic-type infection produced by baculoviruses in insect cells. A schematic representation of the 5-L airlift fermentor (together with its upper head plate), which is marketed by LH Fermentation is shown in Fig. 3. The apparatusconsists of a tem-
Overton et al.
232
headplate
vessel
-
draft tube (chimney)
-
temperature control jacket
---
sparging
support
-silicone
outlet stopper
baseplate
Fig. 3. Schematic view of a 5-L airlift fermentor designed and marketed by LH Fermentation. Components of the fermentor (A) and the arrangement of connections to the head plate at the top of the apparatus (B) are indicated. perature-control jacket connected to a circulating water bath, an inner draft tube through which the suspended cells rise and subsequently fall downward between its outer wall and the inner wall of the fermentor, a vessel head plate with 12 openings, a silicon rubber stopper for draining, and a l/4 in. stainless tube for sparging at the bottom of the growth chamber.
TIMP Expression
in an Airlift
233
Fermentor
0 7
0 13
ITEM
DESCRIPTION
ITEM
DESCRIPTION
1
AI
out
9
Closed
2
Air out
10
Closed
3
Closed
11
Closed
4
Closed
12
Closed
5
Bung closure
13
Sample
6
Closed
14
Closed
7
02 Electrode
15
Top plate
6
Media
16
Vessel
port
addition
clamp
retaining
and screws screw
Fig. 3. (continued). 1. To prepare the fermentor for autoclaving, the draft tube is inserted into the vessel, and the head plate is attached; the l/4 in. stainless-steel tube is inserted through the rubber stopper leaving 3 in. outside the vessel that are attached to a check valve, and then a Millex FG50 0.22~pm filter. This inlet is connected to oxygen and air for oxygenation and sparging purposes. 2. Silicon tubing is attached to the other stainless-steel bottom harvest port. For autoclaving, the tubing is coiled and wrapped in aluminum foil. 3. Next, the head plate is assembled (see Fig. 3). For each closed port a stainless-steel septum with O-ring is screwed onto the headplate. The ends of the tubing are covered with aluminum foil. An oxygen electrode and a pH electrode (if necessary) are placed in the head plate.
234
Overton
et al.
4. The ends of all tubing are covered with a milk filter and wrapped with aluminum foil to prevent contamination of the growth chamber following sterilization. Clamps were put on all tubing Inlets and outlets, except for one air vent, in order to allow for expansion of air inside the fermentor during sterilization. 5. The assembled fermentor is autoclaved at 123°C for 60-70 min.
1. 2. 3.
4.
5.
3.5. Cultivation of Sfs Cells in a 5-L Airlift Fermentor Immediately after autoclaving, the vessel is removed, and air is put mto the vessel through the filter connected to the sparging port so that ambient air will not be drawn into the vessel as it cools. Instrumentation for growing Sf9 cells in the airlift bioreactor consists of the service module and the dissolved oxygen (DO) module. The service module blends gassesand monitors gas flow. The vessel is allowed to cool and Ex-Cell401 culture medium is transferred into the vessel through a 0.22~pm Millipak 60 media filter connected to the media addition tube, Add media until the bottom of the DO probe is covered. Clamp the medium addition tubing and maintain the sterility of its end (see Note 2). Before adding the cells, the DO probe is calibrated. Air is bubbled through the media via a 0.22~pm filter for several hours or overnight; the DO meter is set at 100%. The sparge rate is maintained at about 150 mL/ min. Mixing is accomplished by sparging air into the draft tube of the fermentor through the quarter-inch stainless-steel Y tube attached to check valves at the bottom of the fermentor. The flow rate is adjusted to 150 mL/min. The air is filtered with two 0.22-p filters in tandem in each gas line. Oxygen is pulsed into the sparge air, as needed, to keep the oxygen at 50% air saturation. The large size of the bubbles delivered through the sparge tubes results in poor oxygen transfer; however, initially we had trouble utilizing a sparger with small bubbles because of increased sheer stress on the cells. The temperature is maintained at 28°C by circulating water through a water jacket surrounding the glass vessel. In Fig. 4, a photograph of assembled 5-L airlift fermentors is shown. The complete setup stands approx 4 ft high with the probes attached. Sf9 cells have previously been adapted to serum-free medium; the cells were seeded in Ex-Cell401 medium (100 mL in a 250-n& spinner flask) at a density of 1 x lo6 cells/ml and split when the density reached 4 x lo6 cells/ml. The cells are cultured in this medium for about 3 wk before they were scaled up for inoculation mto the fermentor. The cells are transferred into a volume of 400-500 mL of Ex-Cell 401 medium in a 1-L
TIMP
Expression
in an Airlift
Fermentor
235
Fig. 4. Photograph showing scale-up of Sf9 cells in a 5-L airlift fermentor for the infection by recombinant TIMP virus. spinner at a concentration of 1 x lo6 cells/ml. The cells are then inoculated into the fermentor when they have reached a cell density of about 3.6 x lo6 cells/ml to give a final concentration of 3.5 x lo5 cells/ml (see Note 3). 6. The cells are aseptically loaded into the fermentor by gravity flow through the cell-addition port. The pH of the culture is initially 6.3 and usually drops to 5.95 by the end of the run. Because of this minimal pH change, the pH is usually not controlled for production runs. 7. The cells experience a l-2 d lag period, depending on the physiological state of the seed cells, before entering into exponential growth (see Note 3). The health of the cells at the time of infection is very important for high-level gene expression and virus production. In Fig. 5, a growth curve of uninfected cells is shown. If the cells were not infected, they grew to a density of about 1 x lo7 cells/ml. 8. The cells in logarithmic growth double in 24-26 h until a density of about 5 x lo6 cells/mL is attained. At this density, the cells were about 98% viable; however, viability decreased above 5 x lo6 cells/ml. Sf9 cells are normally infected at a density of 2-5 x lo6 cells /mL.
236
Overton et al.
0
2
4
6
6
10
12
Days
Fig. 5. Growth of uninfected St9 cells in a 5-L airlift fermentor. The cells were grown in Ex-Cell401 medium containing 50 l.&/mL gentamicin. Air flow was set at 150 mL/min, oxygen was fed into the sparging gas to keep dissolved oxygen at 50% air saturation, pH was not controlled, and temperature was maintained at 28OCby circulating water through the jacket of the fermentor. 9. In Fig. 6, growth curves for three production runs of TIMP are shown. The cells were infected at MOIs of 3, 10, and 20 PFU of recombinant virus/ cell. Once infected, the cells stop dividing as can be seen by the leveling off of the growth rate. We routinely harvested the TIMP at 72-96 h postinfection and have observed little variation in the levels of TIMP produced by cultures infected at the different MOIs.
3.6. Purification of TIMP from Sfs Cell Culture Medium 1. Cells are pelleted by centrifugation at lOOOg,and the culture medium is concentrated approx lo-fold using a SpinTrex concentrator fitted with a 10,000 Dalton mol-wt cut-off filter. 2. Virus is pelleted at 20,OOOgfor 2 h and prepared for future inoculums. The virus pellets from 5 L of culture medium are resuspended in 300 mL of Ex-Cell 401 medium and filter-sterilized through a 0.45~pm membrane. The virus titer is determined by plaque dilution assay and is usually about 3 x log PFU/mL. There have been reports of defective interfering (DI)
TIMP Expression
in an Airlift
237
Fermentor
time of infection
+-
0 mol = 3
Ammoi= 0 mol=20
0
2
4
6
a
10
12
Days Fig. 6. Growth of Sf9 cells infected at different MOIs in a 5-L airlift fermentor. Cells were grown in Ex-Cell401 medium containing 50 pg/rnL gentamicin. Dissolved oxygen was kept at 50% air saturation, and pH was not controlled. The d of infection is noted by an arrow. Cultures were harvested 72-96 h postinfection. particles formed upon high passage number of virus stocks (26). A decrease in protein production, owing to the formation of DI particles has not been observed in our system, since we have only used secondpassage virus stocks for infection. From a single 5-L airlift fermentor run, enough virus is obtained to infect eight fermentors. 3. Following removal of the virus, the concentrated supernatant is dialyzed against two 5-L changesof 20 mM Tris-HCl, pH 7 (buffer A), and the sample is centrifirged in a Sorvall SA-600 rotor at 1OOOgto remove insoluble material. 4. The resulting supernatant is then applied to Q-sepharose and S-sepharose fast-flow columns connected in series at a flow rate of 6 mL/min. The Q-sepharose flowthrough contains the original TIMP activity, but the S-sepharose flowthrough material does not. The columns in series are disconnected, and TIMP is eluted from the S-sepharose with eight column volumes of a O-500 mM ammonium sulfate gradient in buffer A. TIMPcontainmg fractions, identified by inhibitory activity of collagenase as well as immunoblot analysis using polyclonal anti-TIMP antibody, are pooled, diluted fivefold with buffer A, and reapplied to a Mono-S HRlO/lO col-
Overton et al.
238
MW ( kW - 97 - 67 - 45 - 31 - 21
-14
Fig. 7. Coomassie-stained SDS gel showing TIMP after each step in the purification process. umn equilibrated in buffer A. TIMP is eluted from the Mono-S column with a 25 column volume gradient spanning 0-0.2M ammonium sulfate. 5. Pooled TIMP activity is then concentratedand applied onto a Sephacryl S- 100 HR column (26 mm x 100 cm) equilibrated in 50 mM HEPIZS, and 50 mM NaCl, pH 7.5, and run at a flow rate of 0.5 mL/min. TIMP activity is pooled, concentrated in a Centriprep 10 ultrafiltration unit, and stored at -70°C. 6. The TIMP-containing fractions from each step in the purification process are electrophoresed through a SDS-polyacryamide gel and stained with Coomassie blue (see Fig. 7). At the end of the purification, the TIMP is
TIM.. Expression
in an Airlift
Inhibition 100
239
of collagenase and 19 kDa truncated collagenase by TIMP
Assay 1
“I
Fermentor
1 2nMEnzvme
1
,~I
05
10
20
I
I
I
40
80
16
I
32
I
63
I
125250
I
I
500
[TIMP], nm
Fig. 8. The biological activity of TIMP was demonstrated in a collagenase chromogenic substrate assay. TIMP inhibits collagenase and a 19 kDa truncated form of collagenase, in a dose-dependent manner. >98% pure. On average, 2540 mg of purified TIMP can be obtained from a 5-L airlift fermentor run. 7. TIMP has two potential N-linked glycosylation sites. Analysis of the recombinant TIMP reveals three major glycosylated forms of TIMP (Fig. 7). The carbohydrate moieties were removed by treatment of TIMP with endoglycosidase F at a ratio of 10 U/mg protein in 100 mM sodium phosphate, 50 mM EDTA, and 0.5% NP-40, pH 6.1, for 20 h at 37°C. The endoglycosidase-treated TIMP is shown in Fig. 7 and has the expected mol wt of deglycosylated TIMP of about 22 kDa.
3.7. Irhiibition
of CollagenaBe
by TIMP
1. Inhibition of collagenase activity is monitored using either full-length or truncated recombinant type I collagenase and the synthetic thioester substrate Ac-Pro-Leu-Gly-thioester-Leu-Gly-OEt (27) which in turn reacts with 5,5’-dithio-bis[Znitrobenzoic acid] (DTNB) following hydrolysis at the Gly-thioester-Leu linkage. This forms a basis for a calorimetric assay with collagenase. 2. In a 24-well microtiter plate, 100 p,L of TIMP at various concentrations were added to 100 p.L of 15 nM purified collagenase and incubated for 15 min at room temperature. The reaction was initiated by adding 100 pL of 300 mA4 DTNB and 300 PM thioester substrate. The reaction was moni-
Overton et al.
240
tored during conversion of the first 10% of substrate to product by the change per minute in absorbance at 405 nm. All assayswere performed in 20 mA4 Tris-HCI, 200 mM NaCl, 5 mM CaCl,, 10 pM Zn2+, and 0.01% Brij-35. In Fig. 8, the inhibition of collagenase by TIMP in a chromogenic substate assay is shown.
4. Notes 1. At times, the 5-L airlift fermentor was under pressure because of moisture on the vent filter. This problem was alleviated by using two filters, not in tandem. The line to one filter was closed and only opened if a pressure build-up was noticed. Also, we have used glasswool-plugged filters and have not had a problem with moisture in these filters. 2. At the present time, we are adding 0.5% fetal bovine serum to the culture medium in the fermentor. The cells look healthier, the run can go an extra d becausethe cells do not die asquickly, and there appears to be an increase in protein yield. Foaming is not a problem with the addition of this amount of serum. 3. The initial cell density in the fermentor is very important. We have experienced extended lag periods with densities <2 x lo5 cells/ml. At the present time we are seeding the fermentor at 3.5 x 1O5cells/ml and have decreased the total run time to 7 d. 4. We are also attempting to scale up protein production in 15-L Bellco and Wheaton stirred flasks. The flasks were fitted with an oxygen probe, and oxygen was maintained at 50% air saturation by sparging in oxygen through a stainless-steel or sintered glass tube. The cells were stirred at 70 rpm. TIMP has been produced in this system, but our initial run yielded a lower protein level than that produced using the 5-L airlift fermentor. In our initial run, cell growth was not optimum because of clumping of cells. Very little clumping occurred in the airlift fermentor. We have recently tried inoculating the 15-L vessels with cells grown in a 5-L airlift fermentor. Using this approach, the clumping problem was avoided. We have obtained cell densities of 2 x lo6 cells/ml and are working on optimizing conditions to improve cell densities. 5. We have also tried increasing protein production by using the Trichoplusia ni cell line, Tn5B14. This cell line is available from Invitrogen and is marketed as High 5 cells. These cells were infected with PHT in a 25-cm2 flask, and a 5- to lo-fold increase m protein production was observed as compared to that produced by infected Sf9 cells. The Tn5B14 cells clumped and grew poorly in suspension, We have not yet attempted to scale them up in the airlift fermentor.
TIMP Expression
in an Airlift
Fermentor
References 1. Smith, G. E., Summers, M. D., and Fraser, M. J. (1983) Production of human beta interferon in insect cells infected with a baculovirus expression vector. Mol. Cell. Biol. 3,2156-2165.
2. Pennock, G. D., Shoemaker, C., and Miller, L. K. (1984) Strong and regulated expression of Escherichia coli P-galactosidase in insect cells with a baculovirus vector. Mol. Cell. Biol. 4,399-406. 3. Luckow, V. A. and Summers, M. D. (1988) Trends in the development of baculovirus expression vectors. Biotechnology 6,47-55. 4. Luckow, V. A. (1991) Cloning and expression of heterologous genes in insect cells with baculovirus vectors, in Recombinant DNA Technology and Applications (Prokop, A., Bajpai, R. K., and Ho, C., eds.), MC Graw-Hill, New York, pp. 97-170. 5. O’Reilly, D. R., Miller, L. K., and Luckow, V. A. (1992). Baculovirus Expression Vectors. A Laboratory Manual. W. H. Freeman and Company, New York, NY. 6. Maeda, S. (1989) Expression of foreign genes in insects using baculovirus vectors. Ann. Rev. Entomol. 34,351-372.
7. Maiorella, B., Inlow, D., Shauger, A., and Harano, D. (1988) Large-scale insect cell-culture for recombinant protein production. Biotechnology 6,1406-1410. 8. Murhammer, D. W. and Goochee, C. F. (1988) Scaleup of insect cell cultures: Protective effects of pluronic F-68. Biotechnology 6, 1411-1418. 9. Godwin, G., Belisle, B., De Giovanni, A., Kohler, J., Gong, T., and Wojchowski, D. (1989) Serum-free growth and recombinant EPO expression in Spodoptera frugiperda (Sf9) insect cells. In Vitro 25(3), 19A. 10. Kamen, A. A., Tom, R. L., Caron, A. W., Chavarie, C., Massie, B., and Archambault, J. (1991) Culture of insect cells in a helical ribbon impeller bioreactor. Biotechnol. Bioeng. 38,619-628. 11. de Gooijer, C. D., van Lier, F. L. J., van den End, E. J., Vlak, J. M,, and Tramper, J. (1989) A model for baculovirus production with continuous insect cell cultures. Appl. Microbial.
Biotechnol. 30,497-501.
12. Kompier, R., Tramper, J., and Vlak, J. M. (1988) A continuous process for the production of baculovirus using insect-cell cultures. Biotechnol. Left. 10,849-854. 13. van Lier, F. L. J., van den End, E. J., de Gooijer, C. D., Vlak, J. M., and Tramper, J. (1990) Continuous production of baculovirus in a cascade of insect-cell reactors. Appl. Microbial.
Biotechnol. 33,43-47.
14. Hink, W. F. (1982) Production of Autographa californica nuclear polyhedrosis virus in cells from large-scale suspension cultures, in Microbial and Viral Pesticides (Kurstak, E., ed.), Marcel Dekker, New York, pp. 493-506. 15 Cavegn, C., Blasey, H. D., Payton, M. A., Allet, B., Li, J., and Bernard, A. R. (1991) Expression of recombinant proteins in high density insect cell cultures. 1lth ESACT Meeting, Brighton. 16. Webb, Z. (1982) Degradation of collagen, in Collagen in Health Disease (Weiss, J. B. and Jayson, M. I. V., eds.), Churchill-Livingstone, New York, pp. 121-134. 17. Alvarez, 0. A., Carmichael, D. F., and DeClerck, Y. A. (1990) Inhibition of collagenolytic activity and metastasis of tumor cells by a recombinant human tissue mhibitor of metalloproteinases. .I. Natl. Cancer Inst. 82,589-595.
242
Overton et al.
18. Kaczorek, M., Honore, N., Ribes, V., Dehoux, P., Cornet, P., Cartwright, T., and Streeck, R. E. (1987) Molecular cloning and synthesis of biologically active human tissue inhibitor of metalloproteinases in yeast. Biotechnology $595-598. 19. Docherty, A. J. P., Lyons, A., Smith, B. J., Wright, E. M., Stephens, P. E., Harris, T. J. R., Murphy, G., and Reynolds, J. 3. (1985) Sequence of human tissue inhibitor of metalloproteinases and its identity to erythroid-potentiating activity. Nature 318, 66-69. 20. Murphy, G. and Reynolds, J. J. (1985) Current views of collagen degradation. BioEssays 2,55-60.
21. Welgus, H. G. and Stricklin, G. P. (1983) Human skin fibroblast collagenase inhibitor, Comparative studies in human connective tissues, serum and amniotic fluid. J. Biol. Chem. 258, 12,259-12,264. 22. Vialard, J., Lalumiere, M., Vernet, T., Briedis, D., Alkhatib, G., Henning, D., Levin, D., and Richardson, C. (1990) Synthesis of the membrane fusion and hemagglutinin proteins of measles virus, using a novel baculovirus vector containing the P-galactosidase gene. J. Virol 64,37-50. 23. Miyazaki, J., Takaki, S., Araki, K., Tashiro, F., Tominaga, A., Takatsu, K., and Yamamura, K. (1989) Expression vector system based on the chicken P-actin promoter directs efficient production of interleukin-5. Gene 79,269-277. 24. Summers, M. D. and Smith, G. E., (1987) A manual of methods for baculovirus vectors and insect cell culture procedures. Texas Agric. Exp. Stu. Bull. 1555. 25. Laemmli, U. K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227,680-685. 26. Wickham, T. J., Davis, T., Granados, R. R., Hammer, D. A., Shuler, M. L., and Wood, H. A. (1991) Baculovirus defective interfering particles are responsible for variations in recombinant protein production as a function of multiplicity of infection. Biotechnol. Z&t. 13,483-488. 27. Weingarten, H., Martin, R., and Feder, J. (1985) Synthetic substrates of vertebrate collagenase. Biochemistry 24,6730-6734.
CHAPTER14
Expression of Foreign Genes in Bombyx mori Larvae Using Baculovirus Vectors Prabhakara Shizuo G. Kamita,
I? Choudary, and Susumu Maeda
1. Introduction Baculovirus expression vector systems based on the nuclear polyhedrosis viruses of Autographa californica (AcNPV) and Bombyx mori (BmNPV) are in wide use (I). Our laboratory originally designed the BmNPV system with the objective of using the silkworm, B. mori, as an in vivo host (la,2). The silkworm belongs to the order, Lepidoptera (butterflies and moths), consisting of herbivorous insect species that feed on foliage plants in the wild. Because of its economic importance in silk production, the silkworm has been domesticated for thousands of years. The larvae have traditionally been cultivated on leaves of the mulberry bush (Morus alba). Furthermore, in recent years, the silkworm has become an ideal multicellular eukaryotic model system for basic research. Although both the AcNPV and BmNPV systems provide highlevel expression of foreign genes using larval hosts (3,4), the silkworm larva offers several additional advantages,i.e., it is easy to rear, it is large and easy to manipulate, it has a relatively short life cycle (approx 7 wk), and its genetics and biology have been well documented. The availability of automated rearing equipment and the fact that the larvae are nonallergenic to human handlers make scale-up and mass production of recombinant proteins under sterile conditions very attractive. From: Methods in Molecular Biology, Vol. 39: Baculovirus Expression Protocols Edited by: C. D. Richardson Q 1995 Humana Press Inc., Totowa, NJ
243
244
Choudary, Kamita,
and Maeda
This chapter describes the experimental protocols necessary for use of the BmNPV expression system with silkworm larvae as hosts. A brief description of the BmNPV expression vectors and the production of recombinant virus (5) is also included, and differences from the AcNPV system are discussed. Procedures for the cultivation of silkworms in the laboratory, their infection with recombinant virus, and the purification of foreign gene products from the larvae are outlined. 2. Materials 2.1. Chemicals, Buffers, and Equipment Materials available in a standard molecular biology or biochemistry laboratory are generally adequate for use of the BmNPV/silkworm expression system. Special equipment, chemicals, buffers, and solutions for expression experiments using silkworm larvae, are listed below: 1. 1000X kanamycin (60 mg/mL): Dissolve 1 g of kanamycin in 16.7 mL of sterilizedwater understerileconditions,andstorein 1-mL aliquotsat -20°C. 2. 1M dithiothreitol (DlT): storeat -20°C. 3. Phenylthiourea. 4. Camel-hair brush and rearingboxes. 5. A recombinantBmNPV of interest. 6. 2% Bleach, 70% ethanol,and 2% formaldehyde. 7. 1-mL disposablesyringesequippedwith 30-gageneedles. 8. Dissecting pins, scissors,forceps, l-in-thick Styrofoamboard for dissection, 20- and 25-gageneedles,absorbentpaper,latex gloves, glasshomogenizer, and sterile 1S-mL microfuge tubes. 9. Transfer vectors for use with BmNPV/silkworm expression system (see Note 1). 10. Genomic DNA from BmNPV preparedfrom budded virus or occlusion bodies (seeNote 2). 2.2. Cell Lines and Culture Medium for the BmNPV Expression System Although several cell lines from the silkworm are currently available (see Note 3), BmN (BmN-4) cells (5,6) are most commonly used, BmN cells are maintained at 26-28°C in TC- 100 medium (6,7), supplemented with 10% fetal calf serum (FCS) (see Note 4). Cells are subcultured (1:2 to 1:3) every 4-7 d by scraping monolayer cells from the surface of the tissue culture dish using a rubber policeman and diluting in fresh complete medium, TC-100 medium, available commercially from Life
Gene Expression
in Bombyx mori Larvae
245
Technologies Inc. (GIBCO/BRL, Gaithersburg, MD, cat. # 11600-061). It is recommended that the commercial media be supplemented with tryptose broth (Difco Laboratories, Detroit, MI, cat. # 0062-01-4) at a concentration of 2.6 g/L. Alternatively, as done in our laboratory, TC100 medium can be made according to previously published recipes (5,7). 2.3. Silkworm Eggs Silkworm eggs and artificial diet can be obtained from the commercial sources described below, and stored for a few months at 5°C for convenience without losing viability. Silkworms are generally reared on artificial prepared diet or mulberry leaves in clean cardboard or plastic boxes at room temperature (23-27°C). In some strains of B. mod, diapause (hibernation) during the egg stage can last several months, severely reducing growth and metabolic rates, and present difficulties in rearing larvae for expression experiments. The length of diapause in silkworm varies in response to rearing conditions and genetic background. Diapausing silkworm eggs can be artificially induced to hatch by simple chemical treatments. The duration and nature of different developmental stages of the silkworm are also influenced by various environmental conditions, such as temperature, light, and nutrition. The supplier should be consulted for precise details concerning storage conditions and hatching procedures. Although over 320 different strains are available from various B. mori genetic stock centers, well-characterized, commercial strains (mainly hybrids) are generally easier to culture and are more suitable for use in expression experiments. Commercial strains are available in Japan, the United States, Canada, and India. Some of the suppliers are listed below: 1. National Sericultural Experiment Station, Oowashi, Tsukuba 305, Japan (Fax: 81-298-38-6028). 2. Institute of Silkworm Genetics, Faculty of Agriculture, Kyushu University, Higashi-ku, Fukuoka 812, Japan(Fax: 81-92-641-2928). 3. Katakura-Kogyo Co. Ltd., 5-25 Chuou, Matsumoto 390, Japan(Fax: 81 263-33-0549). 4. KaneboSilk EleganceCo. Ltd., 2615Hokuho, Kamitaraga,Kasugai,Aichi 486, Japan(Fax: 81-568-81-3080). 5. Institute of Silkworm Genetics and Breeding, 1053 Iikura, Ami-machi, Inashiki-gun, Iharaki 300-03,Japan(Fax: 81-298-89-2356). 6. Forest Pest ManagementInstitute, Forestry Canada,P.O. Box 490, Sault Ste. Marie, Ontario, TGA SM7, Canada(Tel. 705-949-9461).
246
Choudary, Kamita,
and Maeda
7. Member-Secretary, The Central Silk Board, United Mansion, MG Road, Bangalore 560003, India (Fax: 91-812-582177). 8. Entomological Society of America, P.O. Box 177, Hyattsville, MD 20781 (Fax: 301-731-4538).
2.4. Artificial
Diet for Rearing
Silkworm
Larvae diet in a ready-to-use form is generally superior to diet pre-
Artificial pared in the laboratory. Although some workers prefer to raise silkworms on mulberry leaves, this approach does not favor year-round cultivation of the insects. Some of the commercial suppliers of B. mori artificial diet include: 1. Kataknra-Kogyo Co. Ltd., 4-5-25 Chuou, 390 Matsumoto, Japan (Fax: 81263-33-0549). 2. Nihon Nosan Kogyo Co. Ltd, 20-4 Hinode 2-Chome, Funabashi, Chiba 273, Japan (Fax: 81-474-33-7680). 3. Yakurt Co. Ltd., Kunitachi, Tokyo 186, Japan. 4. Carolina Biological Supply Co., Burlington, NC, USA 27215 (Fax: 919584-3399) (cat. # L-905A). The use of ready-to-use diet is strongly recommended; however, diet can also be prepared in the laboratory by the following procedure using the components shown in Table 1 (7a): 1. Grind together all ingredients, except vitamins and antibiotic, add to water, and mix well. Pour into an autoclavable container. 2. Cover the container, and autoclave at 120°C for 20 min. 3. Cool to approx 45OC,add filter-sterilized vitamins and antibiotic, and mix well. 4. Pour into flat, sterile pans to a depth of about 1 in., and let solidify at room temperature. 5. Cover with Saranwrap, and store refrigerated (2-5OC, with humidity).
2.5. BmNPV
Virus
and Transfer
Vectors
1. BmNPV T3 (I) is a well characterized isolate of BmNPV and is commonly used in expression experiments. A DNA fragment library (8,9), physical maps for restriction endonucleases (8), and sequence information for nearly the entire genome (IO), as well as of important genes (1,9,1113) are available for the T3 isolate. 2. Transfer vectors: Two types of BmNPV transfer vectors, categorized as Group A and Group B, are available. Group A vectors are useful for the production of authentic gene products using the translational start codon
Gene Expression
247
in Bombyx mori Larvae
Table 1 Compositions of the Artificial Diets for Silkworm Larvae (7~)
Nutrient For Instar
Diet containing mulberry leaf powder l-4 5
Soybean oil, refined Soybean meal, defatted Potato starch Glucose Ascorbic acid Sorbic acid Citric acid Cellulose powder Salt mix, Wesson’s Vitamin B mix Antibiotics &v Cholesterol Mulberry leaf powder Sucrose p-sitosterol K2HP0, Morin
Diet without mulberry leaf powder, g/L 1 2-4 5
120 25 27 6.7 0.7 13.4 69.3 10.0 +
13.6 204.5 68 45.5 9.1 0.9 18.2 13.6 +
10.0 100.0 33.3 6.7 0.7 1.7 113.3 11.7 +
10.0 133.3 33.3 6.7 0.7 1.7 113.3 11.7 +
25.0 0.7 83.3
2i.7 0.9 113.6
50+.0
5;o 33.3 1.7 3.3 0.33
5.0
-
-
-
33.3 1.7 3.3 0.7
11.5
230.8 76.9 46.2 7.7 0.77 1.9 7.7 + 1;2 -
-
1.9
+Add at a concentration commonly used for insect artificial diet (76).
of the inserted gene (cDNA, genes without introns, or synthetic genes). Group B vectors are useful for the production of polyhedrin-fused proteins
from truncated genes or genes without introns fused to the N-terminal portion (3-133 bp) of the polyhedrin gene using the translational start of the polyhedrin gene. DNA sequence information of the multiple cloning sites useful for insertion of the foreign genes in both types of transfer vectors is available (5). Commonly used transfer vectors are as listed below: a. Group A transfer vectors (multiple cloning sites into which foreign gene may be inserted are indicated) i. pBE274 (EcoRI). ii. pBK273 (KpnI, SacI, and EcoRI). iii. pBE284 (EcoRI, XbaI, and A&I). iv. pBK284 (KpnI, SacI, &TORI, X&I, and A&I). v. pBE520 (EcoRV, XbuI, and A&I). vi. pBM030
(B&II,
SucI, SmuI, EcoRI, NcoI, EcoRV, XbaI, and AutI).
248
Choudary,
Kamita,
and Maeda
b. Group B transfer vectors (reading frame of the polylinker containing EcoRI, XM, and AatI sites is indicated in relation to the 5’ coding sequence of the polyhedrin gene). i. pBF4, pBF37, pBF79, pBF133 (G 1AAT 1ICI 1AGA 1TAG I GCC IT). ii. pBF5, pBF48, pBF81, pBF129 (GA 1ATT 1CIA 1GAT 1AGG 1CCI’). iii. pBF14, pBF23, pBF68, pBF124 (GAA 1TIC 1TAG 1ATA 1GGC 1CT).
2.6. Production of Recombinant Mouse IL-3 and Human @Interferon 1. Column chromatography packings: ACA54, QAE Sephadex (Pharmacia, Uppsala, Sweden), C8 reverse phase (Pharmacia), Sepharose 4B (Pharmacia). 2. Acetonitrile (HPLC grade), 0.1% trifluoroacetic acid (J. T. Baker, Philipsburg, NJ). 3. 0.2M Sodium acetate, pH 7.3,0.5M NaCl, 20% (w/v) glycerol. 4. O.lM Citric acid, pH 2.2,0.5M NaCl, 20% (w/v) glycerol.
3.1. Construction
3. Methods of Recombinant
Transfer
Vectors
1. Digest l-5 pg of the transfer vector with the appropriate restriction endonuclease(s) in a total reaction volume of 50 p,L. 2. In order to increase the insertion rate of the foreign gene, treat the digested transfer vector with either bacterial alkaline phosphatase or calf intestinal phosphatase, followed by extraction with an equal volume of phenol-chloroform (one time), and chloroform (one time). 3. Ligate 1 PL (containing 0.024.10 pg of DNA) each of the digested transfer vector and appropriately digested (and purified) foreign DNA fragment in a total volume of 12 j.tL, 4. Transform 40 PL of competent Escherichiu coli cells with 5 pL of the ligated DNA (keep chilled on ice for 10 min, heat shock at 43.5OC for 30 s, add 100 l,tL of culture medium without ampicillin, and incubate at 37°C for 30 min), and plate on an LB-agar plate containing ampicillin (50 jtg/mL). 5. Identify a clone carrying the recombinant transfer plasmid by (a) restriction endonuclease analysis of plasmid DNAs prepared by a minilysis procedure, and/or, (b) hybridization using the inserted DNA fragment as a labeled probe. 6. Propagate the recombinant transfer plasmid using a large-scale procedure. Purify the DNA by alkaline lysis methods for maxi-preparation of DNA plasmids. In order to construct a recombinant virus, approx 5 ltg of purified recombinant transfer vector is required for cotransfection.
Gene Expression
in Bombyx mori Larvae
3.2 Isolation
of Recombinant
249 Virus
3.2.1. Cotransfection of BmN Cells with Transfer Vector and Wild-Type BmNPV DNA
Cotransfection, mediated by the calcium phosphate-DNA coprecipitation method (1,5), is generally used to introduce DNA into insect cells. Treatment of host cells with Lipofectin (Life Technologies Inc.) may also be used for cotransfection. Preparation of viral DNA for cotransfection has been described previously (5) (see Note 2). All of the procedures for cotransfection outlined below should be conducted in a biosafety hood, using sterile technique. All stock solutions should be filter-sterilized and stored at -20°C. Microfuge tubes, Pasteur pipets, pipet tips, and HZ0 should be autoclaved. Cells should be incubated in a humidified tissue-culture incubator at 27”C, unless otherwise indicated. 1. Seed2-3 x 104BmN cells in 4 mL of complete TC-100 medium in a 60mm tissue-culturedish, and incubatefor at least 2 h (preferablyovernight) prior to use. 2. Make Solutions A and B freshly in separate1.5~mLmicrofuge tubes: Solution A (250 pL): wild-type BmNPV T3 DNA (2 l.tg), recombinant transfer vector DNA (5 pg), 2M CaCIZ(30 pL). Sterile distilled water to make up the final volume to 250 p.L. Solution B (250 pL): 50 mM Hepes,pH 7.1,280 mM NaCl(245 pL), 35 mil4 NaH,P04 (2.5 PL), 35 mM Na2HP04 (2.5 pL). 3. Using a Pipetman (P-1000), add Solution B dropwise to Solution A over a period of 20 s, while gently bubbling air into the mixture through a sterile Pasteur pipet.
4. Incubate the cotransfectionmixture at room temperaturefor 30 min. 5. Add the cotransfectionmixture gently to the previously seededBmN cells, and add 10 i.tL of 1000X kanamycin (60 mg/mL). 6. Incubate the cells at 27OC for 6-16 h, replace the culture medium with fresh TC-100 containing 10% PBS, and continue incubation. 7. Collect the culture medium within 5-7 d after confirmation of viral repli-
cation (polyhedralproduction), and plaquepurify the recombinant virus. 3.2.2. Isolation
of Recombinant
Viruses by Plaque Assay
The medium from the cotransfected cultures generally contains virus at a titer of 106-lo* PFU/n& of which O.l-2% is recombinant BmNPVs.
250
Choudary, Kamita,
and Maeda
For plaque assays, the cotransfected culture medium is diluted ( 102-lo4 times), so that approx 500 plaques/60-mm culture dish are formed. Infected cells are overlaid with medium containing SeaPlaque agarose (FMC Corporation, Pine Brook, NJ) to immobilize the budding virus. Recombinant plaques are identified after 5-7 d using light microscopy. All procedures should be conducted under sterile conditions, unless otherwise noted. 1. For plaque assay,seed BmN cells at 50% confluency (2 x lo6 tells/60-mm culture dish), at least 2 h (preferably overnight) prior to infection. At least 3 dishes/sample are necessary (12 dishes, four dishes/dilution are suggested). 2. Assay lo-fold serial dilutions of the cotransfected cell supernatant. Generally, 100 lt,L each of cell supernatant diluted lo*, 103, and lo4 times in TC100 medium without FCS are used to infect one dish (from step 1). For infection, remove all but 200 l,tL of the culture supernatant from the dishes, and infect each dish with 100 l.tL of the appropriate dilution. Incubate for 1 h at room temperature. Rock the dish gently every 15 min so that the cells do not dry up. 3. During the l-h incubation period, prepare a 0.75% SeaPlaque agarose overlay containing 5% FCS as follows: Agarose Solution: Agarose = (number of dishes + 1) x 34 mg Water = (number of dishes + 1) x 1.13 mL Media-Solution: 1.33X TC-100 = (number of dishes + 1) x 3.44 mL FCS = (number of dishes + 1) x 0.23 mL 1000X kanamycin = (number of dishes + 1) x 9 PL Prepare the agarose solution in a microwavable container, large enough to hold both agarose and media solutrons. Prepare the media-solutron m a sterile 50-mL centrifuge tube, and incubate at 37°C. Ten min prior to the completion of the Incubation period, microwave the agarose solution until completely melted, and allow to cool to 37°C. After the incubation period is complete, pour the media solution into the container of agarose solution, mix well, and allow to cool to 35OC. Completely remove the infection solution from the dishes (using a Pipetman P-1000), and overlay the cells with 4.8 mL of the media agarose solution causing the least possible disturbance. Allow the agarose overlay to solidify and dry in the sterile hood for 20 min with the lids slightly alar. 4. Incubate the dishes at 27°C in a humidified incubator (if a humidified incubator is not available, the dishes should be sealed with Parafilm).
Gene Expression
in Bombyx mori Larvae
251
5. Plaques should become visible to the naked eye in 5-7 d. Visualization of recombinant plaques can be facilitated by adding 2.0 mL of a second agarose overlay containing 0.02% neutral red (add 48 pL/dish of 1% neutral red stock to the media solution and subtract 48 pL/dish of water from the agarose solution) after 5-7 d (see Note 5). 6. Characterize plaques using an inverted microscope. Recombinant plaques are characterized by infected cells with obvious cytopathic effect (CPE), and lack polyhedral inclusion bodies (PIBs). On the other hand, the cells in wild-type plaques also show CPE, but contain nuclear PIBs (from 5 to 50 black inclusions/cell). Initially, scanthe dishes containing as many plaques as possible without overlap, for the presence of putative recombinant plaques that appear clear against the red background. Observe clear plaques under higher magnification (100-200X) to confirm whether or not they are recombinant plaques (i.e., note absence of inclusion bodies). Mark recombinant plaques at low magnification (50X) using a felt-tipped lab marker (Sharpie) from underneath the dish (see Note 6). 7. Collect marked recombinant plaques by piercing the agarose overlay with a sterile cotton-plugged Pasteur pipet and carefully suck out the agarose plug into the pipet tip. Transfer the agarose plug into a sterile 1.5~mL microfuge tube containing 500 p,L of TC-100 media with 1% FCS, and vortex. Subject 5-100 pL of this solution to a second round of plaque assay as described above. If the second plaque assaydoes not produce pure isolated plaques (lo-50/dish), a third plaque assay may be necessary. Select pure recombinant virus isolates as described above. 8. Propagate pure recombinant virus by infecting dishes containing 2 x lo6 tells/60-mm tissue-culture dish (see step 1) with 100 pL of the plaque solution as described above (step 7). The cells in theseinfections should be free of PIBs and produce about 4 mL of viral solution containing approx lo8 PFU/mL. 3.2.3. Propagation
of NPVs in Established
Cell Lines
Viral stock solutions should contain at least lo* PFU/m.L (preferably 4
x lo8 PFU/mL). BmNPV-based recombinant viruses should only be stocked at -80°C or 5”C, not at -20°C. 1, Seed BmN cell monolayers (4 x lo6 tells/60-mm dish or 3 x lo7 cells/l50mm tissue-culture dish) at least 2 h prior to infection. Infect BmN cells with 5 PFU/cell of virus by removing all but 200 pL of medium from the 60-mm dish (or all but 500 pL from the 150~mm dish), and infect for 1 h with gentle rocking every 15 min, so that the cells do not dry out. After 1 h of infection, add 4 mL of complete TC100/60-mm dish (15 n&/150-mm dish), and incubate at 27OC.
252
Choudary, Kamita,
and Maeda
2. After 3 d of incubation at 27*C, transfer the culture medium from the infected cells into a sterile centrifuge tube, centrifuge for 10 min at lOOOg,at 5”C, and collect the supernatant. 3.3. Expression of Foreign Genes in Established Bombyx mori Cell Lines 1. Seed about 3 x lo7 BmN cells/l50-mm culture dish for at least 2 h (preferably overnight) prior to infection. 2. Remove the culture medium, and infect with 5 PFU of virus/cell (375 l.tL of a 4 x lo* PFU/mL stock). Incubate for 60 min at room temperature, with gentle rocking every 15 min. 3. Add 15 mL of TC-100 containing 10% FCS, and incubate at 27°C in a humidified incubator. 4. In order to minimize protein interference from the growth media during the purification process, remove the TC-100 medium, and replace with 15 mL of Grace’s (Life Technologies Inc., cat. # 11300-027, and so on) or defined media without FCS. This should be done at 36 h postinfection (i.e., 24-36 h prior to collection). 5. For secreted proteins, collect supernatant 60-72 h postinfection, and centrifuge at 1OOOgfor 5 min (to remove cells and cell debris). 6. For nonsecreted (cell-associated) proteins, scrape the cells from the dish using a rubber policeman if necessary, and transfer the cells and medium into a 50-mL centrifuge tube. Centrifuge at 200-500g for 5 min, and remove the supernatant. If necessary, wash the cells once with chilled PBS, recentrifuge, and discard the supematant. 3.4. Rearing of Silkworm Larvae The silkworm (shown in Fig. 1) is a holometabolous insect with four distinct developmental stages: egg, larva, pupa, and adult (moth). During the egg stage, the developing silkworm undergoes diapause or hibernation (see Note 3). Following diapause, the life cycle of the silkworm lasts about 49 d; about half this time is spent in the larval stage. The silkworm can feed only during the larval phase, which is further subdivided into five stages called instars (or stadia), interspersed by four meltings (Fig. 2A). The molting stage is characterized by the following behavior: increased activity (movement), feeding cessation, development of a transparent cuticle, and formation of a newly developed head. While molting, the larva stops feeding for a period of 24 h during the first molting, increasing up to 36 h, following the fourth molting. At the end of this process, the larva sheds its old cuticle (ecdysis) and develops a new one. Six to 7 d after the final (4th) molt, the fifth instar larva reaches a maximum length of approx
Gene Expression
in Bombyx mori Larvae
253
Fig. 1. Photograph of the silkworm larvae.
70 mm and weighs about 5 g after voracious feeding. At this time, feeding ceases, the cuticle becomes transparent and yellowish, and the larva continuously spins silk thread and forms a cocoon over a 2-d period. One or 2 d after the cocoon is complete, the larva undergoes ecdysis and enters the pupal stage. In about lo-12 d, the adult moth emerges, mates, and lays eggs. To rear B. mori larvae for use as hosts for foreign gene expression, follow the steps given below: 1. Remove eggs (commercially obtained, over-wintered eggs, or HCl-treated eggs) from cold storage (2-5OC). In general, 50-200 larvae are adequate for one experiment. 2. For laboratory-raised eggs, sterilize eggs by soaking in 2% formaldehyde solution for l-2 min. Repeatedly wash (l-2 min) the formaldehyde off the eggs with tap water. Air dry. 3. Incubate for several d at 25°C at high humidity (a plastic box containing a ball of water-soaked Kimwipes is sufficient). Larvae will hatch a few days after the color of the eggs changes to dark blue. 4. Use a tine camel-hair brush to collect hatched larvae (collect within a day of hatching) onto thinly sliced (
254
Choudary, Kamita,
A
_-
rp-
3 Y ml
1st Instar (lo+ days) (4 days)
J -
and Maeda
Adult (1 + days)
2nd lnstar (3 days)
t
i
tziG@
3rd lnstar (4 days)
Pupa (lo-12 days)
I
B
l-3 day-old 5th lnstar
J
n
Injection of virus
Collection of hemolymph
Fig. 2. (A) Life cycle of the silkworm. (B) Infection of the silkworm by injection with a recombinant virus and collection of hemolymph from a proleg. 5. Feed the larvae thin slices of artificial diet every day, or every other day, making sure that moist diet is always available. Remove the feces (frass) every 3-5 d. Except during molting, keep 1st and 2nd instar larvae in a humidified environment. Also, continually increase the size of the
Gene Expression
in Bombyx mori Larvae
255
rearing container as larvae grow in size. Keep 4th and 5th instar larvae in a relatively nonhumid environment, e.g., on the laboratory bench. Early 5th instar larvae (i.e., l-3 d after molting or about 18-20 d after hatching) are at an ideal stage for infection with BmNPV. At this stage, each larva should weigh about 1 g. 6. To obtain eggs from moths, isolate larvae prior to pupation (larvae with an almost transparent skin, and shrunken to about 50% of 5th instar larvae, which are of maximum size), and place them in dry cardboard boxes layered with a sheet of old newspaper. Several days after formation of the cocoon, remove pupa by carefully cutting one end of the cocoon with a razor blade. Store pupae in a dry place at room temperature. Moths should emerge in lo-12 d. Pupae can be stored at 5OC for a few days to delay emergence. One to 2 d prior to emergence the compound eyes of the silkmoth can be seen as black patches through the pupal cuticle (as seen in the pupal stage of the drawing in Fig. 2A). Female moths are easily distinguished from male moths by the larger size of their abdomen. In males, the structure of genitalia is pronounced. Female moths live for only 1 or 2 d after they emerge: however, males can be stored at 2-5°C for several days and can be mated two or three times. On the same day that female moths emerge, they should be allowed to mate and lay eggs in a confined area (e.g., under a 16-0~. Dixie cup placed on top of a piece of cardboard). Eggs will be laid within a day after mating. 3.4.1. Prevention
of Diseases
Several species of fungi, viruses, protozoans, and parasitic flies can cause severe epidemic outbreaks in silkworm colonies. In the laboratory, viral diseases caused by contamination of NPVs used for expression experiments constitute a major diseaseproblem. After the 3rd instar, silkworm larvae are relatively resistant to oral infection by NPVs. Furthermore, recombinant NPVs lacking the polyhedrin gene are generally noninfectious. Since oral infection is the primary route of viral infection, it is important to prevent contamination of early instar colonies with polyhedra. Once a silkworm colony becomes infected with wild-type NPVs, the laboratory generally becomes heavily contaminated, making it difficult to rear silkworm colonies for several months. To prevent infection, eggs can be surface-sterilized in 2% formaldehyde for l-2 min as discussed previously. Larvae can also be reared under semisterile conditions in sterilized containers. Viral-infected larvae used in the expression experiments should be kept away from healthy colonies, preferably in a separate room. Diet should be cut and handled
256
Choudary,
Kamita,
and Maeda
with a sterilized or detergent-cleaned knife. Should viral contamination occur, the entire colony should be destroyed (by autoclaving) and the rearing area should be decontaminated using 2% bleach or 2% formaldehyde. It is frequently observed that fungi or physical discomforts, such as high humidity or uncomfortable temperatures, interfere with the optimal growth of silkworms, resulting in a significant reduction in the quality as well as quantity of expressed foreign gene products. If fungal contamination should occur, immediately transfer larvae to a clean container of fresh diet and discard the contaminated rearing container and diet. 3.5. Use of Insect Diapause as a Means
to Store
Eggs and Regulate
Hatching
Silkworms undergo diapause (hibernation) during the egg stage. Silkworms are classified into uni-, bi-, or multivoltine strains, depending on the length of their diapause. Commercially available silkworms are mostly of the bivoltine type, whereas silkworms obtained privately in the United States are generally a mixture of the uni- and bivoltine strains. Eggs laid by silkworms reared on artificial diet, however, are often a mixture of diapause and nondiapause eggs. Nondiapause eggs hatch at room temperature in about 10 d after oviposition. Commercially available eggs are artificially-treated diapause eggs (i.e., ready to hatch like nondiapauseeggs). The hatching of diapause eggs can be controlled so that larvae emerge anywhere between 2 wk and several months. The color of silkworm eggs changesafter oviposition; diapause eggs turn dark gray within a few days after oviposition, whereas nondiapause eggs remain light brown until 6-8 d after oviposition, i.e., 2 d before hatching, when they change to dark blue. Nonfertilized eggs are light yellow in color, and often appear crushed or dented over time. The hatching of diapause eggs can be controlled by the following three procedures in order to vary the times of larval emergence. 3.5.1. Hatching
within
2 MO
Diapause eggs can be artificially induced to hatch by treatment with hot hydrochloric acid just before or shortly after the onset of diapause as follows: 1. Use eggs depositedon egg cards, 20-30 h (stored at 25°C) or 24-48 h (storedat 20°C) after oviposition. 2. Preparea 16% (v/v) (5.W) solution of hydrochloric acid (specific gravity = 1.064),and heat to 46°C (seeNote 7).
Gene Expression
in Bombyx mori Larvae
257
3. Fix the eggs on egg cards by soaking in 2% (v/v) formaldehyde solution for 1 min. Briefly wash egg cards with water, and gently blot dry with a Kimwipe. 4. Dip the egg cards in hot (46OC) 16% HCl for 4-7 min. 5. Wash in running water for 10 min to remove traces of acid, and immediately air-dry. 6. Let stand at 25°C for 20-30 h before storing at 2-5°C (for no longer than 30 d). 7. Larvae will hatch 8-10 d after transferring egg cards from 2-5°C to 25°C. 1. 2. 3. 4. 5. 6. 7.
3.5.2. Hatching Between 2 and 5 MO Use eggs stored at 25°C for 2 d (for hatching between 1 and 2 mo) to 3 d (for hatching between 2 and 3 mo) after oviposition. Keep eggs at 2-5°C for l-3 mo, and transfer to 25OC 10 d before larvae are needed for hatching. Prepare a 20% (v/v) (6.3N) solution of hydrochloric acid (specific gravity = 1.10) and heat to 47.8OC. Fix the eggs on egg cards by soaking in 2% (v/v) formaldehyde solution for l-2 min. Briefly wash egg cards with water, and gently blot dry with a Kimwipe. Dip the eggs cards in hot (478°C) 20% HCl for 5 min. Wash in running water for 10 min to remove traces of acid, and immediately air-dry. Incubate eggs at 25OCunder humid conditions until they hatch (8-10 d).
3.5.3. Hatching Between 4 and 10 MO 1. Keep eggs at 20-25°C (room temperature) for at least 60 d after oviposition, This period can be extended to 4-6 mo. Transfer eggs to 2-5°C and store for 50 d to several months. 3. Store eggs at 2-5”C, until required for hatching. 4. Transfer the eggs to 25°C 8-10 d before larvae are needed.
3.6. Production of a Foreign Gene Product in Silkworm Larvae Infected with a Recombinant Virus In order to prevent viral contamination, wear disposable latex gloves and cover the lab bench with absorbent paper during the following steps: 1. Starve an appropriate number of 5th instar larvae (l-2 d after molting) for a period of 3-12h. 2. Prepare rearing boxes. Layer thinly sliced artificial diet on the bottom of a clean cardboard box, of appropriate size, lined with paper towels.
258
Choudary, Kamita,
and Maeda
3. Prepare the viral suspension. Make a 1:lO dilution of the stock viral solution (with a titer of approx 10s PFU/mL) using TC-100 medium without FBS but containing 6 mg/mL kanamycin, so that the final concentration is 104-lo5 PFU/lO pL, (e.g., mix 100 PL viral stock, 800 p.L TC-100 without FBS, and 100 pL 1000X kanamycin in a sterile 1.5-n& microfuge tube). Viral stock for injection can be stored at -80°C for several months. 4. Anesthetize larvae by immersion in an ice water bath for 5-20 min. Blot dry 10-20 larvae at a time using Kimwipes. Inject 10 yL of the virus suspension (104-105 PFWO pL) just underneath the dorsal cuticle (inject longitudinally being careful not to pierce the midgut) of each larva using a 1-mL syringe, fitted with a 30-gage needle. 5. Transfer larvae to the rearing boxes, and incubate at 25°C for 3-4 d with regular feeding as necessary. Collect the recombinant protein after larvae show symptoms of severe viral infection (e.g., yellowish body color, feeding cessation, slow movement, and so forth), but before the larvae die. 6. Secreted proteins are collected from the hemolymph (see Note 8). To collect hemolymph, bend the larvae backwards, and hold the head and tail of the larvae with one hand so that the abdomen and prolegs are sticking outward (Fig. 2B). Hold the larvae over a 1.5~mL microfuge tube, collect hemolymph by piercing the larval proleg using a 25-gage needle plugged with parafilm, and allow the hemolymph to drop into the microfuge tube. To prevent melanization, the tube should contain l/l0 vol of 10-100 rnit4 DTI’ to give a final concentration of l-10 mM. A few crystals of phenylthiourea can also be used to prevent melanization; however, phenylthiourea is toxic to cultured cells. Centrifuge the samples at 300g for 1 min, transfer the supernatant into fresh microfuge tubes, and store samples at -80°C. Alternatively, the hemolymph can be collected into a 1.5-n& microfuge tube placed on dry ice and stored at -8OOC. 7. Cell-associated proteins are collected from the fat body. Although a majority of the larval tissues support viral replication and presumably express foreign gene products, the fat body is used for the purification of nonsecreted foreign gene products. Place larvae, dorsal side up, on a piece of Styrofoam or dissecting dish, stretch out, and pin down the head and tail. Make a longitudinal incision from tail to head. Using forceps, pinch off the midgut at either the head or tail end, and remove the entire midgut and midgut contents. Collect the fat body tissue using a scraper onto a piece of weighing paper. Slide fat body along the weighing paper so that the fat body and hemolymph separate, and transfer the fat body into a 1.5-n& microfuge placed on dry ice. Store the tubes at -80°C.
Gene Expression
in Bombyx mori Larvae
259
3.7. Purification of Foreign Gene Products from the Silkworm Larvae The procedures followed for purification of foreign gene products from the hemolymph of infected larvae are similar to the biochemical procedures commonly used for purifying proteins from other organisms. However, melanization of the hemolymph sample must be prevented, so that aggregation of proteins does not occur, interfering with the purification process. Once melanization occurs, further purification is nearly impossible. Dithiothreitol (DTT), phenylthiourea, or heat treatment can prevent melanization almost completely. Protein aggregation can also be prevented by rapid chromatographic purification of the recombinant protein from the components involved in the melaninization reaction cascade. 3.7.1. Purification
of Secreted Recombinant Proteins from the Hemolymph 1. Thaw frozen samples, vortex, and centrifuge immediately at 8000g for 1 min at 5°C (or at 1OOgfor 5 min at 5OC). 2. Treat the supernatant appropriately to prevent melanization. If DTT or phenylthiourea is not used, the excess sample must be purified or frozen immediately. Molecular sieving, ion-exchange, or affinity column chromatography may be used to purify the recombinant protein, depending on its physico-chemical characteristics. Once the purification process is initiated, melanization will not occur. 3.7.2. Purification
of Cell-Associated Recombinant Proteins from the Fat Bodies 1. Homogenize the fat bodies containing the recombinant protein in an appropriate buffer using a glass homogenizer, Polytron, or equivalent. Generally, the buffer should also contain protease inhibitors (phenylmethylsulfonyl fluoride, leupeptin, and pepstatin A). Melanization of the proteins present in the homogenized fat body usually does not occur unless there is a severe contamination of hemolymph in the fat body sample. 2. Centrifuge the homogenized fat bodies at lOOOg-10,OOOgfor 2-10 min. Follow an appropriate purification procedure, e.g., molecular sieving, ion exchange, or affinity chromatography, precipitation, and so on, based on the characteristics of the expressed protein.
260
Choudary, 3.8. Examples
of Purified
Produced
Kamita,
Recombinant in Larvae
and Maeda
Protein(s)
The major advantage of expressing foreign genes in larvae is the extremely high levels of expression achievable compared to expression in cell lines, In vivo expression levels of 50-1000 times higher than in vitro levels are common. For example, mouse IL-3 (mIL-3) can be produced at about 500 times higher levels in larvae than in cultured cells (14). In most cases, protein degradation has not been encountered in the silkworm hosts, probably because of the fact that the silkworm stores secreted proteins in its body cavity filled with hemolymph, which contains various protease inhibitors. Cell-associated recombinant proteins are not as highly expressed in larvae as are their secreted counterparts. These proteins are generally produced in larvae at about lo-fold higher rates than in established cell lines; however, hepatitis B surface antigen was produced at about 1000 times higher levels in larvae than in the in vitro system. Protein degradation may occur at higher levels in fat bodies than in cell lines; however, most foreign proteins expressed in larvae undergo normal posttranslational modifications. For example, C-terminal amidation occurs more efficiently in larvae than in cell lines, since the enzyme that catalyzes amidation is synthesized in the fat bodies of Lepidopteran insects. 3.8.1. Production
of Recombinant
Mouse IL-3 in Larvae
1. Two- to 3-d-old 5th instar silkworm larvae (1500 larvae) were injected with a recombinantBmNPV carrying mousemIL-3 cDNA (14) andreared on artificial diet. 2. Four to 5 d postinfection, hemolymph was collected in centrifuge tubes (l-l.3 niL/tube) on ice. Every 10-20 tubes,about l/10 vol of 50 mM DTT was addedinto eachtube. The solutions were briefly vortexed and centrifugedat 3000rpm for 2 min. The supematantwascollectedin a 15-mL polypropylenecentrifugetube(4 mL/tube)andstoredat -8OOCuntil purification. 3. The frozen hemolymph (4 tnL) was thawed, vortexed, and centrifuged at 1OOOg for 5 min at 5°C. 4. The supematantwas appliedto an ACA54 gel-exclusion column (1.4 x 90 cm), equilibrated
with buffer A (50 mM Tris-HCl,
pH 7.5), containing
100
mA4NaCl. Fractions containing mIL-3 activity (identified by immunoreaction with mIL-3 antibody or by Coomassieblue staining on an SDSpolyacrylamide gel) were concentrated by an 80% ammonium sulfate precipitation and dialysis againstbuffer A.
Gene Expression
in Bombyx mori Larvae
261
Fig. 3. Purified mIL-3 from the hemolymph. Aliquots of the fractions were separated on a 12.5% SDS-polyacrylamide gel. Lane 1, M, standards with M, size @Da) shown at the far left; lane 2, hemolymph from infected larva; lane 3, ACA54 fractions; lane 4, QAE-Sephadex fractions; lane 5, C8 reverse-phase fractions. Arrows indicate the three species of mIL-3. 5. The dialyzed solution was applied to a QAE-Sephadex column (Pharmacia Biotech Inc., Piscataway, NJ) pre-equilibrated with buffer A. 6. The void volume was concentrated with an Amicon filter (Amicon Inc., Beverly, MA, YM5) and applied to a C8 reverse-phase column (Pharmacia Biotech Inc.). Purified mIL-3 was eluted at an acetonitrile concentration of 35% and 0.1% trifluoroacetic acid. 7. In an alternative approach, the hemolymph sample (20 mL) was applied directly onto a large CS reverse-phase column (Pharmacia Biotech Inc.) and eluted at an acetonitrile concentration of 35% and 0.1% trifluoroacetic acid. Fractions showing mIL-3 activity were collected. 8. Analysis by SDS polyacrylamide gel electrophoresis demonstrated one to three protein bands that represent differently-glycosylated mIL-3 (Fig. 3) (see Note 9). 3.8.2. Production of Human a-Interferon in Silkworm Larvae 1. Hemolymph samples were prepared from silkworm larvae infected with a recombinant BmNPV carrying the human a-interferon gene (I). 2. The supernatant sample was applied onto an affinity Sepharose 4B column constructed with monoclonal antibody against human a-interferon. The column was washed several times with 0.2M sodium acetate, pH 7.3,0.5M NaCl, and 20% glycerol.
262
Choudary,
Kamita,
and Maeda
3. Recombinant interferon was eluted from the column with O.lM citric acid, pH 2.2, OSM NaCI, and 20% glycerol. On SDS-polyacrylamide gel electrophoresis, the purified recombinant protein appeared as a single major band of the M, size expected of the h a-IFN (I) (see Note 10).
4. Notes 1. The BmNPV transfer vectors described in this chapter are available from our laboratory. 2. Genomic DNA from BmNPV can be prepared from budded virus or occlusion bodies as described in Chapter 5 of this book. More details are also presented in reference (5). 3. The following cell lines provide an alternative to the commonly used BmN cell line. BoMo (15): BmNPV replicates well in BoMo with high production of PIBs. Two subclones of BoMo are available. Although BoMo15A cells attach loosely to the surface, they are often found suspended. Bm36 (16): The origin of the cell line may not be B. mori; BmNPV replicates in this cell line; however, polyhedral production is poor. Bm36 is generally propagated on solid media. 4. Each lot of fetal bovine serum (FBS) should be tested for its ability to support cell growth before proceeding with further experiments. 5. Step five of the procedure can be omitted after becoming familiar with the identification of recombinant plaques. 6. Recombinant viral plaques can be identified using the procedures outlined in Chapter 6 using AcNPV. 7. Concentrated HCl has a concentration of 12N and is 38% (v/v). It has a specific gravity of 1.19. 8. On the average, each larva produces about 0.4 mL of hemolymph. 9. It was estimated that the hemolymph of silkworm contained 6 x lo* units/ mL of m&3. Based on the specific activity calculations, it was estimated that each larva produced 0.5 mg of recombinant IL-3 in its hemolymph (14). 10. It was estimated that the hemolymph from each silkworm contained at least 50 pg of recombinant a-interferon. 11. Foreign genes can also be expressed using the AcNPV vector and their lepidopteran hosts, e.g., Tricoplusiu ni and I-h&this virescens. These species, however, are significantly smaller than B. mori and often cannibalistic so that special rearing conditions are required. Munduca sex&, which are larger (weighing up to 10 g) than B. mori, can also be used as an expression host for the AcNPV vector (I 7); however, in this host expression rates are comparatively low and AcNPV does not replicate normally. Procedures for using AcNPV (18) and rearing its hosts (4,7u, 7b,l9) are available elsewhere.
Gene Expression
in Bombyx mori Larvae
263
Acknowledgments The authors are grateful to Kei Majima for helpful suggestions on developing protocols and preparing the illustrations, Bruce D. Hammock and Carl W. Schmid for discussions, and Carole Hicks for secretarial assistance. The work discussed in this chapter was supported in part by grants from the USDA (#91-37302-6186), NIEHS Superfund Basic Research Prog (#P42-ESO-4699), and EPA (#CR-819047-01-0). P. V. C. is a Rockefeller Foundation Biotechnology Career Fellow. References 1. Maeda, S. (1994) Expression of foreign genes in insect cells using baculovirus vectors, in Insect Cell Biotechnology (Maramorosh, K. and McIntosh, A., eds.), CRC, Boca Raton, FL, pp. 1-31. la. Maeda, S., Kawai, T., Obinata, M., Fujiwara, H., Horiuchi, T., Saeki, Y., Sato, Y., and Furusawa, M. (1985) Production of human alpha-interferon in silkworm using a baculovirus vector. Nature 315,592-594. 2. Maeda, S. (1989) Expression of foreign genes in insects using baculovirus vectors. Ann. Rev. Entomol. 34,35 I-372.
3. Price, P. M., Reichelderfer, C. F., Johansson, B. E., Kilbourne, and E. D., Acs, G. (1989) Complementation of recombinant baculoviruses by coinfection with wildtype virus facilitates production in insect larvae of antigenic proteins of hepatitis B virus and influenza virus. Proc. Natl. Acad. Sci. USA 86,1453-1456. 4. Medin, J. A., Hunt, L., Gathy, K., Evans, R. K., and Coleman, M. S. (1990) Efficient, low-cost protein factories: expression of human adenosine deaminase in baculovirus-infected insect larvae. Proc. Natl. Acad. Sci. USA 87,2760-2764. 5. Maeda, S. (1989) Gene transfer vectors of a baculovirus, Bombyx mori nuclear polyhedrosis virus, and their use for expression of foreign genes in insect cells, in Invertebrate Cell System Applications, vol I (Mitsuhashi, J., ed.), CRC, Boca Raton, FL), pp. 167-181. 6. Volkman, L. E. and Goldsmith, P. A. (1982) Generalized immunoassay for Autographa californica nuclear polyhedrosis virus infectivity in vitro. Appl. Environ. Microbial.
44,227-233.
7. Gardiner, G. R. and Stockdale, H. (1975) Two tissue culture media for production of lepidopteran cells and nuclear polyhedrosis viruses. J, Znvertebr. Pathol. 25,363-370. 7a. Ito, T. and Kobayashi, M. (1978) Rearing of the silkworm, in The Silkworm: An Important Laboratory Tool (Tazima, Y., ed.), Kodansha, Tokyo, pp. 83-102. 7b. Singh, P. and Moore, R. F. (1985) Handbook of insect rearing, ~01s. I and II. Elsevier, New York. 8. Maeda, S. and Majima, K. (1990) Molecular cloning and physical mapping of the genome of Bombyx mori nuclear polyhedrosis virus. J. Gen. Virol. 71, 185 1-1855. 9. Maeda, S., Kamita, S. G., and Kataoka, H. (1991) The basic DNA-binding protein of Bombyx mori nuclear polyhedrosis virus: the existence of an additional arginine repeat. Virology 180,807-810.
Choudary, Kamita,
and Maeda
10. Gomi, S., Chua, J. W., and Maeda, S. (1994) Program Abstr. Ann. Meet. Am. Sot. Viral. Abstr. P17-5. 11. Kamita, S. G., Majima, K., and Maeda, S. (1993) Identification and characterization of the p35 gene of Bombyx mori nuclear polyhedrosis virus that prevents virusinduced apoptosis. J. Viral. 67,455463. 12. Maeda, S., Kamita, S. G., and Kondo, A. (1993) Host range expansion of Autographa califomica nuclear polyhedrosis virus (NPV) following recombinantion of a 0.6~kilobase-pair DNA fragment originating from Bombyx mori NPV. J. Viral. 67,6234-6238. 13. Ohkawa, T., Majima, K., and Maeda, S. (1994) A cysteine protease encoded by the baculovirus Bombyx mori nuclear polyhedrosis virus. J. Viral. 68, in press. 14. Miyajima, A., Shreurs, J., Otsu, K, Kondo, A., Arai, K., Maeda, S. (1987) Use of the silkworm, Bombyx mori, and an insect baculovirus vector for high-level expression and secretion of biologically active mouse interleukin-3. Gene 58, 273-281. 15. Inoue, H. and Mitsuhashi, J. (1984) A Bombyx mori cell line susceptible to a nuclear polyhedrosis virus. J. Seric. Sci. Jpn. 53, 108-113. 16. Quiot, J. M. (1982) Establishment of a cell line (S. P. C. Bm36) from the ovaries of Bombyx mori L. (Lepidotpera). Sericologia 22,25-3 1. 17. Gretch, D. G., Sturley, S. L, Friesen, P. D., Beckage, N. E., and Attie, A. D. (1991) Baculovirus-mediated expression of human apolipoprotein-E in Manduca sexta larvae generates particles that bind to the low density lipoprotein receptor. Proc. Natl. Acad. Sci. USA 88,8530-8533.
18. O’Reilly,
D. R., Miller, L. K., and Luckow, V. A. (1992) Baculovirus Expression Manual. W. H. Freeman, New York, pp. l-347. 19. Shorey, H. H. and Hale, R. L. (1965) Mass-rearing of the larvae of nine noctuid species on a simple artificial medium. J. Econ. Entomol. 58,522-524. Vectors: Laboratory
CHAPTER15 Expression of Foreign Proteins in Dichoplusia ni Larvae Jemey A. Medin, Karen Gathy, and Mary Sue Coleman 1. Introduction A large number of eukaryotic proteins have been successfully overproduced in insect cells infected with a natural pathogen, a baculovirus, in which foreign gene coding sequenceswere placed under the control of the viral polyhedrin gene promoter. This expression system is effective because many proteins are appropriately modified after translation, and production levels are often high in the host cells. However, a potential disadvantageof using standardrecombinant baculovirus infection of insect cells is the cost associated with a large-scale tissue-culture operation. The insect cells commonly used for baculovirus propagation are a continuous ovary-derived cell line, Spodoptera frugiperda. Although these cells grow efficiently on flat surfaces, either in plates or in T-flasks made of polystyrene (I), or in roller bottles (2), they can also be cultured in suspension in spinner flasks. The use of the latter-type culture container permits growth of reasonably large quantities of cells by using equipment normally available in most tissue-culture laboratories. However, the high-oxygen requirement for optimal cell growth in this insect line necessitates a large surface area and therefore limits the volume of the culture (3). The oxygen requirement is best satisfied by using an air-lifttype fermenter since insect cells are fragile and cannot withstand the shear forces that result from the introduction of air by bubblers found in bacterial fermenters (4). If a large air-lift-type fermenter is used, then From: Methods In Molecular Biology, Vol. 39: Baculovlrus Expression Protocols Edited by: C. D. Richardson Q 1995 Humana Press Inc., Totowa, NJ
265
Medin,
Gathy, and Coleman
media cost must also be considered. The Sf9 cells are normally cultured in a defined medium containing 10% fetal bovine serum. Some strains of Spodopteru have been adapted to minimal serum requirements (4), and these cells may be employed in some circumstances. To circumvent large-scale tissue culture, we have investigated the infectivity of a number of recombinant baculoviruses and the stability of their respective gene products in insect larvae. The polyhedrin promoter has been shown to be very active in larvae (particularly Trichoplusia ni) and routes of infection have beenper OS(5) and by direct injection into the larvae (6). The feeding route of infection requires that the virus particles be encapsidated, and variable infection rates are observed depending on the feeding schedule of the larvae. Injection of virus into larvae near the set of four prolegs at the fourth instar of development resulted in negligible loss of larvae owing to inoculation injury and expression of recombinant proteins that varied from 1 to 10% of total soluble insect protein (6). We and our colleagues have now expressed four enzymes in the active state and eight mutationally altered proteins (Table 1). Adenosine deaminase, terminal deoxynucleotidyl transferase, HIV reverse transcriptase, and human UMP synthase were produced in larvae. The steady-state expression level varies with each construct introduced into the virus and may be related to the protease resistance of the protein. However, an unanticipated benefit that results from the low optimal temperature of host insect culture (2526°C) was observed from these studies. Mutationally altered proteins that exhibit potential thermolabilities are readily produced and isolated from larvae (7). Purification of recombinant proteins expressed in larvae is straightforward and can often be accomplished using the same procedures developed for mammalian tissue protein purification (6,7). The only standard modification suggested is removal of viral DNA from crude cell homogenates by treatment of the extracts with protamine or streptomycin sulfate. As an example for yields of purified proteins, we routinely obtained 20-35 mg pure human adenosine deaminase from 100 infected insect larvae. The yield of mutationally altered proteins is usually lower than the yields from wild-type recombinants, an observation that may reflect the inherent instability of the former proteins. However, as a result of our current range of experience with this system, it is our expectation that this procedure will be generally applicable for the production of a wide array of recombinant proteins.
Production
of Foreign Proteins in T. ni Larvae
267
Table 1 Foreign Proteins Expressed in Trichoplusia ni Protein”
Extract activityb
Human aclenosine deaminase wt F117 ~161 *21k
~264
+++ ++ ++ + + + ++
~272
’
N214 ~214
AZ2 Human terminal deoxynucleotidyl wt Wt (fusion)
Extract antigenb
transferase
~227
HIV reverse transcriptase wt Human LIMP syntbase wt
+ + +
+ + +
+++
+++
++
++
*Various mutants of adenosine deaminase were expressedin cabbage loopers (6). Mutations are indicated as an amino acid substitution at a given residue number. Terminal deoxynucleotidyltransferase was expressed either as the wild-type enzyme or as a protein fused to five extra amino acids at its amino-terminus (7). The effect of mutating its Cys at residue 227 to a Gly was monitored. The HIV reversetranscriptasewas produced by Robert Evans, Department of Chemistry, Williams College, and the UMP synthase produced in the laboratory of Mary Ellen Jones, Department of Biochemistry, University of North Carolina. me highest level of antigen produced in extracts (t++) represented 5-10% of the soluble larvae protein as measured by Western blots. By kinetic analyses the wild-type recombinant enzyme activity was indistinguishable from enzyme isolated from natural tissue sources. As an example of recombinant enzyme yields, we have obtained expressron of 1 mg wild-type ADA per larvae and 2.6 mg wild-type HIV reverse transcriptase per larvae (R.Evans, personal communication) in crude extracts. Intermediate levels (t+) of antigen represented l-2% of soluble protein and the lowest levels of antigen (t) represented less than 1% of soluble protein.
2. Materials 1. The insect cell line Spodopteru fmgiperdu, clone 9 (American Type Culture Collection, Rockville, MD), or any other insect line developed for expression of recombinant baculovirus (as an example, High Five insect cells, BTI-TN-SB 1-4, Invitrogen, San Diego, CA). 2. The recombinant virus stock is expanded for insect inoculation by infection of 1.2 x lo7 cells in a large (75 cm*) T-flask. The viral stock (1 mL) is added to cells growing in the flask in a final volume of 25 mL complete Grace’s medium, and allowed to incubate at 27OCfor 5 d or until complete
268
3.
4. 5.
6.
Medin, Gathy, and Coleman cell lysis is observed. The remaining cells and debris are removed by centrifugation at 800g for 10 min, and the virus stock is stored at 4°C. The titer of virus in the medium is consistently found to be l-2 x lo* PFU/mL for all isolates. After long-term (>2 yr) storage at 4”C, viral stocks exhibit no loss of PFU. Insect diet components: Dry pinto beans (any commercial food store); ascorbic acid (Sigma [St. Louis, MO] #AO278); sorbic acid (ICN [Costa Mesa, CA] #102937); methyl paraben (Sigma #H5501); alphacel (ICN #/900453); vitamin mixture (ICN #904654, this is a special order vitamin diet fortification mixture pg AD-43 animal diets); Brewer’s yeast (ICN #103312); agar (Sigma#A9915); styrofoamcups with lids (8 oz., squat style, 10.5-cm top diameter); formaldehyde 37% (Fisher [Pittsburgh, PA] ACS certified #579); green blotter paper (any office supply house). A premixed formulation of diet for Trichoplusiu ni larvae in powdered form can also be obtained commercially (Bio-Serv, Frenchtown, NJ). Trichoplusiu ni pupae (to establish the insect colony): the pupae may be obtained in most entomology departments, from the authors’ laboratory (University of North Carolina), from the laboratory of David Jones (University of Kentucky), or the Forest Pest Management Institute (Saulte Ste Marie, Ontario, Canada). Protease Inhibitors, including benzamidine hydrochloride, phenylmethylsulfonyl fluoride (PMSF), EGTA, leupeptin, and trans-epoxysuccinyl-L-leucylamido-(4-guanidino)butane (E-64) can be purchased from Sigma (St. Louis, MO). Pefabloc SC is from Boehringer Mannheim (Indianapolis, IN).
3. Methods To use the insect larvae as a host for expression of foreign proteins by infection with recombinant baculovirus, it is necessarythat a small insect colony be maintained for production of the larvae. This is a relatively simple operation, and the materials required are inexpensive and readily available. The steps below illustrate in detail the establishment of a colony and the infection of larvae with the virus preparations. 3.1. Preparation of the Insect Diet A single batch of the insect diet is sufficient for 20 cups, each containing 20-35 larvae. If multiple batches of diet are to be made, each should be mixed separately. 1. Soak 450 g of pinto beans overnight in 550 mL of distilled water. Rinse beans with fresh water and drain right before making the diet.
Production
of Foreign Proteins in T. ni Larvae
269
2. On the next day, boil 600 mL of distilled water, and remove vitamins from the refrigerator. 3. Weigh out the diet ingredients: 9 g ascorbic acid, 2.5 g sorbic acid, 6 g methylparaben (p-OH-benzoic acid methylester), 20 g alphacel, 55 g vitamins, and 150 g Brewer’s yeast. 4. To a Waring-type blender (high capacity4 L) add 600 mL of cold distilled water and the diet ingredients. Blend for 30 s on high speed. Add the washed, soaked beans, and blend for 2 min until the ingredients have been thoroughly mixed. 5. Add 6.2 g of agar to the boiling water from step 2, and dissolve the agar completely (about 5 min). Add the agar solution to the blender and mix for an additional 1.5 min. Add 5 mL of 37% formaldehyde and mix for 30 s in the blender. 6. Pour diet into Styrofoam cups to a depth of at least 2 cm, and cool for 30 min without the covers. 7. When the diet is solidified and cooled, staple an egg sheet(see Section 3.2.6.) to the inside of the lid. The egg sheet will be about 2 cm2 and should contain about 35 eggs. The sheetsshould be protected from moisture.
3.2. Establishing a Mating Colony for Insect Egg Production The mating colony is essential for production of a steady supply of insect eggs that will be used for hatching into larvae. The colony is contained in a small box (40 x 40 cm) with a hole cut for a light source that is on a 24-h timer device.Thehoxmaybeasirnplecardboarclcontaineroramomelaboratew0oden box. Photographs of cabbage loopers and adult moths are shown in Fig. 1. 1. Cover a paper cylindrical quart container with a fiberglass mesh screen lid, and line the container with rolled green blotter paper (15.5 x 27 cm). 2. Rinse Trichoplusiu ni pupae with 10% chlorine bleach for 10 min. This step kills any viruses on the external parts of the pupae. Rinse the pupae in distilled water for 10 min. They are now ready for separating by sex. 3. The Trichoplusiu ni males have a two-dimensional light coloring difference near the last segment line that resembles a moustache or bow tie. The female has a pair of three-dimensional bumps near the last segment line. Place the pupae under a dissecting microscope, and look for these features. Figure 2 shows a detail of the characteristics to use for sexing the pupae. It is easiest to use the marking on the males to determine sex. Place the pupae in the quart container with the mesh screen (20 females and 20 males). It is important that these numbers be maintained since smaller insect numbers, even in the same proportion, will not result in efficient mating.
270
Fig. 1. Photographs of Trichoplusia their moths.
Medin,
Gathy, and Coleman
ni larvae at 5th instar together with
4. Place the containers (one or two are sufficient) in a box with a 25-w incandescent bulb on a 14-h light/lo-h dark cycle: 6 AM to 8 PM light and 8 PM to 6 AM dark. Soak a cotton pad with 10% sucrose, and place on the screen of the container. Resaturate the pad every morning and evening. This serves as food for the adults as they are mating. 5. When the adults start laying eggs, remove the liner daily and replace. The moths will continue to lay eggs for about S-10 d. Discard moths and unwanted eggs by freezing and autoclaving (see Note 1). 6. Cut the egg sheets into three strips about 10 in. by 2 in., and soak in 1 L 7.4% formaldehyde for 20 min. Soak with distilled water for 40 min, air-
Production
of Foreign
Proteins
Fig. 2. Sexing of Trichoplusia
in
T. ni
271
Larvae
ni pupae. The male is shown
on the left, and
the female on the right. See text for detailed description. dry on paper towels for 5 h or overnight, and cut and staple to the lids so that there are about 35 eggs/diet cup. The cups should be stored on the laboratory bench at room temperature where the eggs will hatch in 3 or 4 d.
3.3. Irqjection
of
Larvae
with Recombinant
Virus
1. Selecting larvae at the appropriate developmental stage is important for success in expression of the recombinant proteins from viral infection (Fig. 3). Larvae should be large enough to facilitate injection and yet not ready for pupae formation. Ten days after the eggs hatch, check the larvae twice a day to identify the appropriate stagefor injection. Larvae should be selected in between the 4th instar and the 5th instar stages (14-20 d old). Protein synthesis capacity in the insect is optimal at this stage. The 4th instar lasts 2-3 d, and the 5th instar lasts 4-6 d. When the larvae are at the end of the 4th instar, they stop feeding and walk to the top of the containers. The larval color turns bright green, since they are not feeding and they are easy to distinguish. These larvae should be collected for injection. When the larvae complete the molting process, they return to the diet mix and feed for about 2-4 d. At the end of the 5th instar stage, larvae stop feeding again, wander back to the cup lid, and spin a silk cocoon, Shortly thereafter, they molt to the pupal stage.
cgqs hatch brwe .-.02
Ern
Inject larvae on the lost day of 4th tnstor or 1st day of 5th m&r l
5th mstar (Pupal stage) lasts 1 day fmshe$ ~pmnmq cocoon and pupates
‘ONa*
-25
cm
Collect larvae
lotwe 3-4 days after m~ed~ons - 0 3 qmmo. w 3 cm
.Lawe ahould not be m the wandermq stage of the 5th mstor when collected so great core should be mmntmncd rhen seelectmg fw ,n,ecbon
Fig. 3. Schedule of selected events in the life cycle of Trichoplusiu ni.
Production
of Foreign Proteins in T. ni Larvae
273
Fig. 4. Injection site of Trichoplusia ni larvae. The 27.5-gage needle should be l/8-1/4 cm deep, needle point down and bevel up. See text for description of injection procedure. 2. Place selected larvae on ice for 15-30 min. This slows movement and makes the injection step easier. 3. Inject 10 pL recombinant baculovirus stock/larvae using a 1-mL syringe and a 27.5-gage needle. The virus titer for this step should be about lo* PFWrnL. The injection site (illustrated in Fig. 4) is forward along the side near the set of four prolegs, just under the skin (l/8-1/4 cm deep). There may be a small amount of leakage from the injection site, but this is not deleterious to the larvae. However, if the needle is inserted too deeply, the larvae will expire. Place injected larvae back into diet cups (with lids) and return to shelf (see Note 2). 4. After 3-4 d (the time allowed for recombinant protein expression may vary with particular protein and should be monitored), place larvae in -80°C freezer for later use in protein isolation. Discard any larvae that have turned black (see Notes 3 and 4). 5. If the colony is to be continuously maintained, uninjected larvae should be allowed to pupate in the diet cup. The pupae are removed from the silk, sexed as before and placed in the light box chamber for mating and egg production. 3.4. Protein Purification from Insect Larvae 1. Frozen larvae should be homogenized using a Polytron/Tissuemizer-type device, Proteases may pose a problem, and thus many investigators use multiple protease inhibitors throughout purification steps. The insect is known to exhibit an alkaline-activated protease, so we routinely maintain buffer pH at 6.8 or less. In addition, we use PMSF (100 pg/mL), EGTA (1 mM), benzamidine (50 yg/mL), and Pefabloc (1 mM). If oxidation of the crude extract is evident (discoloration to brown or black) 10 mM mercaptoethanol should be included in the crude extract. Other protease inhibitors that are useful are leupeptin (2 p,g/mL) and E-64 (1 mM), which can inhibit cysteine proteases (see Note 5). 2. Centrifuge the crude extract at 18,000g for 30 min, and filter the supematant through a double layer of cheesecloth to remove lipid. The clarified solution should be a yellow-green color. In order to remove viral DNA,
274
Medin, Gathy, and Coleman
add 1 mg of protamine sulfate/20 mg total protein. The protamine sulfate should be neutralized prior to use by dissolving (with mixing) in several milliliters of water to which a drop of O.lM NaOH has been added. After addition of protamine sulfate, stir the supernatant in the cold for 1 h. Centrifuge at SOOOgfor 20 min at 4OCto pellet DNA. 3. At this stage, any conventional protein purification procedure may be employed to recover the recombinant protein of interest. Protein expression should be monitored by activity measurements or Western blots. The inclusion of His6 (8), immunotags (9), or glutathione-S-transferase gene fusions (10) can greatly facilitate the purification process.
4. Notes 1. Approximate insect growth rates and life cycle (see Fig. 3 for schematic): day 1, eggs laid; day 4 eggs hatch; days 14-20 larvae injection; days 18-24 collection; days 17-25 pupate; days 30-35 moths emerge. 2. With experience, a single individual can inject at least 200 larvae/h. 3. A black oxidation product is sometimes observed in the insects and in crude supernatant on extended storage. Larvae that are dead or totally black in color should not be collected. 4. Injected larvae are usually collected on day 4. However, expression levels of individual recombinant proteins should be experimentally verified by Western blots or activity measurements. By day 5, the larvae will be totally infected by virus, and the tissue integrity will be lost. 5. Foreign protein production will vary greatly, depending most probably on the stability of the expressed product in larval cells.
Acknowledgments This work was supported by a grant from the National Cancer Institute
CA2639 1. We are grateful for the editorial assistanceof Judith A. Lesnaw. References 1. Weiss, S. A. and Vaughn, J. L. (1986) Cell culture methods for large-scale propagation of baculoviruses, in the Biology of Buculoviruses, vol. II (Granados, R. R. and Federici, B. A., eds.), CRC, Boca Raton, FL, pp. 63-87 2. Weiss, S. A., Smith, G. C., Kalter, S. S., and Vaughn, J. L., (1981) Improved method for the production of insect cell cultures in large volume. In Vitro 17, 495-502. 3 Streett, D. A. and Hink, W. F. (1978) Oxygen consumption in Trichoplusia ni (TN-368) insect cell line infected with Autographa californica nuclear polyhedrosis virus. J. Invertebrate Path01 32, 112-l 13. 4. Maiorella, B., Inlow, D., Shanger, A., and Harano, D. (1988) Large-scale insect cell-culture for recombinant protein production. Bioflechnology 6, 1406-1409.
Production
of Foreign Proteins in T. ni Larvae
275
5. Price, P. M., Reichelderfer, C. F., Johansson, B. E., Kilbourne, E. D., and Acs, G. (1989) Complementation of recombinant baculoviruses by coinfection with wildtype virus facilitates production in insect larvae of antigenic proteins of hepatitis B virus and influenza virus. Proc. Natl. Acad. Sci. USA. 86,1453-1456. 6. Medin, J. A., Hunt, L., Gathy, K., Evans, R. K., and Coleman, M. S. (1990) Efficient, low-cost protein factories: Expression of human adenosine deaminase in baculovirus-infected insect larvae, Proc. Natl. Acad, Sci. USA 87,2760-2764. 7. Medin, J. A. and Coleman, M. S. (1992) Lack of functional significance of cys227 and cys234in terminal deoxynucleotidyltransferase. J. Biol Chem. 267,5199-5201. 8. Hochuli, E., Bannwarth, W. Dobeli, H., Gentz, R., and Stuber, D. (1988) Genetic approach to facilitate purification of recombinant proteins with a novel metal chelate absorbant. Bio/Technology 6, 1321-1325. 9. Quelle, D. E., Lynch, K. J., Burkert-Smith, R. E., Weiss, S., Whitford, W., and Wojchowski, D. M. (1992) Phosphorylatable and epitope-tagged human erythropoietins: Utility and purification of native baculovirus-derived forms. Prot. Expression
Purification
3,461-469.
10. Peng, S., Sommerfelt, M., Logan, J., Huang, Z., Jilling, T., Kirk, K., Hunter, E., and Sorscher, E. (1993) One-step affinity isolation of recombinant protein using the baculovirus/insect cell expression system. Prot. Expression Purification 4,95-100.
CHAPTER16
Use of Baculoviruses as Biological Insecticides Jenny S. Gory and David H. L. Bishop 1. Introduction Baculoviruses have been isolated from a wide range of invertebrates. Their development as pest control agents spans over 40 years, although records of their discovery and use date from considerably earlier (I). The baculoviruses from Lepidoptera (butterflies and moths), Hymenoptera (sawflies only), and Coleoptera (beetles) present the best options for pest control, and thus most research on baculovirus insecticides has concentrated on these orders. Baculovirus insecticides have been used in a wide range of situations from forests and fields to food stores and greenhouses, Baculoviruses have several advantages over conventional insecticides that make them highly acceptable control agents. Probably the most important is their specificity. They have narrow host ranges (sometimes limited to one or two species), and do not infect beneficial insects, making them very suitable for use in integrated control programs. Most also possess the capacity to persist in the environment, which can be utilized in the development of more ecologically based long-term control programs. Baculoviruses are particularly suitable for use in developing countries, since they can be produced locally and their production in vivo (although labor-intensive) requires little in the way of capital expenditure. More recently, interest in baculoviruses has expanded in relation to their potential for genetic modification, primarily to increasetheir efficacy. This area is still in its infancy, and the development of genetically modified baculoviruses will not be considered here (however, see Cory [2], Possee From: Methods in Molecular Biology, Vol. 39: Baculovitus Expression Protocols Edited by: C. D. Richardson Q 1995 Humana Press Inc., Totowa, NJ
277
278
Cory and Bishop
et al. [3], and Cory et al. [3uJ), although most factors that are relevant to their usein the field will be the sameas for naturally occurring baculoviruses. The use of baculoviruses as biological insecticides is a broad subject area. The rationale taken in this chapter has been to outline the pathway required for the development of a control program utilizing baculoviruses, from isolation to field trials, and also to describe aspects of the theoretical background (see Fig. 1). The further development of baculoviruses for use in integrated control programs and their commercialization are beyond the scope of this chapter. 2. Materials The majority of baculoviruses have been isolated from Lepidoptera: over 500 to date (4). Whether this reflects a genuine host bias or is merely a product of the fact that there are large numbers of lepidopterists (compared to other insect specialists) and that baculovirus infected larvae are usually very obvious, remains to be seen. However, it does indicate that for Lepidoptera at least, there is a high probability of isolating a baculovirus from a given species. Many pest-control programs rely on the isolation of local strains of baculoviruses. This can have some advantages: it will circumvent any problems that may occur if an “exotic” virus is imported for wide-scale release, and may result in the selection of a virus better adapted to a particular host or ecosystem. Isolation of wild-type baculoviruses from insect pests can vary in difficulty. In families, such as the Lymantriid moths (for example, the gypsy moth, Lymantria dispar, and the Douglas fir tussock moth, Orgyia pseudosugata), virus epizootics are common and isolates are not difficult to collect, particularly in high-density populations. The apparent rarity of epizootics in some other species, however, does not preclude the successful use of a baculovirus for their control. The current means of characterizing a newly isolated baculovirus is by restriction endonuclease digestion of its genomic DNA. The resulting DNA profile gives a good indication of the uniqueness (or otherwise) of the isolate. It is also highly likely that any wild-type isolate will be a mixture of several genotypically distinct varieties, for example, the lepidopteran Helicoverpa spp. nuclear polyhedrosis viruses (NPVs) (5) and the stem borer Chile spp. granulosis viruses (GVs) (6). The relevance of this heterogeneity to successful pest control has yet to be ascertained.
Phase I
Phase II
Phase IV
Application strategy
Phase V
Phase VI
Intaractmn with other control agents
El Laboratory effkacy testmg Formulauon testing
Fig. 1. Schematic pathway for the development of baculoviruses boxes refer to those categories dealt with in the text.
Registration
as control
agents. Numbered
Gory and Bishop
280
3. Methods Every pest:crop system will present its own unique problems, but the general approach and areasfor study will be similar for most pest control programs. The most important of these areas are discussed below. 3.1. Screening
and Eflcacy
The first step in the initiation of any pest-control program using baculoviruses is the screening of available isolates to assesstheir efficacy as potential control agents. This can only be accomplished by insect bioassay. Production and registration of baculoviruses also require the establishment of a standard bioassay system. Bioassay of baculoviruses basically describes the relationship between the virus and the insect in terms of overall mortality or rate of kill. The standard dose-response curve of insects to baculoviruses is sigmoidal. When setting up a bioassay, a range finding assay should be carried out first, using a large range of doses, to assessthe shape of the response and to pinpoint the linear region of the curve. A second assay is then set up with the doses arranged symmetrically around the LC5,-,or LDsO point (see below). The spacing of the doses and the number of test insects can vary, but in general, if the number of doses (and subjects) is low, the doses should be widely spaced and vice versa. Several parameters can be measured using the bioassay: The lethal concentration (LC)50 and lethal dose (LD)50 are those most commonly used in dose-mortality studies. The LC5a is the concentration of virus that the test insects feed on or drink that results in 50% being killed. An LD5c is the dose at which there is 50% mortality of the test insects. An LCsOtells you less about the response than the LDsO, since in the former, the amount of virus ingested is dependent on the behavior of the insect and so is inherently more likely to vary between tests, whereas the LDsO is independent of differences in individual behavior patterns, making this technique more suitable for comparative experiments. The most common method of analysis for dose-mortality data is probit analysis, which is used to determine the LD5a and LCsO values, although other models are also used (73). Analysis of the time-mortality response (time to death for 50% of the test insects) can be measured either as the lethal time (LT)50, where test insects are continually exposed to virus, or as the survival time (ST)50,
Baculovirus
Insecticides
281
where the inoculum is only received at the beginning of the assay. Probit analysis is not valid for analyzing time-mortality assays, and other transforms, such as the logit, should be used for these data. Many factors will affect the outcome of bioassays, including the instar used, temperature, the choice of insect diet, and larval weight. In order to increase the precision of the assay, it is important to minimize heterogeneity within the system, in particular, by using test insects within a very narrow weight range. The assay must be standardized, particularly if batches of virus are being tested against a standard, as in a potency assay. Several insect bioassay techniques have been developed, and there are numerous variants of these to account for the different feeding requirements of particular insect species. It is not possible to make valid comparisons between different bioassay techniques, since each will generate different values. The two most commonly used bioassays will be described here. 3.1.1. Diet Plug Assay Many Lepidoptera, and in particular, noctuid species, will feed on an artificial or semiartificial diet, which makes insect bioassays much easier to handle than when foliage is used. Details of a range of diets and the insects that have successfully been reared on them can be found in Singh and Moore (9). 1. Place a small plug of diet (of a size suitable for the chosen instar to eat within 24 h) in the base of a 96-well microtiter plate (a larger plate will be needed for the larger instars). 2. Apply 1 l.L of the virus suspension to the diet plug with a micropipet. 3. Place larvae individually in wells and cover. It is usually best to seal the
microtiter plate with damptissuesto preventdehydrationof larvae or diet, and to reduce the likelihood of escape. 4. Place the microtiter plate at constant temperature for 24 h. Transfer larvae to small individual pots that contain enough diet to maintain them through
to pupation. Only transferthoseinsectswhich have completely consumed their diet plug and havethereforetakenup a measureddoseof virus. Seal and perforate the lids. 5. Check after 24 h to remove handling deaths, and then monitor and record daily until death or pupation. 6. With well-infected larvae, diagnosis can often be made visually. However, with early deaths or small larvae, it is often difficult to do this. The easiest
method of confirming diagnosis of NPVs is to make a thin smear of the
Gory and Bishop
282
midgut region, stain with Giemsa, and observeat xl000 under oil immersion. Granulosis viruses and subgroupC baculoviruseswill needfurther confirmation using alternativetechniques,such asELISA, dot blotting, or by the transmissionelectron microscope. 3.1.2. Droplet Feeding Assay
This particular system was developed by Hughes and Wood (10) and has since undergone numerous improvements (8). The technique was developed for neonates, however, it has also been successfully used for later instars (IOa). Its main advantage is that the dose is ingested over a very short time, and infection can thus be regarded as synchronous. This is of particular importance in the measurement of LT,e or ST,, responses. 1. Preparationand selection of the larvae: Larvae are chosenfrom a narrow window of hatch time, 3-5-h old. They are then selected for vigor by allowing them to climb a ramp of blotting paper. 2. Larvae are placed on a plastic surface that has been ringed with a 2: 1 mix of mineral oil and Vaseline to prevent them from escaping. The virus suspension is then placed in small drops near the larvae. The suspension should contain a small amount of blue food coloring. 3. Larvae that have fed on the virus solution should be obvious because of the blue dye in then gut. These larvae are transferred carefully to individual pots containing diet. 4. Monitor as appropriate. For LTsOor ST,, analysis, this may well need to be every few hours.
The bioassay carried out in this manner can be used to estimate the LC,,. The amount of fluid ingested (and thereby the dose) needs to be estimated for evaluation of an LD,,. This can be done using 32P(IO), by fluorescence spectroscopy (II), or gravimetrically. 3.2. Virus
Production
At present, natural viruses for pest control are produced in vivo. The use of an in vitro system for virus production is some time away, because the yield of virus/unit volume is still too low to be considered economically feasible and the virus often loses infectivity for the insect host after passagethrough cell culture. With larger-scale production programs such as for the corn earworm HeEicoverpa zea NPV (12) and gypsy moth, L. dispar NPV (I3,14), the factors affecting the optimization and cost effectiveness of in vivo virus production have been studied in more detail. The most important aspects of the production process are: choice
Baculovirus
Insecticides
283
of host, dose/instar relationship, rearing conditions, and virus processing, all of which are outlined below. For a more detailed discussion of these and other areas, see Shapiro (14) and Carter (15). 3.2.1. Choice of Host It is usually preferable to use the natural virus host, which, in order to reduce the likelihood of contamination, should be reared in the laboratory rather than collected from the field. It is also a considerable advantage to use a host species that has been adapted to an artificial diet. Alternative hosts must be sought if the natural host is unsuitable for laboratory culture, for example, where the dietary requirements are too stringent, where a long obligatory diapause is required, or if it produces irritant, urticacious hairs. In these circumstances, it is important to confirm the identity of the progeny virus, since alternative hosts might select other strains from the original (mixed) inoculum or switch to a homologous virus if this is present in a latent state within the alternative host culture. 3.2.2. Instar-Dose Relationship The aim of the production process is to use the minimum quantity of inoculum to produce the maximum yield and quality of virus. This will require several range-finding experiments in order to optimize the process. For instance, Kelly and Entwistle (16) compared the use of third, fourth, and fifth instar cabbage moth, Mumestru brussicae larvae for NPV production, Third instars clearly gave the larger yield, which did not vary within the dose range tested (6.25 x 107-6.25 x lo* polyhedral inclusion bodies [PIBs]/mL). Other factors can also affect the viral product. For example, activity of virus appears to vary depending on when it is harvested from the larvae. NPV collected from H. zea larvae was found to be more active and more abundant when it was recovered from dead larvae rather than from live ones (12). Similarly, in L. dispar, the activity of the NPV obtained from different instars varied; it was found to increase up to the fourth instar and then decrease in the fifth (17). 3.2.3. Rearing Conditions In general, the most important environmental consideration is temperature. Other factors, such as humidity, appear to have little affect on virus production, except perhaps in terms of creating favorable conditions for the growth of undesirable contaminants. The optimum tempera-
284
Gory and Bishop
ture range appears to be between 20” and 26°C. Shapiro (14) investigated gypsy moth, L. dispar, NPV production at temperatures between 23” and 32”C, and found that at the highest temperatures, both yield and activity of the NPV were lower. Similarly, Kelly and Entwistle (16) looked at the effect of temperature on the rearing of both M. brussicae and the pine beauty moth, PanoZisJlammea NPVs in M. brassicae larvae over the range of 20” to 30°C. The results again tended to point to a reduction in yield with an increase in temperature. The other key factors for virus production are size of container and the number of larvae reared/container. Harvesting the virus-infected larvae is the most time-consuming, and thus expensive, stage of the virus production process, so it pays to use larger containers, although this must be set against the cannibalistic nature of many species. 3.2.4. Processing
The processing of virus-killed larvae and the level of purification used greatly affect the final yield. To liberate the virus from the larval bodies, the insects are blended with diluent and then filtered through muslin. Both the ratio of body weight to diluent and time spent blending have a significant affect on the final virus yield. Shapiro (14) showed with L. dispar NPV that using 1 g larval weight with 10 mL distilled water resulted in 95% PIB recovery after filtration. At lower dilutions, a greater proportion of the virus was lost. Similarly, with blending time, 5 s in a blender resulted in 75% PIB recovery, whereas 15 s resulted in 95%. A comparison of semipurification and full purification has been made for the production of M. brassicae and the brown tail moth, Euproctis chrysorrhoea, NPVs (16,18). Semipurification involved blending and coarse filtration, followed by low-speed contrifugation to remove debris and a higher-speed spin to pellet the virus. Both the low- and high-speed spins were repeated to ensure maximum virus recovery. For full purification, these steps were followed by centrifugation at 30,OOOgthrough a discontinuous 50-60% sucrose gradient. With M. brussicae NPV, the fully purified product resulted in a loss of 25-30% of the virus and E. chrysorrhoea NPV with a 35% loss of PIBs as compared to the semipure preparation. The authors calculated that the ratio of PIBs produced to inoculum used was > 1000: 1 for E. chrysorrhoea NPV, 2200: 1 for M. brassicae NPV, 2150: 1 for 0. pseudosugatu NPV, and 4000: 1 for L. dispar NPV.
Baculovirus
Insecticides
285
3.3. Formulation
The formulation of baculoviruses, although important, has not really received much attention. Formulation of baculovirus insecticides basically falls into two areas: that relating to storage stability and factors important to field application. Storage stability is very much dependent on the method of processing, but it has been little tested and there are few guidelines on it. The main aims are to produce a stable preparation in which the viability of the baculovirus is preserved or even enhanced. Most baculoviruses are processed for use as sprays, and research into the production of solid preparations is much rarer. For small-scale trials, storage stability is not usually a problem since macerated larvae mixed with water or partially purified preparations are often very effective in the field and will keep reasonably well for short periods provided they are refrigerated or, even better, frozen. This is not as practical for larger quantities of virus, where there is a need to move away from time-consuming steps, such as centrifugation, and where it is important to keep contaminant levels low. Most baculovirus products are produced as concentrated wettable powders. The main methods for large-scale processing and formulation are spray- or air-drying after dilution with an inert carrier (e.g., the commercial H. zea NPV product “Elcar”), the more expensive process of freeze-drying (lyophilization) with a carbohydrate, often lactose (e.g., US Forest Service formulations of L. dispar NPV and 0. pseudosugata NPV), or acetone precipitation in lactose. In terms of stability, the indications are that spray-drying produces the best-quality product, although freeze-drying powders also retain activity. Preparations produced by acetone precipitation are not favored, because they tend to lose activity in storage. In general, NPVs appear to be more stable than GVs, and aqueous preparations lose activity more rapidly than dry formulations. For further details, see reviews by Couch and Ignoffo (19) and Young and Yearian (20). Formulation for the field must provide good residual activity on the target site, and is thus dependent on the nature of the substrate and the effect of environmental parameters (temperature, humidity, sunlight, and so on). Little is actually known about the effect of substrate on virus viability, but of the environmental parameters that might affect them, only ultraviolet (UV) radiation between 290-320 nm is thought to have any major influence in inactivating the virus. Baculoviruses completely
286
Coy and Bishop
lose activity in a matter of h in direct sunlight. The addition of UV protectants and other compounds to baculovirus formulations have been rather random and unsystematic in terms of their assessment. In general, the types of compounds that have been added to baculovirus formulations to act as UV protectants fall into two categories: UV reflectants and UV absorbers. The former are primarily metallic oxides, and the latter are a varied mixture of substances that include antievaporants (to delay droplet evaporation during spraying), feeding stimulants, spreaders/wetting agents (usually detergents added to reduce aggregation) and stickers (although adhesion to many plant surfaces appears to be very strong) (I). Whatever additives are included in baculovirus formulations, care must be taken to check that they have no detrimental effects on virus activity. Since baculoviruses break down in alkaline conditions, pH is obviously particularly important, Additives must also be tested to see that they do not encourage aggregation or reduce adhesion of PIBs to the target substrate. They must also be compatible with the spray equipment and, where applicable, with other pesticides in the final tank mix. 3.4. Pest Biology
The two most important aspects of baculovirus application are temporal and spatial. However effective a virus isolate is in the laboratory, it will fail in the field if it is not applied in the right place and at the right time. Factors relating to pest biology and behavior can be crucial to the success of pest-control programs. Knowledge of insect behavior on the crop after hatch, its distribution within the crop canopy in each instar and the area of foliage ingested per instar will allow precise and effective use of viral pathogens. The optimal quantity of virus required per unit area of host plant can be estimated and used together with the behavioral information to select the most appropriate application methodology and strategy (see Sections 3.5. and 3.6.). For Lepidopteran pests, the damage is usually done in the larval stage, with most being carried out in the final two instars. To minimize crop damage, the baculovirus must be applied as early as possible so that death occurs before the insect reaches the final instars. If the pest is exposed continually during its life cycle and is thus accessible to the baculovirus spray, then the development of the control program does not usually present a problem. However, many pests are
Baculovirus
Insecticides
287
protected from insecticidal sprays at some time during their life cycles, for example, stem borers, leaf curlers, and soil-dwelling cutworms, which demands that both timing and deposition of virus be very accurate so that the insect ingests a lethal dose of virus before it enters its refuge. In some cases, this may necessitate the development of specialized application techniques. 3.6. Method of Application Baculoviruses need to be ingested by the insect in order to initiate infection. Thus effective application should distribute virus to the pest’s feeding sites and give good coverage, so that the insect’s opportunity for acquiring a lethal dose of virus is maximized. Most baculoviruses are applied as sprays, which has meant that they can be, and are, applied using the same equipment as conventional pesticides. In forests, the normal method for application is from the air using airplanes or helicopters fitted with either hydraulic nozzles on a boom or spinning-disk atomizers. For agricultural crops, most spraying is carried out from the ground using boom and nozzle equipment with either flat-fan or hollow-cone jets. In orchards, plantation crops, and some specialized situations, such as in nature reserves or roadside verges, small-scale application equipment, such as back-pack mist blowers and hand-held sprayers with either hydraulic nozzles or spinning-disk applicators, can be useful. Electrostatic sprayers have received little attention with regard to the application of microbial pesticides, but they have some disadvantages since canopy penetration is not usually very deep. Because it is possible to apply baculoviruses using the techniques mentioned above, there is a tendency to use whatever equipment is available, rather than determining what application criteria are most suitable for the particular crop-pest situation. Failure to obtain adequate viral infection in the field may be the result of ineffectual application, rather than innate problems of the host-virus system itself. Judicious timing of application is part of this process, since larger quantities of virus are needed as the larvae age and decrease in susceptibility, for example, as has been studied in detail for the pine beauty moth, P. flummea (21). In this example, it was found that prehatch virus application was ineffective as compared to posthatch sprays. There was only a small window for application after hatch if virus mortality was to be high and crop damage avoided.
288
Cory and Bishop
The optimal droplet size, droplet density, and dose/drop should be estimated for each particular system, and then the spray equipment developed to deliver it. A key factor in application seems to be spray droplet size. With chemical sprays, small droplets (51-100 pm) give better surface coverage on foliage with a resulting increase in pest mortality (22). This has also been found to be the case with many baculoviruses, presumably because it increases the chances of the insect encountering a PIB-containing droplet, although there are some notable exceptions to this, such as with Helicoverpa control on cotton (1). There is a move toward controlled droplet application using low-volume and ultralow-volume equipment, such as spinning-disk sprayers, which are now commonly used for the application of baculoviruses in forests. It is particularly important to keep tank volume down in the forest situation because of the high costs of aerial spraying and the logistic difficulties of carrying large quantities of fluid to the site. The depth of the crop in forests can also be very large. Thus it is important to consider and measure canopy penetration in relation to the distribution of insects on the crops when deciding on suitable application systems. For some groups of pests, standard spray application is not going to be effective or is undesirable for other reasons. One such example is the group of noctuid pests known as cutworms. Because these insects spend most of their lives hidden in the soil, normal spray application is not particularly effective, and such alternatives as development of baits, usually based on bran, are being investigated instead (22a). Stored product pests represent another special case: water-based sprays may encourage the growth of fungal agents in stores. Thus techniques using dusts are being studied for delivery of viruses to stored product pests. Other application methodologies include the release of infected insects, a technique that has been particularly successful with nonpersistent subgroup C baculoviruses (see Section 3.6.2.). Further novel suggestions include luring adults to light or pheromone traps, where they become covered in a baculovirus dust, which is transferred on to their eggs, and subsequently infecting the crop of parasitoid wasps that have been contaminated with virus solution, who will act as vectors for the baculovirus.
Baculovirus
Insecticides
289
3.6. Application Strategy The majority of baculovirus applications fall into one category, inundativb release, and they are therefore in direct competition with chemical pesticides. Other more long-term control strategies have been investigated and are discussed below. In general, ecologically based control strategies have received little attention, since they require a good understanding of the long-term effects of baculoviruses. The use of baculoviruses as classical biological control agents (where introduction of an exotic virus results in self-sustaining control) is not likely to be possible in many circumstances. This is primarily because in most habitats, the virus will be rapidly removed from the host environment. Virus is only likely to persist in stable crop systems and where there is an element of vertical transmission via the host or alternative hosts. The only example with baculoviruses that can be said to fall into this category is the outbreak of an NPV in the spruce sawfly, Gilpinia bercyniue, in Canada, where an NPV was accidentally introduced into the population (23). The population has not resurged since the viral epizootic, a period of almost 50 yr. Application of baculoviruses at more strategic times to slow down the pest population build-up is a more likely option, and may be a more effective and less costly alternative to other control methods in more stable habitats. Early spraying (i.e., preoutbreak) is now the recommended strategy for Douglas fir tussock moth, 0. pseudosugatu, control in North America. The main strategiesfor applying baculoviruses fall into three categories: 1. Inundative release; 2. Inoculation; and 3. Manipulation of resources. 3.6.1. Inundative Release This strategy involves the application of a very large quantity of virus when the pest has reached outbreak proportions, with the expectation that the control will only be of limited duration. Currently this is by far the most common approach to applying baculoviruses in both forest and agricultural situations. In forests where single, univoltine pests are often the target, one application is usually sufficient. In agricultural crops, where pest complexes are more common and multiple generations can occur throughout the year, applications will have to be more frequent,
290
Gory and Bishop
Alternatives to blanket, high-dose coverage have been suggested,for example, more frequent applications of lower-dose sprays in agricultural situations, but these have been little studied so far. Another suggestion is that of lattice introductions, which makes useof the dispersive capacity of the virus. With information on the rate of dispersal in a specific virus:host ecosystem, the virus could be introduced in strips from which it would spread to the intermediate areas of infestation, thus saving both virus and flying time. 3.6.2. Inoculation This approach is based on long-term regulation where it is anticipated that the effect will not be permanent and will require reintroduction of virus at a later stage. Virus is introduced into the population at low levels with the intention that it will build up within the population over a number of generations and suppress pest numbers. The most successful example of this strategy is control of the rhinoceros beetle, Oryctes rhinoceros in the Pacific region, The baculovirus that has been isolated from 0. rhinoceros is a subgroup C baculovirus. Since representatives of this group have no protein coat, they have limited persistence in the environment and are not suitable for spray application. However, they can be very effectively introduced into the environment by releasing infected hosts. Infected adult beetles continually excrete virus and in this way contaminate feeding and breeding sites, thereby passing on the infection to both adults and larvae. The virus can also be transmitted during mating and in the females this severely reduces fecundity (24). 3.6.3. Manipulation of Resources The natural body of baculovirus inoculum in the environment is manipulated in such a way that it is brought into repeated contact with the pest and so maintains numbers at economically acceptable levels. This strategy is only suitable in more permanent habitats and makes use of the persistent capacity of the virus. This technique requires an intimate knowledge of the dynamics of the baculovirus:host relationship and has as yet only been applied successfully to the lepidopterous pasture pests Wiseana spp. in New Zealand. Where the pasture was more frequently ploughed, pest outbreaks were more common because any virus remaining in the soil was rapidly removed from the pest environment. By instigating a minimal disturbance regime, the virus pool could be manipulated so that it was available to infect Wiseane, and in this way, the pest was kept below economically damaging levels (25).
Baculovirus
Insecticides
291
4. Results Although baculoviruses have been isolated from several orders of insects, they have only been used to control pest species from three of them: Lepidoptera (from which the majority of baculovirus isolations have been made), Hymenoptera (diprionid sawflies) and Coleoptera. Control of the sawflies is particularly successful because of their high degree of sensitivity, the gregarious nature of many species, and the presence of vertical transmission (26). Only one virus has been developed for the control of Coleopteran pests, that of the rhinocerous beetle, 0. rhinoceros (see Section 3.6.2.). This virus has been used very successfully to control &y&es spp. in coconut plantations in the Pacific regions and elsewhere, and its use may well be extended in order to control other susceptible dynastid beetles. Overall, the most successful examples of baculovirus usage are to be found in the control of forest pests, a fact that is mainly related to the higher damage threshold in forests. The development of baculovirus against forest pests is frequently undertaken by government agencies. For instance, in the US and Canada, the forest services have registered baculoviruses for use against the Douglas fir tussock moth, 0. pseudosugata, gypsy moth, L. dispar, pine sawfly, Neodiprion sertifer, and the red headed sawfly, Neodiprion Zecontei , all of which have been used successfully in large-scale control campaigns (27). However, the largest potential market for baculoviruses is in the agricultural sector. In recent years, several small companies have started to produce baculovirus insecticides, but they have yet to take over a significant share of the market. The first baculovirus to be commercialized was the NPV from the corn earworm, H. zea, for use on cotton. After a promising start, the sales of this product fell, primarily because of the increase in use of synthetic pyrethroids among cotton growers. With the development of resistance of pyrethroids, Helicoverpa NPV may again have a role to play in control of this pest. Other baculoviruses that have been looked at in terms of commercial development are the GV of the codling moth, Cydia pomonella, various Spodoptera NPVs, and the broad host range alfalfa moth, Autographa californica, NPV (27). Many other baculoviruses are being developed on a more local level. A particularly successful program is being carried out in Brazil, where NPV is being used over hundreds of thousands of hectares for the control of the velvet bean caterpillar, Anticarsia gemmatalis. The Chinese have made large collections of
292
Gory and Bishop
native isolates and are developing many of them for control of agricultural pests. Russia has also developed the production of several viruses, mainly for forest pests, to commercial levels. For more information on the use of baculoviruses as control agents, see reviews by Young and Yearian (28), Entwistle and Evans (l), and Huber (29). However, despite the obvious potential of baculoviruses for the control of many agricultural pests, their use is still limited. A key factor in this is their speed of action; because the viral infection necessarily takes time to develop, some feeding damage can occur and in many agricultural crops this is, at present, unacceptable. The issue is more complex, however. Chemical insecticides are easy to use, and result in a rapid and obvious knock-down effect. Baculoviruses require more care and effort in application, including monitoring of the pest populations, and do not give such a dramatic affect. The expectations of farmers need to be changed before they gain acceptance. In terms of commercialization, their narrow host range is seen as a drawback. Equally, the likelihood of developing more ecologically based, long-term control strategies that utilize the persistent effect is also not likely to appeal to the more commercially minded manufacturers of pesticides. Thus, investment in baculoviruses is not as great as for other control methods, and their development is primarily left to publicly funded institutions. However, the current awareness of the dangers of using broad-spectrum, persistent chemical pesticides and the problems of pesticide residues in foodstuffs, together with the general move towards more environmentally friendly methods of pest control may well result in recognition of the great potential of baculoviruses, and we may see a more serious development of natural baculovirus insecticides in agriculture. References 1. Entwistle, P. F. and Evans, H. F. (1985) Viral Control, in Comprehensive Insect Physiology, Biochemistry and Pharmacology (Kerkut, G. A. and Gilbert, L. I., eds.), Pergamon Press, Oxford, pp. 347412. 2. Cot-y, J. S. (1991) Release of genetically modified viruses. Rev. in Med. Virol. 1, 79-88. 3. Possee, R. D., King, L. A,, Weitzman, M. D., Mass, S. D., Hughes, D. S., Cameron, I. R., Hirst, M. L., and Bishop, D. H. L. (1992) Progress in the genetic modification and field-release of baculovirus insecticides. Proceedings of Second International Conference on the Release of Genetically-Engineered Microorganisms (REGEM 2).
Baculovirus
Insecticides
293
3a. Cory, J. S., Hirst, M. L., Williams, T., Hails, R. S., Goulson, D., Green, B. M., Caley, T. M., Possee, R. D., Cayley, P. J., and Bishop, D. H. L. (1994) Field trial of a genetically improved baculovirus insecticide. Nature 37,138-140. 4. Martigoni, M. E. and Iwai, P. J. (1986) A Catalog of Viral Diseases of Insects, Mites and Ticks. 4th ed. USDA, Forest Service, General Technical Report, PNW-195. 5. Gettig, R. R. and McCarthy, W. J. (1982) Genotypic variation among wild isolated of Heliothis spp. nuclear polyhedrosis viruses from different geographic regions. Virology 117,245-252.
6. Easwaramoorthy, S. and Cory, J. S. (1990) Characterization of the DNA of granulosis viruses isolated from two closely related moths, Chilo infuscatellus and C. sacchriphugus indicus. Arch. Virol. 110,113-l 19. 7. Finney, D. J. (1978) Statistical Methods in Biological Assay. 3rd ed. Charles Griftin & Co., London. 8. Hughes, P. R., and Wood, H. A. (1986) In vivo and in vitro bioassay methods for baculoviruses, in The Biology of Baculoviruses, vol. II (Granados, R. R. and Federici, B., eds.), CRC, Boca Raton, pp. l-30. 9. Singh, P., and Moore, R. F., (1985) Handbook of Insect Rearing, ~01s. I and II. Elsevier, Amsterdam. 10. Hughes, P. R. and Wood, H. A. (1981) A synchronous peroral technique for the bioassay of insect viruses. J. Invert. Pathol. 37,154159. lOa.Smits, P. H. and Vlak, J. M. (1988) Biological activity of Spodopteru exigua nuclear polyhedrosis virus against S. exigua larvae. J. Invert. Puthol. 51,102-l 14. 11. van Beek, N. A. M. and Hughes, P. R. (1986) Determination of fluorescence spectroscopy of the volume ingested by neonate lepidopterous larvae. J. Invert. Pathol. 48,249-25 1. 12. Ignoffo, C. M. and Couch, T. L. (1981) The nucleopolyhedrosis virus of Heliothis species, in Microbial Control of Pests and Plant Diseases 1970-1980 (Burges, H. D., ed.), Academic, London, pp. 329-363. 13. Lewis, F. B (1981) Control of the gypsy moth by a baculovirus, in Microbial Control of Pests and Plant Diseases 1970-1980 (Burges, H. D., ed.), Academic, London, pp. 363-377. 14. Shapiro, M. (1986) In vivo production of baculoviruses, in The Biology of Buculoviruses, vol. II (Ganados, R. R. and Federici. B., eds.), CRC Press, Boca Raton, pp. 3 1-61. 15. Carter, J. B. (1989) Viruses as pest control agents, in Management and Control of Invertebrate Crop Pests (Russell, G. E., ed.), Intercept Ltd., Andover, pp. 165-209. 16. Kelly, P. M. and Entwistle, P. F. (1988) In vivo mass production in the cabbage moth (Mamestra brassicae) of a heterologous (Panolis) and a homologous (Mamestra) nuclear polyhedrosis virus. J, Virol. Methods 19,249-256. 17. Shapiro, M., Robertson, J. L., and Bell, R. A. (1986) Quantitative and qualitative differences in gypsy moth (Lepidoptera: Lymantriidae) nucleo-polyhedrosis virus produced in different aged larvae. J. Econ. Entomol. 79, 1174-l 177. 18. Kelly, P. M., Speight, M. R., and Entwistle, P. F. (1989) Mass production of Euproctis chtysorrhoea (L.) nuclear polyhedrosis virus J. Virol. Methods 25,93-100.
Gory and Bishop 19. Couch, T. L. and Ignoffo, C. M. (1981) Formulation of Insect Pathogens In: Microbial Control of Pests and Plant Diseases 1970-1980. (Burges, H. D., ed.), Academic, London, pp. 165-209. 20. Young, S. Y., III and Yearian, W. C. (1986) Formulation and application of baculoviruses, in The Biology of Baculoviruses, vol. II (Granados, R. R. and Federici, B., eds.), CRC, Boca Raton, pp. 157-179. 21. Cory, J. S. and Entwistle, P. F. (1990) The effect of time of spray application on infection of the pine beauty moth, Panolisjlammea (Den and Schiff) (Lepidoptera: Noctuidae) with nuclear polyhedrosis virus. J. Appl. Entomol. 110,235-241. 22. Matthews, G. A. (1979) Pesticide Application Methods, Longman, London. 22a.Bourner, T. C., Vargas-Osuna, E., Williams, T., Santiago-Lavarez, F., and Cory, J. S. (1992) A comparison of the efficacy of nuclear polyhedrosis and granulotis viruses in spray and bait formulations for the control of Agro segetum (Lepidoptera. Noctuidae) in maize. Biocontrol Sci Technol. 2,315-326. 23. Neilsen, M. M., Martineau, R., and Rose A. H. (1971) Diprion hercyniae (Hartig), European spruce sawfly (Hymenoptera: Dipriondae); Biological control programs in Canada 1959-1968 Tech. Bull. No 4 Commonw. Inst. Biol. Control, 136-143. 24. Bedford, G. 0. (198 1) Control of the rhinoceros beetle by baculovirus, in Microbial Control of Pest and Plant Diseases (Burges, H. D., ed.), Academic, London, pp. 409-426. 25. Kalmakoff, J. and Crawford, A. M. (1982) Enzootic virus control of Wiseana spp. in a pasture environment, in Microbial and Viral Pesticides (Kurstak, E., ed.), Marcel Dekker, New York, pp. 435448. 26. Cunningham, J. C. and Entwistle, P. F., (1981) Control of sawflies by baculoviruses, in Microbial Control of Pests and Plant Diseases 1970-1980 (Burges, H. D., ed.), Academic, London, pp. 279-407. 27. Cunningham, J. C. (1988) Baculoviruses: their status compared to Bacilus thuringiensis as microbial insecticides. Outlook on Agriculture 17, 10-17. 28. Young, S. Y., III. and Yearian, W. C. (1982) Control of insect pests of agricultural importance by viral insecticides, in Microbial and Viral Pesticides (Kurstak, E., ed.), Marcel Dekker, New York, pp. 387-423. 29. Huber, J. (1986) Use of baculoviruses in pest management programs, in The Biology of Baculoviruses, vol. II (Granados, R. R. and Federici, B., eds.), CRC, Boca Raton, pp. 18 l-202.
CHAPTER17
Molecular Approaches to AIDS Vaccine Development Using Baculovirus Expression Vectors C. Yong Kang 1. Introduction The acquired immune deficiency syndrome (AIDS) is caused by human immunodeficiency virus type 1 or 2 (HIV-l or HIV-2). The disease is characterized by a high susceptibility to opportunistic infections or to malignant diseases, such as Kaposi’s sarcoma. The major immunologic abnormality is the selective depletion of CD4+ T-cells together with decreased capacity to secrete or respond to lymphokines, such as interlet&in 2. During infection, HIV induces both a humor-al and cell-mediated immune response to its viral components. However, the response cannot cope with progressive infection. An antibody response to the virion surface proteins, env (gp120, gp41), and the core protein, gag (~24, p17), is observed in most instances 4-8 wk after infection with the virus. The antibody response clears the virus from the circulation, but does not eliminate cells that become latently infected. As infected individuals progress from the asymptomatic state to produce AIDS-related complex (ARC) and subsequently to full-blown AIDS, anti-p24 titers are reduced and the viral antigen appears in the circulation of patients. The significance of this reduction in p24 antibody for disease progression is presently not known. High levels of antibodies to env gene products are also found in HIV-l positive patients, but no real relationship has been noted between their titers or their neutralizing capacities and progression to AIDS (1,2). Therefore, vaccination of individuals with env protein alone From:
Methods in Molecular Biology, Vol. 39: Baculovirus Expression Protocols Edited by: C. D. Richardson 0 1995 Humana Press Inc., Totowa, NJ
295
may be successful in eliciting an antibody response,but may not be good enough to control the disease process. We have also discovered that the anti-vifresponse is only observed in seraof asymptomatic patients (3). The antibodies in patients with ARC and AIDS have no detectable ant&if activity, as measured by Western blot. A number of laboratories speculate that the loss of anti-vifantibody correlates with disease progression, A cell-mediated immune response to the various viral antigens also appearsduring HIV- 1 infection. Two cell-mediated immune mechanisms are thought to play important roles in the clearance of virus infected cells, antibody-dependent cellular cytotoxicity (ADCC), and cytotoxic T-cell (CTL) responses, which are part of the cell-mediated immune response. In viral infections, ADCC seemsto be responsible for the lysis of infected cells that have bound antiviral antibody directed against gp12O/gp160 expressed on the cell surface (4). In another study, it was found that antibody reactivity with p24 nucleoprotein of the virus correlated better with ADCC activity than reactivity to gp120/160 (.5), suggesting that antibody to this internal protein might also play a role in ADCC. Patients with AIDS have lower ADCC activity than sera from HIV positive healthy individuals (5-8) whereas others have found no significant reduction in ADCC activity of effector cells isolated from AIDS patients (9). The cytotoxic T-cell response to both envelope and internal proteins of viruses, in general, have been shown to be important in recovery from diseaseand possibly in elimination of virus from the host (10,11) Walker et al. (12,13) have demonstrated that cytotoxic T-cells to env, gag, and pal can be detected in AIDS patients, but how those responsesrelate to diseasehas not been elucidated. This information is very important in understandingthe pathogenesis of the diseaseand can serve to help in vaccine development. The goal of my group is to develop an AIDS vaccine through the introduction of HIV-specific structural proteins into the body in order to induce both humoral and cell-mediated immune responses. Our approach is to produce a vaccine containing viral surface proteins, internal structural proteins, and proteins involved in transcription and replication of HIV. Majority of candidate vaccines, already licensed for human clinical trial by the US Food and Drug Administration, contain an HIV surface glycoprotein, gp 120, or its precursor gp 160, which may produce neutralizing antibodies against HIV. However, it is not clear if such a vaccine will work because some apparently healthy, yet HIV-infected individuals who already have detectable levels of antibodies against surface proteins still develop
Expression
of AIDS-Related
Proteins
AIDS. We will assessthe properties of these proteins in terms of being superantigens, since this class of antigen can causesdepletion of specific T-cell subsets (14). In the subsequent work, we have expressed high levels of the gag precursor, envelope (env), polymerase (pal), and vifproteins from both HIV-l and HIV-2 in Spodopteru Frugiperdu (Sf9) cells using recombinant Autographa cluifomicu nuclear polyhedrosis virus. 2. Materials 1. Sf9 insect cells can be obtainedform the American Type Culture Collection (Rockville, MD). 2. TClOO medium or Grace’s supplementedmedium can be obtained from GIBCO Life Technologies(GrandIsland, NY). 3. The pAcYM1 vector camefrom David Bishop (15). 4. SeaPlaqueagarosewaspurchasedfrom PMC Bioproducts(Rockland,ME). 5. Pooled HIV-positive human antisera were obtained from the National Institutes of Allergy and Infectious Diseases,AIDS Researchand ReferenceReagentProgram (Bethesda,MD). 6. [35,!+Methioninecanbe purchasedfrom Amersham(Arlington Heights,IL). 7. The plasmid plBM containing gag-pal came from R. C. Gallo (NIH, Bethesda,MD). 8. Plasmid HXB-2D containing the entire HIV-l genome(16) was obtained from R. C. Gallo (NIH, Bethesda,MD). 9. Acrylamide, bis-acrylamide, SDS, TEMED, and TRIS base, were purchasedfrom Bio-Rad (Richmond, CA). 3. Methods The structural protein genes of HIV-l and HIV-Z have been expressed in large quantity in Sf9 cells using a baculovirus expression system. In the baculovirus expression system, the polyhedrin gene of the baculovirus is replaced with a gene from HIV to make a recombinant baculovirus that in infected cells directs production of desired proteins. As a result, infected Sf9 cells produce large quantities of group-specific antigens, gag, reverse transcriptase, pal, envelope glycoprotein, gp120, and vifprotein. 3.1. Expression of Gag Protein The construct designed to express the HIV-2 gag precursor proteins is shown in Fig. 1. The transfer vector pAcYMl-gag was constructedby using crossover linker mutagenesis in order to achieve high-level expression of the HIV-2 gag protein precursor. The crossover linker mutagenesis is a convenient method (2 7,18) to delete noncoding flanking sequencesand to add essential sequences, such as the putative ribosome binding site.
Kang
298
,:$I PlBM
Cut wth crossover
Sphl B dephosphorylation linker IlgatiOn & transformation
,I’ ,’ oas
,’
I ,,*’
:’
Isolation of Stul Sacll fragment Insert into Hmcli site of pUC19
I Cut with crossover
EcoRl linker
and Kpnl llgatlon & transformation
lsolatlon
of EsmHl
IWO
Insert
Into
fragment ,IlU,
pAcYMl
Fig. 1. Construction of recombinant baculovims containing the HIV-2 gag gene. The gag gene was isolated from plBM (23), which contains the entire HIV-2NIH-Z sequence, by StuI and Sac11digestion, A 1.6-kb S&I - Sac11fragment of the plBM genome was purified and subcloned into the HincII site of pUC 19. pUC19-gag was digested with EcoRI and KpnI, and ligated with a synthetic oligonucleotide crossover linker containing an EC&I sticky end, a BarnHI site, a putative Sf9 ribosome binding site (P), and translation initiation codon ATG (TI), followed by 11 additional nucleotides from the coding sequence representing the N-terminus of the gag protein. pUC19-gag was further modified to delete the gag-pal overlapping sequencesand to add translation termination codon (TT) and a BamHI site. The pUCl9-gag was digested with SphI and ligated with another crossover linker containing the translation termination codon (TAG) (TT), which deletes 93 amino acids at the C terminus of the gag protein, and introduces a BamHI site and an $hI sticky end. The BamHI
Expression
of AIDS-Related
Proteins
Recombinant baculovirus-expressing gag has been isolated, and the protein has been shown to produce 100 nm particles at the surface of infected insect cells that are secreted into the culture media. 1. The plasmid plBM containing the gag-pal genes of HIV-2 was cut with StuI and Sad, and a DNA fragment containing gag and a small portion of pal at the 3’ terminus were ligated into the HincII site of pUC 19 (Fig. 1). pUC19gag was cut with &&I and KpnI and ligated with the 5’ oligonucleotide containing EcoRI/~amHI/SfP cell ribosome binding site/initiation codon/ and a short piece of the adjacent gag sequence.The duplicated gag sequence forms a loop structure that is removed by crossover repair following transformation into E. coli. The modified pUC19-gag plasmid containing the new ATG was subsequently modified at the 3’ end of the gene in a similar manner using an oligonucleotide containing SphIIBamHYtermination codon TAG/a short piece of adjacent 3’ gag sequence. In this way, the 5’ terminus of the pal gene was deleted. The modified gag gene was cut from pUC 19 as a BarnHI fragment and ligated into the BamHI cloning site of the baculovirus expression vector, pAcYM1. This construct contains a 1278-bp open reading frame of the gag gene which can code for 425 amino acids. 2. Cotransfection of pAcYMl-gag and wild-type AcNPV viral DNA is performed with methods described in Chapter 6 of this book. Occlusion bodynegative plaques were isolated by visual screenmg. 3. Sf9 cells are inoculated with the recombinant baculovirus (AcNPV-HIV-2 gag) and the infection was allowed to proceed for 1,2,3, and 4 d. 4. Sf9 cells (105) are harvested at the various times, lysed with SDS sample buffer, subjected to electrophoresis in 12.5% acrylamide gels containing SDS, and stained with Coomassie blue. 5. Lysate proteins can also be resolved by SDS-polyacrylamide electrophoresis (PAGE), blotted onto nitrocellulose, and detected with Western blots using pooled antisera from AIDS patients. 6. A band migrating at 41 kDa is seen in the extracts of Sf9 cells infected with the recombinant virus; this band is not seen m extracts from wild-type AcNPV or mock-infected cells. 7. The time-course experiment reveals that gag pr41 does not accumulate in infected cells. Electronmicroscopy also demonstrates that gag particles (100 nm diameter) budded from the surface of infected insect cells (19). fragment was inserted into the BamHI site of the transfer vector pAcYM1 (15). The transfer vector pAcYMl-GAG was used to cotransfect Sf9 cells with wildtype AcNPV DNA and recombinant baculovirus, AcNPV-HIV-2gag, was isolated. Modified from Luo et al. (19).
Kang
300
B
A MCWl23
MCW
12
3
9266423l2114-
Fig. 2. gag Protein expression in recombinant AcNPV-HIV-2 gag virusinfected Sf9 cells and the detection of extracellular gag particles in cultured supernatant. The lysate of AcNPV-HIV-2 gag virus-infected Sf9 cells from 3 d postinfection was subjected to SDS-PAGE in 12.5% polyacrylamide gels, and proteins were stained with Coomassie Blue. Sf9 cells were infected with recombinant baculoviruses at an MO1 of 5 PFU/cell and incubated at 27OC for 72 h. The cell culture supernatant was collected after centrifugation at 1OOOgfor 20 min. Secreted particles in the culture supernatant were collected by ultracentrifugation at 100,000g for 1.5 h, resuspended in PBS containing 0.1% Tween20, 10 pg/mL aprotinin, and left at 4OC overnight. Coomassie blue-stained SDS-PAGE of intracellular and extracellular gag protein is shown in panel A. Panel B shows Western blot analysis of the gag protein with pooled serum from AIDS patients. gag Particles were isolated from a 2060% sucrose gradient after treatment with 0.1% Tween-20 and 10 p&nL aprotinin at 4°C overnight. M, marker protein; C, cell control; W, wild-type AcNPV-infected cells; Lane 1, cell lysate infected with AcNPV-HIV-2 gag virus; Lane 2, extracellular gag particles present in the pellet obtained by ultracentrifugation of cultured supernatant; Lane 3, purified gag particles. The arrow indicates the major gag pr41 protein.
Sf9 cells are infected with recombinant virus at a multiplicity of infection of 5 PFU/cell and incubated at 27°C for 72 h. The cell culture supernatant is collected after centrifugation at 1OOOgfor 20 min. 8. Particles in the clarified culture supernatant are collected by ultracentrifugation at 100,OOOgfor 1.5 h, and the pellets are resuspended in PBS containing 0.1% Tween 20 and 10 pg/rnL aprotinin at 4°C overnight. The suspension is subjected to SDS-polyacrylamide electrophoresis, and gels are either stained with Coomassie blue or analyzed with Western blots using antisera from AIDS patients (Fig. 2). The gag precursor protein (41 kDa) and a 26-kDa degradation product are evident.
Expression of AIDS-Related
Proteins
Fig. 3. Transmission electron microscopy of HIV-2 gag particles. (A) Pelleted material from infected Sf9 cell supernatants showing both gag and baculovirus particles (arrow B shows baculovirus in longitudinal and transverse section). (B) Thin section of purified gag particles from 20-60% sucrose gradient. (C) Negative staining of purified gag particles fixed with glutaraldehyde and osmium tetroxide. The white bars in the figure represent 100 nm. 9. Pelleted gag particles are purified by ultracentrifugation on a 20-60% (w/v) sucrose gradient at 100,OOOgovernight. Two well-separated bands are evident; the higher band sediments at about 40% sucrose, and the lower band sediments at about 52% sucrose. Electronmicroscopy reveals the upper band to be the gag p41 particles whereas the lower band contains particles that had the typical rod-like structure of baculoviruses (Fig. 3). The identity of the upper band can be confirmed biochemically through PAGE and Western blot analyses (Fig. 2). 10. The level of production of total gag p41 is estimated to be 30-50 mg/5 x lo* cells/L of culture fluid (see Note 1). 3.2. Expression of pol Protein We have used the same strategy to express thee different cassettes of
the pal gene: 1. A deleted polymerase (Dpol) cassette containing a deleted protease, the reverse transcriptase, and integrase coding regions was constructed first as shown in Fig. 4 starting from the plasmid pHXB-20. This cassettestarted from the first AUG codon in the pal open reading frame, deleting 103
302
Kang
pAcVM1
Fig. 4. Construction of the HIV-1 pal open reading frame. The Dpol cassette shows deletion of 273 bp from the 5’ terminus of the pal open reading frame. The &$I1 -MI fragment of plasmid pHXB-2D (16) containing the HIV-l pal coding region was isolated and inserted into BumHI and MI sites of pUC18. The resulting recombinant plasmid pUC18Dpol 1 was cut with S&I and dephosphorylated. A synthetic double-strand crossover linker containing an SstI cohesive end, a BarnHI site, the putative ribosome binding site, CCTATAAAT, which was derived from nucleotides -9 to - 1of the polyhedrin gene (19), and 15 nucleotides of homology searching sequence that overlap with the 5’ terminus of the Dpol gene was ligated to the SstI site, and the resulting construct was used to transform E. coli. Recombinant plasmid (pUC18Dpol2) was isolated, digested with S’hI, dephosphorylated, and ligated with another crossover linker DNA containing SphI cohesive end, a BarnHI site, and 15 nucleotides of a homology searching sequence that recognizes the 3’ terminus of the pal gene. The resulting recombinant plasmid (pUC1 S-Dpol3) contains the putative ribo-
Expression
of AIDS-Related
Proteins
303
amino acids at the amino terminus of the protease, but retaining 64 amino acids at the carboxyl terminus of the protease. In addition, we also included the putative ribosome binding site CCTATAAAT in front of the translation initiation codon. To make this construct, the BgZII-SuZI fragment containing the HIV-1 pol coding region was isolated and inserted into the BarnHI and SuZI sites of pUC18. The resulting plasmid was cut with S&I, the ends were dephosphorylated, and the linearized construct was annealed to an oligonucleotide containing SstIIBamHIZSf9 cell ribosome binding site (CCTATAAAT)/initiation codon/ nucleotides of the 5’ terminus of the Dpol gene. Deletion of the 5’ noncoding sequences (i.e., 103 amino acids of the protease) was accomplished by crossover linker ligation. Similarly, the noncoding sequences of the pal gene at the 3’ end were removed by crossover linker ligation with an oligonucleotide containing SphYBumHY termination codonZl5 nucleotides from the 3’ terminus of the Dpol gene. A BumHI fragment containing the coding sequenceof Dpol was isolated from pUCl8-Dpo13 and inserted into the BumHI site of pAcYM1. Through this process, we deleted all noncoding sequencesat both 5’ and 3’ termini of the pal gene, and added BumHI restriction sites at either end. This cassette contains genetic information for 912 amino acids. 2. When we express the Dpol cassettein Sf9 cells, large quantities (approx 80-100 mg/5 x lo* cells/L culture) of unprocessed lOO-kDa protein accumulate in recombinant AcNPV-infected Sf9 cells (see Fig. 7 later in this chapter). The lOO-kDa protein is immunoreactive with antibodies present in HIV-l and HIV-2 positive human sera (Fig. 7b) (22). 3. The full-length (Fpol) cassette contains the full-length pal open reading frame, which consists of the virus-specific protease, reverse transcriptase, RNase H, and the endonuclease/integrase (Fig. 5). Since the pal open reading frame does not have its own translation initiation site, we inserted an ATG in front of the pal open reading frame. The BgZII-Sal1 fragment of pHXB-2D was isolated and ligated with a synthetic double-stranded DNA oligonucleotide linker consisting of a translation codon and 12 nucleotides from the protease gene missing upstream of the BgZII site, a ribosome binding site (CCTATAAAT), a BumHI site, and an SstI cohesive end, and inserted into pUC18 to give pUCl&Fpoll. A 900-bp SstI-EcoRV fragsome binding site (P) followed by the pal open reading frame starting with the first ATG (TI) in the pal gene and ending with the translation termination codon TAG (TT). This Dpol cassette is flanked with BumHI sites. The BumHI fragment was isolated, inserted into the BumHI site of the pAcYM1 baculovirus transfer vector (15), and pAcYMl-Dpol DNA was used to cotransfect Sf9 cells along with wild-type AcNpV DNA to produce recombinant baculovirus.
304
Kang
Fig. 5. Construction of recombinant AcNPV HIV- 1pal containing the entire open reading frame. The full-length pol gene (Fpol) was constructed as follows: the BgZII-SalI fragment of plasmid pHXB-2D was isolated and ligated with a synthetic double-stranded DNA linker that provides 12 nucleotides missing immediately upstream from the BgZII site plus a translation initiation codon (TI), a putative ribosome binding site (P), a BarnHI site, and an &I cohesive end. The DNA was inserted into the &I and SalI sites of pUC18. The resulting recombinant plasmid (pUC 1S-Fpol 1) was cut with SstI and EcoRV, and a 900bp fragment was isolated. The 900-bp SstI-EcoRV fragment and a 5.4-kbp SstIEcoRV fragment of pUClS-Dpol3 (Fig. 4) were ligated and used to transform E. coli. The resultant plasmid (pUCl&Fpol2) contains the putative ribosome binding site (P) followed by the newly introduced translation initiation codon (TI), the full-length pal open reading frame, and the translation termination codon (IT), and flanked by a BumHI site at either end. This cassette(Fpol) was isolated and inserted into the BumHI site of the pAcYM1 baculovirus transfer vector and was used to cotransfect Sf9 cells along with wild-type AcNPV DNA to produce recombinant baculovirus.
Expression of AIDS-Related
4.
5.
6.
7.
Proteins
ment was inserted in place of the SstI-EcoRV fragment in pUCl%Dpol3 to yield pUCl8-Fpo12. The BumHI fragment from this plasmid was ligated into pAcYM1, cotransfected into Sf9 cells with AcNPV DNA, and recombinant virus expressing Fpol was isolated. The resulting recombinant AcNPV Fpol virus is capable of synthesizing a product of 1016 amino acids.When we expressthe Fpol genein Sf9 cells using recombinant AcNPV, we cannot detect aprecursor polyprotein band by staining with Coomassieblue (Fig. 7A). However, Western blot analysis with HIVpositive human sera shows 66-, 51-, and 34-kDa processed proteins (Fig. 7B). This result suggeststhat the HIV protease must be active and cleaves the precursor polyprotein to produce the final products ~66, p5 1, and ~34. We observe that cells infected with recombinant baculovirus containing Fpol cassette lyse early in infection, We believe that HIV protease is extremely toxic to the cells, and therefore, cells cannot synthesize large quantities of the polyprotein. A cassettecontaining only the reverse transcriptase (RT) has a deletion of both protease and endonuclease/integrase genes (Fig. 6). It is able to code for a protein of 563 amino acids that contains reverse transcriptase and RNase H activities (20) (Fig. 6). The plasmid PUC18Dpo13 was cut with S&I, dephosphorylated, and the 5’ terminal protease portion of the gene was removed with crossover linker mutagenesis using an oligonucleotide containing an SstVBumHVSf9 ribosome binding site (CCTATAAAT), an initiation codon, and 15 nucleotides from the 5’ end of the RT. The integrase from the 3’ end was removed by crossover linker mutagenesis using an oligonucleotide containing an sph1 cohesive end, a BarnHI site, a termination codon, and 15 nucleotides with homology to the 3’ end of the RT. The BarnHI fragment containing the RT was inserted into pAcYM1 and used to generate AcNPV-RT recombinants. AcNPV-RT expresseslarge amounts (approx 100 mg/5 x lo6 cells/L) of a 66-kDa protein in St9 cells that is recognized by antibody present in HIVl-positive human sera (Fig. 7B) (see Note 2).
3.3. Expression
of gp120 env Protein
1. Using the same strategy of construction (Fig. 8), we have expressed gp120 and gp130 of HIV-l and HIV-2, respectively. The plasmid pHKB-2D, which contains the entire HIV-l genome, was used as a source of envelope glycoprotein coding sequence (16). A 2.1-kb ,.%~I-ZGX!III fragment from pHKB2D was isolated, blunt-ended with Klenow fragment and T4 DNA polymerases, and inserted into the HincII site of pUC19. The resulting recombinant plasmid, pUC19-Env, was digested with EcoRI and ligated with a synthetic oligonucleotide crossover linker that contained an EcoRI sticky
306
Kang
crossover
cut with
crossover
Sstl
lmker
knker
kgation
& transformation
8 dephosphorylatlon
hgatlon
8 transformatton lsolatlon
of Barn HI fragment ,42341
Insert
cut
wtth Sphl
mto PAcYMI
& dephosphorylatroq
Fig. 6. Construction of recombinant AcNPV containing 66-kDa RT gene. The pUC 18-Dpol3 from Fig. 4 was cut with SstI, dephosphorylated, and crossover linker mutagenesis was used to remove the 5’ terminal protease portion of the geneand to add a BarnHI site, a putative ribosome binding site (P), and a translation initiation codon (TI) in front of the RT coding sequences. The resulting plasmid pUC18-Dpol4 was then cut with ,!$&I, dephosphorylated, and ligated with another crossover linker containing an SphI cohesive end, a BumHI site, translation termination (TT) codon, and 15 nucleotides of homology searching sequence that recognizes the 3’ terminus of the RT gene. The recombinant plasmid (pUC18-RT) contains the putative ribosome binding site (P) followed by a new translation imtiation site (TI), the RT open reading frame, which also contains C-terminal RNase H coding sequences ending with a translation termination codon (TAA) (‘IT). This RT cassetteis flanked with BarnHI sites. The BumHI fragment was isolated, inserted into the BumHI site of the pAcYM1 baculovirus transfer vector, and pAcYMl-RT DNA was used to cotransfect Sf9 cells along with wild-type AcNPV DNA to produce recombinant baculovirus.
Expression
of AIDS-Related
Proteins
307
Fig. 7. Expression of pal genes in Sf9 cells by recombinant baculoviruses. Sf9 cells infected with recombinant baculoviruses were denatured and electrophoresed in 12.5% polyacrylamide gels. The proteins in gel were visualized by Coomassie Blue staining (A) or Western Blot analysis (B) using standard NationallInstitutes of Health HIV-positive human sera. Lanes 1,2, and 3 represent the Sf9 cells harvested 72 h after infection with recombinant baculovirns carrying Fpol, Dpol, and RT gene cassettes,respectively. Lanes 5, 6, and 7 show the Sf9 cells harvested 96 h after infection using the same set of clones. Lanes 4 and 8 show wild-type AcNPV infected cell lysate. P designates polyhedrin protein. Lane M contains molecular mass markers shown in kDa. Lanes 1 and 5 in Panel B show processed proteins of 66-, 51-, and 34-kDa processed pal gene products. end, a BamHI site, ribosome binding site (CCTATAAAT), initiation codon, and 12 nucleotides of homology with the 5’ terminus of the pUC19-Env coding region. The gp41 coding sequence adjacent to the 3’ end of gp120 was deleted by crossover linker mutagenesis using an oligonucleotide containing a Hind11 stickly end, an internal BumHI site, a translation termination codon, and 12 nucleotides with homology to the 3’ terminal residue of gp120. The gp120 coding region was cut and removed with BumHI and inserted into pAcYM1, cotransfected with AcNPV DNA into Sf9 insect cells, and used to generate recombinant AcNPV-gp 120. 2. Figure 9 shows the nonglycosylated form of gp120 and gp130 of HIV-l and HIV-2, respectively (see Note 3). We have partially purified these
308
Kang I
Xbd
Filled Insert
with RIbnow 614 DNA polymersw into Hlncll Site 01 pUC1S I Insert
into
PAcVMI
Fig. 8. Construction of recombinant baculovirus containing the HIV-l gp120 gene. The plasmid pHXB-2D containing an entire HIV-l genome was used as a source of HIV-l envelope glycoprotein coding sequence (16). A 2.1 kb SstI &&III fragment from pHXB-2D was isolated, filled with Klenow and T4 DNA polymerase, and inserted into the HincII site of pUC19, The resulting recombinant plasmid, pUC19-Env was digested with EcoRI and ligated with a synthetic oligonucleotide crossover linker containing an EcoRI sticky end, a BumHI site, putative ribosome binding site (P), and the translation initiation codon ATG (TI), followed by 12 additional nucleotides of homology searching sequence that overlap with the 5’ terminus of the pUC19-Env gene. pUC19-Env was further modified to delete the remaining gp41 env coding sequence in the pUC19Env gene, and an inframe stop codon was introduced. The pUC19-Env was digested with Pst I and Hind III, and ligated with another crossover linker DNA containing a Hind III sticky end, an internal BarnHI site, a translation termination codon (‘IT), and 12 nucleotide of homology searching sequence that recognizes the 3’ terminal residues in the gp120 gene. The resulting recombinant plasmid contains the putative ribosome binding site (P) followed by the gp120 open reading frame, starting with the first ATG (TI) at the beginning of the signal sequence of the env gene and ending with the translation termination codon TAA (TT). This cassettewas inserted into pAcYM1 (pAcYMl-Env) and used to cotransfect S.frugiperda cells to isolate recombinant AcNPV.
Expression
of AIDS-Related
Proteins
309
I
36
-
I::. :;: \
21-
Fig. 9. Coomassie blue-stained SDS-PAGE of the envelope glycoprotein backbone. Lanes 1,2, and 3 show nonglycosylated gp120 of HIV-l (arrow with l), and lanes 5 and 6 show nonglycosylated gp120 of HIV-2 (arrow with 2). The HIV-2 gp120 was expressed by a recombinant baculovirus containing the HIV-2 glycoprotein gene that was constructed with the same strategy as shown in Fig. 8. cell-associated glycoprotein precursors and used them to immunize rabbits. Antirabbit sera against nonglycosylated gp120 of HIV-l neutralizes HIV-l infectivity (data not shown). We have subsequently included this protein as one of the components of an AIDS vaccine. 3.4. Expression
of vif Protein
In addition to structural proteins of HIV-l and HIV-2, we have also constructed recombinant AcNPV containing the @gene of HIV-l using the same strategy of crossover linker mutagenesis (Fig. 10). 1. An EC&I fragment was isolated from a plasmid containing the entire HIV-l genome (pHXB); the fragment was rendered blunt-ended by filling it in with Klenow fragment DNA polymerase. 2. The blunt-ended fragment was inserted into the HincII site of pUC19 to produce pUC 19-vifl . 3. The plasmid pUC19-vifl was cut with BumHI and XbaI, and ligated to an oligonucleotide consisting of SmaUBamHYSf9 ribosome binding site (CCTATAAAT)/initiation codon/ nucleotides of the 5’ end of vi’ coding sequence; transformation of E. coli was followed by deletion and crossover linker mutagenesis to yield pUC19-vi$L
Kang
310
Cut with Pat: & HindIll
c4ussover
linker
1 ligatlgn
A trensformetion
I ~solatlon
of EcoRl
fragment
I I
cut with Bamlil
Crossover
lmkw llgatlon
II fill m with Klenow
hXbal
& tmnsformatlon
Fig, 10. Construction of recombinant AcNPV containing the HIV-l vu gene. A EcoRI fragment was isolated from the infectious clone pHXB-2D and filled in with Klenow. The fragment was inserted into the HincII site of pUC19 to produce pUC19-vifl. pUC19-vifl was cut with BumHI and XbaI and a double-stranded crossover linker was used to construct pUC19-vifl. The downstream sequence was modified by cutting pUC19-~$2 with PstI and Hind III and further crossover linker mutagenesis. The resulting plasmid (pUC19-vif3) contains the putative ribosome binding site (P) and open reading frame of vifgene, which ends with the translation termination codon TAG. This cassette was flanked with BumHI sites. The BumHI fragment was inserted into pAcYM1 (pAcYMl-vij). The plasmid pAcYMl-vif was cotransfected with wild-type AcNPV DNA to isolate the recombinant baculovirus containing the vif gene.
Expression of AlDS-Related
Proteins
311
4. The 3’ of the vif coding region was modified by cutting pUC19-$2 with P&I and HirzmII, ligating it to an oligonucleotide consisting of Hin&II/ BamHIltermination codon/l2 nucleotides of the 3’ end of vif coding sequence to give pUCl9-vifl. 5. A BumHI fragment was removed from pUC19-vif3 and inserted into pAcYM1, the plasmid was cotransfected into Sf9 cells with AcNPV DNA, and recombinant virus was isolated. 6. Figure 11 shows expression of the vifprotein of HIV-l in Sf9 cells. The vif proteins of HIV-l and HIV-2 are extremely stable in recombinant baculovirus infected Sf9 cells. 7. The infected Sf9 cells synthesize vi’protein until 3 d postinfection and vifprotein remains in the cells without degradation for a week, whereas most of the cellular proteins degrade (Fig. 12). The level of vifprotein expression is approx 150-200 mg/5 x lo8 cells/L at 4 d after infection (see Note 4).
4. Notes 1. The gag p41 particles are highly immunogenic and produce good levels of antibodies. This protein has previously been shown to induce cytotoxic T-cell response (13,21). 2. The truncated HIV-2 gag gene was fused with the neutralizing domain (V3) of gpl20 env gene sequences from HIV- 1 and HIV-2. These fused genes express chimeric proteins that form virus-like particles. Antisera against these particles neutralize homologous HIV infectivity (24). Our results show that precursor gag protein has potential as a carrier for the presentation of foreign epitopes in good immunological context. The gag protein is highly immunogenic and has the ability to carry large foreign inserts; as such, it offers an attractive approach for HIV vaccine development (24). 3. Since HIV- 1 and HIV-2 have a high degree of sequence homology, HIV- 1 pal precursor protein can interact with HIV-2 positive-patient sera (22). Reverse transcriptase has been shown to induce a cytotoxic T-cell response (13). 4. We have observed that the gp120 polypeptide backbone of HIV-l and the gp130 backbone of HIV-2 can be expressed at high levels if we delete the signal sequences of the protein; however, these proteins are cell-associated, and the majority of the proteins are not glycosylated. 5. We found that antibody against vifis produced early in HIV-l infection, and disappears as disease progresses to ARC and AIDS. Our results suggest that continuous presence of anti-vifantibody may be crucial to stopping progression of the disease. We hypothesize that disappearance of
312
Kang M
Cl
2
3
4
M kDa --66
-22
-14
Fig. 11. Expression of HIV-l vif protein in Sf9 cells by recombinant baculovirus. Sf9 cells infected with recombinant baculoviruses containing the vifgene were harvested 24 (Lane l), 48 (Lane 2), 72 (Lane 3), and 96 (Lane 4) h after infection. Lysates of the infected cells were denatured and electrophoresed in 12.5% polyacrylamide gels. The proteins in gel were visualized by Coomassie Blue staining. Lanes M contain markers. anti-vif antibody is partially responsible for disease progression. Thus, I recommend to include vifprotein as one of the components of an AIDS vaccine. 6. In summary, we have expressed high levels of gag particles, pal proteins, nonglycosylated gp120, and vifproteins of both HIV-l and HIV-2 in Sf9 cells using recombinant baculoviruses. Different combinations of these structural proteins are being tested for their ability to protect nonhuman primates against HIV-2 or SIV infection. Our preliminary results show that the rhesus macaque immunized with the combination of proteins produces high levels of antibodies against these proteins. We are currently investigating the cytotoxic T-lymphocyte responses to these proteins.
Expression
of AIDS-Related
Proteins ,.
,.. .‘.
‘..
313 1’
3
;.
.,.
Fig. 12. Expression of HIV-2 vif in Sf9 cells by recombinant baculovirus. Sf9 cells infected with recombinant baculovirus containing the HIV-2 vifgene were denatured and electrophoresed in 12.5% polyacrylamide gels. A: methionine-labeled total cellular protein analyzed by SDS-PAGE. Lanes 1,2,3,4, and 5 represent 35S-methionine labeled protein on d 1, 2, 3,4, and 5 postinfection. B: lysates of AcNFV HIV-2 vifrecombinant virus infected Sf9 cells from d 1, 2,3,4,5,6, and 7 postinfection were subjected to SDS-PAGE in 12.5% polyacrylamide gels, and the proteins were stained with Coomassie blue. The arrow shows the accumulation of HIV-2 vif protein in recombinant baculovirusinfected cells. Lane W represents the 33-kDa polyhedrin protein from wildtype AcNPV-infected cell lysates. Lane C contains uninfected Sf9 cell lysate. Lane M contains molecular-mass markers.
Acknowledgments I would like to thank K. Dimock for the constructive review of this manuscript. This work is supported by grants from the Medical Research Council of Canada, Ontario Ministry of Colleges and Universities and Oxford Virology Ltd. References 1. Weber, J. N., Clapham, P. R., Weiss, R. A., et al. (1987) Human immunodeficiency virus infection in two cohorts of homosexual men: neutralising sera and association of anti-gag antibody with prognosis. Luncet 1,119-122 2. Groopman, J. E., Benz, P. M., Ferriani, R., Mayer, K., Allan, J. D., and Weymouth, L. A. (1987) Characterization of serum neutralization response to the human immunodeficiency virus HIV. AIDS Res. Human Retroviruses 3,71-86. 3. Devash, Y., Reagan, K., Wood, D., Turner, J., Parrington, M., and Kang, C.-Y. ( 1990) Antibodies against AIDS proteins. Nature (Lmd. ) 345,58 1.
4. Lyerly, H. K., Reed, D. L., Matthews, T. J., Langlois, A. J., Ahearner, P. A., Petteway, S. R., Jr., and Weinhold, K. 3. (1987) Anti-gp 120 antibodies from HIV seropositive individuals mediate broadly reactive anti-HIV ADCC. AIDS Res. Human Retroviruses
3,409-422.
5. Rook, A. H., Lane, H. C., Folks, T., McCoy, S., Alter, H., and Fauci, A. S. (1987) Sera from HTLV-III/LAV antibody-positive individuals mediate antibody-dependent cellular cytotoxicity against HTLV-IIULAV-infected T cells. J. Zmmunol. 138, 1064-1067. 6. Ljunggren, K., Karlson, A., Fenyo, E. M., and Jondal, M. (1989) Natural and antibody-dependent cytotoxicity in different clinical stages of human immunodeficiency virus type 1 infection. Clin. Exp. Zmmunol. 75, 184-189. 7. Bender, B. S., Auger, F. A., Quinn, T. C., Redfield, R., Gold, J., and Folks, T. M. (1986) Impaired antibody-dependent cell-mediated cytotoxic activity in patients with the acquired immunodeficiency syndrome. Clin. Exp. Zmmunol. 64,166-172. 8. Wisecarver, J. T., Bechtold, T., Lipscomb, H., Davis, J., Collins, M., Purtilo, D., and Sonnabend, J. A. (1984) Comparison of antibody-dependent cell-mediated cytotoxicity and natural killer activities of peripheral blood mononuclear cells from patients at risk for acquired immunodeficiency syndrome. AIDS Res. 1, 347-352. 9. Katz, J. D., Mitsuyasu, R., Gottlib, M. S., Lebow, L. T., and Bonavida, B. (1987) Mechanism of defective NK cell activity in patients with acquired immunodeficiency syndrome (AIDS) and AIDS-related complex. J. Zmmunol. 139,55-60. 10. Gotch, F., McMichael, A., Smith, G., and Moss, B. (1987) Identification of viral molecules recognized by influenza-specific human cytotoxic T lymphocytes. J. Exp.Med.
165,408-416.
11. Mondelli, M., Vergami, G. M., Alberti, A., Vergami, D., et al. (1982) Specificity of T lymphocyte cytotoxicity to autologous hepatocytes in chronic hepatitis B virus infection: evidence that T cells are directed against HBV core antigen expressed on hepatocytes. J. Zmmunol. 129, 2773-2778. 12. Walker, B. D., Flexner, C., Paradis, T. J., and Hirsch, M. S. (1988) HIV-l reverse transcriptase is a target for cytotoxic T lymphocytes in infected individuals. Science 240,64-66. 13. Walker, B. D., Chakrabarti, S., Moss, B., Paradis, T. J. et al. (1987) HIV-specific cytotoxic T lymphocytes in seropositive individuals. Nature (L.ond) 328,345-348. 14. Imberti, L., Sottini, A., Bettinardi, A., Puoti, M., and Primi, D. (1991) Selective depletion m HIV infection of T cells that bear specific T cell receptor Va sequences. Science 254,860-862.
15. Matsuura, Y., Possee, R. D., Overton, H. A., and Bishop, D. H. L. (1987) Baculovirus expression vectors: the requirements for high level expression of proteins, including glycoproteins. J. Gen. Virol. 68,1233-1250. 16. Ratner, L., Haseltine, W., Patarca, R., Livak, K. J., et al. (1985) Complete nucleotide sequence of the AIDS virus, HTLV-III. Nature (Land.) 313,277-284. 17. Sung, W. L., Zahab, D. M., MacDonald, C. A., and Tam, C. S. (1986) Synthesis of mutant parathyroid hormone genes via site-specific recombination directed by crossover linkers. Gene 47,261-267.
Expression of AIDS-Related
Proteins
315
18. Garson, K., Percival, H. and Kang, C.-Y. (1990) The N-terminal env-derived amino acids of v-rel are required for full transforming activity. Virology 177, 106-l 15. 19. Luo, L., Li, Y., and Kang, C.-Y. (1990) Expression of gag precursor protein and secretion of virus-like gag particles of HIV-2 from recombinant baculovirusinfected insect cells. Virology 179,874-880. 20. Hu, Y.-W., and Kang, C.-Y. (1991) Enzyme activities in four different forms of human immunodeficiency virus 1 pol gene products. Proc. Natl. Acad. Sci. USA 88,4596-4600. 21. Nixon, D. F. and McMichael, A. J. (1989) HIV-gag-specific cytotoxic T lymphocytes. Res. Immunol. 140, 107-I 10. 22. Reagan, K. J., Lile, C. C., Devash, Y., Turner, J., Hu, Y.-W., and Kang, C.-Y. (1990) Use of HIV-l pal gene precursor to detect HIV-l and HIV-2. The Luncet 335,236. 23. Franchini, G., Collati, E., Arya, A., Fenyo, E., Biberfeld, G., Zagury, J. F., Kanki, P. J., Wong-Staal, F., and Gallo, R. C. (1987) Genetic analysis of a new subgroup of human and simian T-lymphotropic retroviruses HTLV-IV LAV-2 SBL-6669 and STLV-III-AGM. AIDS Res. Human Retroviruses 3,1 l-l 8. 24. Luo, L., Li, Y., Cannon, P. M., Kim, S., and Kang, C. Y. (1992) Chimeric gag-V, virus-like particles of human immunodeficiency virus induce virus-neutralizing antibodies. Proc. Natl. Acad. Sci. USA 89,10,527-10,531.
CHAPTER18
Expression
of the Hemagglutinin of Influenza Virus
Protein
Analysis of Protein Modifications
Kaxumichi
Evelyne Kretzschmar, Kuroda, and Hans-Dieter
Klenk
1. Introduction Baculovirus vectors have been developed for the expression of foreign genes in insect cells. Protein yields are often significantly higher than in bacterial, yeast, or vertebrate expression systems (12). It was therefore of interest to test the capacity of this system to direct the complex co- and posttranslational modifications involved in the biosynthesis of most membrane and secretory proteins. To probe this question, the influenza virus hemagglutinin was expressed in Spodupteru frugiperdu (Sf9) cell cultures (3,4) and in larvae of Heliothis virescens (5). The hemagglutinin of influenza A virus is a class I membrane glycoprotein whose structure has been analyzed in great detail (6). As the major viral antigen, as the initiator of infection, and as an important determinant for virus pathogenicity (7), the protein is of considerable biological interest. The biosynthesis of the hemagglutinin involves translation on membrane-bound polysomes and transport through the Golgi complex to the plasma membrane. During translation, the hemagglutinin is translocated into the lumen of the endoplasmic reticulum, freed of its amino-terminal signal peptide, and N-glycosylated with oligomannosidic side chains, The first posttranslational event is assembly of monomers to trimers at either the endoplasmic reticulum or in another pre-Golgi compartment. Another From:
Methods in Molecular Biology, Vol. 39: Baculovlrus Expression Protocols Edited
by: C. D. Richardson
Q 1995
317
Humana
Press Inc , Totowa,
NJ
318
Kretzschmar,
Kuroda, and Klenk
early posttranslational event is attachment of fatty acids to carboxy-terminal cysteine residues. In the Golgi complex, oligosaccharides are trimmed and elongated to complex side chains. Finally, activation of the fusogenic activity of the hemagglutinin by proteolytic cleavage yielding in the two disulfide-linked subunits HA1 and HA2 occurs either in the trans-Golgi compartment or in transport vesicles on the way to the plasma membrane.
2. Materials 1. The Spodopteru frugiperda cell line Sf9 is maintained at 27°C in TClOO medium with 10% fetal calf serum. This cell line is used for the propagation of wild-type and recombinant AcMNPV (3,s). 2. Eggs from Heliothis virescens are disinfected with formaldehyde vapor to avoid inapparent infection of the larvae. Hatchmg larvae areraised on a semisynthetic diet (9) in groups until they reach the 4th instar stage (see Note 1). 3. The expression vectors pAc373 and pAcYM1 were kindly provided by M. Summers, Texas A &M University, and D. H. L Bishop, VNERC Institute of Virology, Oxford, UK. 4. Hemagglutinin cDNA from fowl plague virus, strain A/FPV/RostocW34 (H7Nl), was used (3). 5. PCR-buffer: 500 mM KCl, 100 mM Tris-HCl, pH 8.4, 15 mM MgC12, and 200 pg/mL BSA (DNase-free). Store at -20°C. 6. 1X SSC: 0.15M NaCl, and 0.015M sodium citrate pH 7.0. 7. Buffer for homogenizing larvae: Phosphate-buffered saline (PBS) containing 5000 U of aprotinin/mL, 1 mM phenylmethylsulfonylfluoride (PMSF), or 10 mM iodoacetamide. 8. Triton lysis buffer: Prepare a solution containing 200 mMNaC1, 1% Triton X-100, 10 mM EDTA, 1 mM PMSF, 10 mM iodoacetamide, 5000 U aprotinin, 20 mM (iV-morpholino)ethane sulfonic acid (MES), and 30 mJ4 Tris-HCl, pH 7.2. Store at 4°C. 9. Metabolic labeling: When labeling is done with [9, 10 (n)-3H] palmitic acid, it is freed from ethanol, redissolved in 5 p,L ethanol, and added directly to the medium. [2-3H] mannose is completely dried and taken up in medium deprived of glucose and tryptose. [35a methionine is added directly to the medium (see Note 2). 10. Dithiobis[succinimidylpropionate] (DSP) for chemical crosslinking was purchased from Pierce: make fresh as required (see Note 3). 11. Sucrose gadient: continuous gradients are prepared in a gradient mix from 5 and 20% sucrose solutions containing 100 mM NaCl, l%Triton X-100, 20 mil4 MES, and 30 mM Tris-HCl, pH 7.2. 12. N-tosyl-L-phenylalanine chloromethyl ketone (TPCK) (Sigma, St.Louis, MO). 13. Endoglycosidase H (endo H), glycopeptidase F, endo-/3-N-acetylglu-
Hemagglutinin
14. 15. 16. 17.
Expression
319
cosaminidase (endo D), and a-mannosidase (Boehringer Mannheim, Mannheim, Germany). Biogel P-2, Biogel P-4, and Amberlite AG MB-3 mixed bed resin (BioRad, Munich, Germany). Li Chrosorb Diol(5 pM mesh) (Merck, Darmstadt, Germany). Aminex HPX-87H and Aminex HPX-87P HPLC columns (Bio-Rad, Munich, Germany). Galactose (Gal), mannose (Man), fucose (Fuc), N-acetylglucosamine (GlcNAc) standards (Sigma, Deisenhofen, Germany).
3.1. Full-Length
3. Methods and Truncated Hemagglutinin Used for Expression
Genes
1. The complete hemagglutinin gene (3) with terminal BumHI sites (designated A+) is cloned into the BumHI site of pAc373. The orientations of the hemagglutinin gene in these clones is determined by double cleavage with the restriction endonucleases EcoRI and EcoRV (Fig. 1). 2. To obtain a truncated hemagglutinin (A-) lacking the carboxy-terminal amino acids, including the membrane anchor and the cytoplasmic tail, the complete FPV hemagglutinin gene in pUC8 is digested with Tthl 1l/I and ligated to a translation terminator with the sequence CATAATAGTGA downstream nucleotide 1570. The mutated hemagglutinin gene is excised from the plasmid by BgZII cleavage and recloned into the BumHI site of pAc373 and pAcYM1. The orientation of the hemagglutinin genes in these vectors is determined by double cleavage with the restriction endonucleases EcoRI and EcoRV and nucleotide sequence analysis. 3. Another truncated hemagglutinin lacking the entire HAz-subunit (HA2-) was the result of spontaneous mutations of the hemagglutinin gene in a baculovirus recombinant. The mutant gene was isolated by PCR, cloned in pUC8, and sequenced. The nucleotide sequence data demonstrated that an additional A at position 1036 (amino acid 321) is present. 4. A fragment containing the HA2- coding sequence (Fig. 2) is isolated from recombinant AcMNPV DNA and amplified by PCR following the protocol described by Ausubel and coworkers (10): a. Mix the following: 10 PL 10 x PCR reaction buffer, 10 pL 8 mM dNTP mix solution, the oligonucleotides AGAGGCATTGCGAC as an upstream primer, and TTATATACAAATAGTGCACCGCAT as a downstream primer to a final concentration of 1 PM, 2 pg genomic DNA as a template, and 2.5 U Taq-Polymerase up to 100 l.tL with HzO. b. Denturation is done at 94°C for 2 min, annealing at 48OCfor 2 min, and elongation at 72’ for 2 min in 24 cycles.
320
Kretzschmar,
EcoRV
Kuroda, and Klenk
BarnHI
GTC$ATCAATAljATAGTTGCTGATATCATGG CMT ECORV AGPjlWiTTAAAA1);ATAACCATCTCGCAAAT TATA AAATAAGTATTTTACTGTTTTCGTAACAGTT h2AP TTGTAATAAAAAAACCCGAGATCCGCGGATC BarnHI CCCGGGAGCTCCCGGGAGCTTCGAGCAAAAG CAGGGGATAC~AACACTCAA---
L h+ Fig. 1. Construction of pAc-A+ hemagglutinin expression plasmid. Fowl plague virus hemagglutinin gene, polyhedrin, and other AcMNPV sequences are indicated by solid, blank, and crosshatch bars, respectively. Polyhedrin promoter is indicated by a rectangle. The nucleotide sequence of 5’ flanking region of hemagglutinin gene is also shown.
3.2. Isolation of Recombinant AciWNPV by Adsorption to Erythrocytes An alternative method (see Note 4) has been devised to enrich for tranfected Sf9 cells that produce recombinant influenza hemagglutinin
Hemagglutinin
A+
Expression I I’ Sl ml
321 Cleavage
prptlde
sit\
5""'""
paptide
Henbran
\
1 3’ anchor
A 45
I
3’
0
hydrophobic
1
basic
region
A
oligosaccharide
I
fatty
W
frameshift
linker
region
acid sequence
Fig. 2. Comparisonof wild-type andtruncatedhemagglutininsof FPV. Wildtype hemagglutinin (A+) hasthreehydrophobic regions:the signal peptide,the fusion peptide,andthe membraneanchor.Truncatedhemagglutinin lacking the membraneanchor(A-) hasbeenobtainedby insertion of a translationterminator in the HA gene in the sequencecoding for the membrane anchor. The sequencecoding for the HAt subunit has an additional adenine(A) at position 1036,which results in an incomplete hemagglutinin designatedHA,-. by adsorbing these cells to erythrocytes that have been fixed to plastic cell-culture dishes. Adsorbed Sf9 cells with hemagglutinin molecules on their surface are then released by neuraminidase treatment (3). Transfections (II) and plaque assays(12) were performed as previously described: 1. Pretreatplastic cell-culture dishesasfollows: Add a volume of 2.5 mL of a poly-t-lysine (mol wt 120,000) solution (25 pg/mL in PBS) to a 5.2-cm Petri dish, and incubate for 15 min at room temperature, Wash three times
with 2.5 mL of PBS, add a 1% chicken erythrocyte suspensionin saline (see Note 5), and incubate for 30 min at room temperature. Wash the dish
again three times with PBS.
322
Kretzschmar,
Kuroda, and Klenk
2. Add Sf9 cells (one 5.2-cm Petri dish) that have been cotransfected with pAc373 containing the full-length hemagglutinin gene and AcMNPV DNA, and subsequently incubated for 6 d at 27OC,to the 5.2-cm Petri dish containing adsorbed chicken erythrocytes. After a 30-min incubation at room temperature, wash the dish three times with PBS. 3. Add neuraminidase from Vibrio choZerue (0.2 U in 1 mL PBS) and incubate for 15 min at 37OCto release the adsorbed Sf9 cells. Add these cells together with fresh medium to a new cell-culture dish, and collect the medium after 3 d. Polyhedrin-negative plaques are isolated after two rounds of plaque purification. Cells are infected with the plaque purified virus and analyzed for hemadsorption. Isolates inducing hemadsorption are used as virus stocks.
3.3. Isolation of Polyhedra Containing Recombinant AcMNPV from Infected Sfs Cell Culture for Oral Infection of Larvae Three days after infection, polyhedra are purified from cell cultures that have been inoculated with authentic AcMNPV alone, or with both authentic and recombinant AcMNPV. For coinfection, various ratios of authentic and recombinant AcMNPV are used with multiplicities of infection (MOI) ranging from 0.1 to 2.0 PFU/cell. When the MO1 is 0.1 or 0.2 for authentic and recombinant virus, respectively, conditions are optimal for hemagglutinin expression in larvae of Heliothis virescens. The polyhedra are purified as follows (5): 1. Pellet the cells at 10,OOOgfor 20 min. Incubate the viral pellet at 23°C for 30 min in a 0.5% sodium dodecyl sulfate (SDS) solution, and then by a series of washes in water, in OSM NaCl, and again in water. 2. Finally, pellet the polyhedra through a cushion of 30% (w/w) sucrose (lOO,OOOg,30 min, 4OC), wash with water, and count in a hemacytometer.
3.4. Infection
of Insects
Two different approaches exist for the infection of larvae. First, larvae are infected by parenteral infection. Second, recombinant virus is packaged into polyhedra by double infection with wild-type AcMNPV and can then be used for infection by oral ingestion (5): 1. For parenteral infection, 4th instar larvae are injected into the hemocoel with 20 pL of cell-culture supernatant containing l@-lo6 PFU of authentic or recombinant AcMNPV and raised individually in cups on the semisynthetic diet at 23°C.
Hemagglutinin
323
Expression
2. For oral infection, 4th instar larvae are placed individually in cups with a piece of food containing lo6 polyhedra, small enough to be ingested within 24 h. Thereafter, larvae are transferred to cups with fresh food and incubated at 23°C. 3. Larvae are stripped of their guts to avoid proteolytic degradation and homogenized for 1 min in 0.5 mL of PBS containing protease inhibitors with an Ultra-Turrax blender. After brief low-speed centrifugation, the supernatant and the white upper layer of the pellet are removed from the lower black layer and used for further analyses, For histological studies, intact larvae are fixed in methanol.
3.5. Immunofluorescence
Analysis
of Cell Cultures
Sf9 cells grown on coverslips are infected with recombinant virus and incubated for 2 d at 27°C. All the following steps are performed at room
temperature. 1. Fix the cells with 1% paraformaldehyde in PBS for 15 min, and permeabilize with 0.2% Triton X-100 in PBS (lo-15 min). 2. Incubate the cells for 30 min with a 1:50 dilution of rabbit-FPV-serum for 20 min, and wash three times with PBS. 3. Adsorb fluorescein isothiocyanate-conjugated donkey antirabbit immunoglobulin to the cells for 20 min, and wash off excess antibody with PBS. 4. Mount the cells in a 10% solution of 1,4-diazobicyclo-(2,2,2)-octane and a 3.5% solution of Mowiol (Aldrich, Milwaukee, WI). 5. Examine cells under a microscope equipped with UV optics. Results are recorded photographically.
3.6. Hemadsorption, Hemagglutination and Hemolysis Assays 1. Hemadsorption assays were performed at 5-6 d after cotransfection or 2-3 d after infection with recombinant virus; monolayers of Sf9 cells on 3.2-cm Petri dishes were washed once with cold PBS, and 500 PL of a 1% suspension of chicken erythrocytes in saline were added. After 30 min at 4°C the cells are washed three times with ice-cold PBS and examined microscopically (3). 2. Hemagglutination assayswere done 2 d after the infection with recombinant or wildtype AcMNPV. Sf9 cells were scraped off plastic bottles with a rubber policeman and were centrifuged. Pelleted cells (1 x 106) were suspended in 400 PL of PBS and sonicated for 10 s at 25 W. Hemaggluti-
nation was then determinedin the homogenateby the following titration technique (3):
324
Kretzschmar,
Kuroda, and Klenk
a. Prepare 12 serial twofold dilutions of the probe in a 96-well plate. b. To 50 p,L of each dilution, add an equal volume of a 0.5% suspension of chicken erythrocytes in saline. c. After 30 min at room temperature, the hemagglutination activity of the homogenate was determined asthe reciprocal value of the dilution causing 50% hemagglutination. 3. Hemolysis assays involved mixing a volume of 100 pL from homogenized infected cells (see Section 3.6., step 2) with 100 p.L of a 1% suspension of chicken erythrocytes in saline, incubating for 15 min on ice, and then centrifuging briefly. Add 200 pL of 130 mM NaCl, and 20 rnM sodium acetate at pHs between 5.2 and 5.8 to the pellet. Incubate the mixture for 15 min at 37”C, centrifuge, and measure the optical density of the supernatant at 520 nm (3). 3.7. Immunization with Recombinant
of Chickens Hemagglutinin
For immunization of 30-wk-old chickens, insect cells are used, 3 d after infection with recombinant AcMNPV. Chickens are injected twice intramuscularly (im) with 10’ cells at an interval of 3 wk. Two weeks later, the birds are challenged by im infection with lo4 PFU of FPV. In series 2, chickens are inoculated once im with a cell suspension (10’ cells) emulsified in Freund’s complete adjuvant and 3 wk later challenged with the same FPV concentration as before. Blood samples are obtained before challenge, and the sera are inactivated at 45°C for 30 min. Hemagglutination inhibition and neutralization tests are performed according to standard procedures (3). 3.8. Metabolic
Labeling and Immunoprecipitation of Hemagglutinin Cells (1.5 x 106) from a 3.2-cm Petri dish are infected with recombi-
nant virus (10 PFU/cell) and labeled 2 d after infection with [35s] methionine, [2-3H] mannose, or [9, 10 (n)-3H] palmitic acid: 1. For labeling with [35s] methionine (100 pCi/mL), preincubate cells for 2 h with TC-100 medium lacking methionine and tryptose. When labeling is done in the presence of tunicamycin, Sf9 cells are treated with 1, 2, or 5 p.g/mL tunicamycin (stock solution 1 mg/mL tunicamycin in 0. 1N NaOH) 2 h before labeling. Then label cells with [35s] methionine in the presence of 1,2, or 5 pg/mL tunicamycin. For labeling with [9, 10 (n)-3H] palmitic acid (500 pCi/3.2-cm Petri dish), complete TClOO medium is used. For
Hemagglutinin
Expression
325
labeling with [2-3H] mannose (100 pCi/3.2-cm Petri dish), cells are precincubated for 2 h with medium lacking glucose and tryptose, and labeling is done in the same medium. 2. After the pulse or the chase, which is performed in complete TClOO medium, scrape the radiolabeled cells off the dishes, wash twice with cold PBS, and solubilize pelleted cells obtained from a 3.2 cm Petri dish in 0.5 mL Triton lysis buffer. For analysis of secreted protein, remove medium and mix with l/4 volume Triton lysis buffer. 3. Centrifuge lysates to remove insoluble debris and incubate under shaking with hemagglutinin-specific antisera (1: 100) and protein-A Sepharose (50%, 20 pL/lOO pL cell lysate) overnight at 4OCunder constant shaking. 4. Wash the immunocomplexes three times for 10 min at 4°C under constant shaking, and analyze by SDS-PAGE on 12% acrylamide gels followed by fluorography.
3.9. Chemical
Crosslinking
This procedure is used to analyze oligomers on polyacrylamide
gels:
1. Wash pulse-chase labeled cell lysates twice with cold PBS, and resuspend in 0.4 mL Triton-lysis buffer. Centrifuge the cleared lysates obtained from one 3.2-cm Petri dish. Treat 100 pL of the lysate with 2 pL 40 mM DSP (dithiobis[succinimidylpropionate]) in DMSO for 15 min at 15OC.Stop the reaction by addition of 2 pL 1M NH4HC03. 2. The samples are immunoprecipitated with hemagglutinin-specific antisera as indicated above and analyzed after separation on 6% polyacrylamide gels under nonreducing conditions.
3.10. Sedimentation in Sucrose
of Hemagglutinin Gradients
This procedure is used to analyze the quaternary hemagglutinin.
structure of the
1. Sf9 cells infected with recombinant AcMNPV are incubated at 27OC for 2 d. Cells are then labeled with [35S]methionine (100 pCi/mL), and chase is done in complete TC-100 medium. Cells are washed twice with cold PBS and lysed with 0.4 mL Triton-lysis buffer. 2. The lysate obtained from one 3.2-cm Petri dish is centrifuged to remove debris and loaded on a linear sucrose gradient (5-20%; w/v). 3. Sedimentation is performed in a Beckman SW55 rotor at 45,000 rpm for 13 h at 4°C. Gradients are collected in 11 0.4~mL fractions from the bottom, immunoprecipitated with a rabbit antiFPV serum as described in the protocol above, and analyzed by SDS-PAGE.
Kretzschmar,
326
3.11. Isolation of GZycopeptide, and Chomatographic 3.11.1. Isolation
Kuroda,
and Klenk
Oligosaccharides, Analysis
of Glycopeptides
39 cells expressing wild-type hemagglutinin are labeled with [2-3H] mannose overnight and immunoprecipitated using rabbit-WV serum. To obtain glycopeptides, the immunocomplex is digested with trypsin (14): 1. Digest samples (5-10 x lo3 cpm) of immunoprecipitated hemagglutinin with N-tosyl-L phenylalanine chloromethyl ketone (TPCK)-treated trypsin (stock solution 10 mg/rnL in 1 mM HCl) at a final concentration protem/ typsin of 40: 1 (w/w) for 4 h at 37°C in 5 mL of 0.2 ammonium bicarbonate buffer, pH 8.2. 2. Desalt glycopeptldes obtained by gel filtration. 3.11.2. Isolation of Oligosaccharides
Free carbohydrates from hemagglutinin glycopeptides are isolated by digestion with endo H, and residual material is separated by Biogel P-4.
The endo H-resistant glycans are then liberated by glycopeptidase F (14): 1. For treatment with endo H, dissolve tryptic glycopeptides in 1 mL of a 0.05M sodium phosphate-citrate buffer, pH 5.0, with O.lM NaCl. Then add 0.4-0.8 nkat recombinant endo H (endo-P-ZV-acetylglucosaminidaseH [EC 3.2.1.961) from Streptomyces plicatus produced in Escherichia co& and incubate the mixture at 37°C for 15 h (see Note 6). 2. Separate oligosaccharides released from residual peptides and glycopeptides by Biogel P-4 chomatography. 3. For reduction with sodium borohydride, take up the oligosaccharide samples in 200 p,L water and add l-2 mg NaBH4 (stock solution 10 mg/ mL in HzO). Store the mixture overnight in the dark at room temperature. On the next day, acidify with acetic acid, and remove boric acid by repeated evaporation with methanol. Desalt by gel filtration. 4. For digestion with glycopeptidase F, dissolve endo H-resistant glycopeptides in 50 pL of 20 mA4sodium phosphate buffer, pH 7.2, containing 10 mit4 EDTA, and add 0.1 nkat glycopeptidase F (EC 3.5.1.52) from Flavobacterium meningosepticum. Incubate samples at 37OC for 18 h. Enzyme addition and incubation are repeated once. 3.11.3. Chromatographic Procedures for Glycopeptide and Oligosaccharide Analysis 1, Glycopeptides and oligosaccharides are desalted by a Biogel P-2 column (-400 mesh; 0.6 x 50 cm; Bio-Rad, Munich, Germany) with distilled water as eluant. Fractions of 1.2 mL are collected at a flow rate of 2.4 mL/h.
Hemagglutinin
2. 3.
4.
5.
Expression
327
After addition of scintillation cocktail, aliquots of all fractions are monitored for radioactivity with a Model 4550 (Packard, Meriden, CT) liquid scintillation counter. Neutral monosaccharides are desalted by passage through a column (0.5 x 7 cm) of Amberlite AG MB-3 mixed-bed resin using distilled water as eluant. Unbound radioactive material is collected. For chromatographic separation of oligosaccharides by gel filtration, a Biogel P-4 column (-400 mesh; 06 x 200 cm; Bio-Rad, Munich, Germany) is used with 0.02% aqueous sodium azide as eluant. Chromatography is performed at hydrostatic pressure, and fractions of 0.4 mL are collected and monitored for radioactivity. HPLC-separation of oligosaccharide aldrtols sensitive to endo H IS performed using a column (0.4 x 15 cm) of LiChrosorb Diol(5 pm; Merck, Darmstadt, Germany) and acetonitrile/wate (75:25, by volume) as eluant. Fractions of 0.3 mL are collected at a flow rate of 0.5 mL/min and monitored for radioactivity. The column is calibrated with oligomannosidic standard alditols ManS-9GlcNAcOH and isomaltooligosaccharide alditols (with l-8 Glc units). Radiolabeled monosaccharides released by exoglycosidases from endo Hresistant oligosaccharides are identified by HPLC using Ammex HPX-87H or HPX-87P columns (0.78 x 30 cm; Bio-Rad, Munich, Germany) and O.OlN sulfuric acid or distilled water as eluant at 50 or 80°C, respectively. Fractions of 0.1 mL are collected at a flow rate of 0.5 mL/min. The columns are calibrated with Gal, Man, Fuc, and GlcNAc (HPX-87H) or with Man and Gal (HPX-87P). Monosaccharide standards are detected by continuous flow monitoring of absorbance at 193 nm. Radiolabeled sugars are combined with unlabeled monosaccharide standards and ascertained by liquid scintillation counting of the fractions collected (13). 3.11.4. Digestion with Exoglycosidases
1. For degradation of oligosaccharide alditols with a-mannosidase (EC 3.2.1.24) from jack beans (Sigma, Deisenhofen, Germany), take up the oligosaccharide sample in 100 PL of 50 mM sodium citrate buffer of pH 4.5 (with 0.2 mM ZnCl& and incubate with 25 nkat (see Note 6) of the enzyme for 24 h at 37OC.Enzyme addition and inoculation are repeated once. 2. For treatment with a-fucosidase from beef kidney (EC 3.2.1.5 1; Boehringer Mannheim, Germany), dissolve samples in 50 pL 0.05M sodium citrate buffer, pH 4.5, and incubate with 1 nkat of the enzyme for 24 h at 37OC. 3. For digestion with a-1,Zmannosidase from Aspergillus oryzae, dissolve samples in 50 pL of O.lM sodium acetate buffer, pH 5.3, and incubate with 8.3 pkat of enzyme for 24 h at 37OC.
Kretzschmar,
Kuroda, and Klenk
4. For digestion with P-mannosidase (EC 3.2.1.25) from Polyporus suljkreus, dissolve samples in 100 pL 0.05M glycine/HCl buffer, pH 2.6. After addition of the enzyme (1.7 nkat), incubate samples at 37OC for 24 h. Reaction products are analyzed by Biogel P-4 chomatography (see Section 3.1,1.3., step 3). 3.11.5. Digestion of Immunoprecipitated Hemagglutinin with Endoglucosaminidases H and D and Glycopeptidase F 1. Suspend the immunocomplex in 50 pL phosphate buffer containing SDS (0.1%) and mercaptoethanol(O.5%). 2. After boiling for 5 min, add Triton X-100 at a final concentration of O-5%, and clarify samples by centrifugation. 3. Incubate overnight at 37°C with 17 pkat endo H, 17 pkat endo D (endo+N-acetylglucosaminidase D) (EC 3.2.1.96) from Diplococcus pneumoniae or 0.8 nkat glycopeptidase F (EC 3.5.1.52) from Fluvobacterium meningosepticum or delete enzyme in negative controls. 4. Samples are analyzed by SDS-PAGE on 10% gels and by fluorography. Concerning
3.12. Analysis of Data Protein Modifications of Influenza
HA
By expressing the hemagglutinin of influenza virus in the baculovirus system, it could be shown for the first time that insect cells are able to synthesize and process complex membrane proteins of vertebrate origin. Posttranslational proteolytic cleavage, which is of high importance for the biological activity of the hemagglutinin, was also observed with the FPV hemagglutinin (3,15). It did not take place, however, with another influenza hemagglutinin containing a different cleavage site (4). Sf9 cells were also able to acylate (8) and to glycosylate the FPV hemagglutinin (Fig. 3). Trimerization of the hemagglutinin occurs in Sf9 cells at kinetics two to three times slower than in vertebrate cells (8). As a consequence, the processing steps following trimerization, such as proteolytic cleavage and surface transport, are also retarded and less efficient (Fig. 4). Deletion mutants of the hemagglutinin lacking the membrane anchor (A-) or the entire HA, subunit (HA,) have also been expressed in insect cells (Fig. 2). These mutants do not trimerize, and although both lack the membrane anchor, only HA2- is secreted into the culture medium (16). When analyzed for its biological activities, hemagglutinin expressed in Sf9 cells showed hemagglutination, hemadsorption, and hemolysis, and it reacted with hemagglutinin-specific antibodies.
Hemagglutinin
Expression
329
Fig. 3. Acylation, glycosylation, and proteolytic cleavage of PPV hemagglutinin in St9 cells. Two days after inoculation with recombinant AcMNPV, St9 cells were labeled for 12 h with [2-3H] mannose (Man), [35s]methionine (Met), or [3H] palmitic acid (FA). A+ was immunoprecipitated using rabbit anti-FPV serum and analyzed by polyacrylamide gel electrophoresis followed by fluorography. The hemagglutinin precursor HA and the cleavage products HAi and HA, can be seen. It also protected chickens from infection with FPV (3). Hemagglutinin yields obtained in Sf9 cells obtained by Western blot analysis (Fig. 5) are shown in Table 1. Although the initial studies had already shown that hemagglutinin from insect cells contained N-glycosidic oligosaccharides, it appeared that they were less completely processed than in vertebrate cells (15). A detailed analysis of the carbohydrate structures supported this concept. In this study, radioactively labeled oligosaccharides were liberated by treatment with endoglucosaminidase H and glycopeptidase F. Stepwise degradation with exoglycosidases and chromatographic separation revealed, besides oligomannosidic structures containing predominantly five to nine mannose residues, the truncated oligosaccharides MansGlcNAc, and Mans(Fuc)GlcNAr+ It could also be shown that most of the oligosaccharides on HA1 were truncated, whereas HA2 contained one trun-
330
Kretzschmar,
Kuroda,
and Klenk
Fig. 4. Kinetics of hemagglutinin cleavage in Sf9 and MDCK cells. MDCK cells infected with FPV (A) and Sf9 cells infected with recombinant AcMNPV (B) were incubated at 37°C for 6 h or at 27°C for 2 d. respectively. Cultures were labeled with [35s] methionine for 20 min at 27°C and chased for 0,30,60, 120, and 240 min at the same temperature. After immunoprecipitation with an FPV antiserum, samples were subjected to SDS-polyacrylamide gel electrophoresis followed by fluorography. The extent of cleavage was determined by densitometric analysis of the fluorograms on a Bio-Rad model 620 videodensitometer (C). The amounts of cleaved hemagglutinin observed at different intervals in MDCK cells (A) and in Sf9 (0) cells, as well as the halftimes of cleavage in MDCK cells (A), and in Sf9 cells (O), are indicated (8).
cated and one oligomannosidic side chain. Comparison with the glycosylation pattern obtained in vertebrate cells allowed an assignment of these oligosaccharides to the individual glycosylation sites (Fig. 6). These results showed that 39 cells have the capacity to trim N-glycans to trimannosyl cores that can be further processed by addition of fucose. The complex oligosaccharides found on hemagglutinin obtained from
Hemagglutinin
Expression
Fig. 5. Quantitation of hemagglutinin by Western blot analysis. Serial dilutions from 2-l to 2” containing 13.125-1.64 pg total protein of infected cell lysates of A+ (a), A- (c), and HA,- (d) or cell supernatants of A- (c) and HA,(e) and serial dilutions from 2-l to 2-4 containing 0.525-0.065 g of a purified hemagglutinin-probe as a standard (GP HA) were loaded on a 12% SDS-polyacrylamide gel, subjected to electrophoresis, transferred onto nitrocellulose, and subjected to Western blotting using rabbit anti-FPV serum. The blot was photographed and densitometrically analyzed on-a Bio-Rad model 620 video densitometer. Tbe standard probe was obtained by octylglucoside extraction from purified FPV particles. vertebrate cells are therefore replaced in hemagglutinin from insect cells by truncated oligosaccharide cores (13). It has been reported that recombinant glycoproteins derived from Sf9 cells contain complex oligosaccharides with neuraminic acid, but despite extensive search, we have
332 Hemagglutinin
Kretzschmar,
Table 1 (HA) Production in the Cellular and Secreted Fraction of Sf9 Cells VectorC
Cells
pAc373 pAcYM 1 pAcYM1
Medium
Kuroda, and Klenk
pAc373
Insert A+ AHA,AA-Mu27b AAHA,-
%HA of total protein 3.4 4.5 1.1 4.7 1.6 -
“ug HA obtained from a 3.2-cm Petri dish. *A- hemagglutinin with deleted 5’ noncodmg region (21 HA-speak
cLgHAa 22 35 10
29 16 0.8 3 33 and 18 nucleotides from
cloning events). The vectors pAc373 and pAcYM1 were used for the expression of A+, A-, and HAz-, Fortyeight hours postmfection HA yields were examined by densitometric analysis of Western blots,
and total protein yields by the Lowry method. Each value wasdetermined by at least two different preparations.
obtained no evidence for the presenceof significant amounts of such oligosaccharides in Sf9 cell-grown hemagglutinin (18). Hemagglutinin was also expressed in live insects. For this purpose, larvae of Heliothis virescens were infected either by parenteral injection or by feeding of the baculovirus vector. For peroral uptake, recombinant virus occluded in polyhedra was used that was obtained by coinfection of Sf9 cells with wild-type and recombinant AcMNPV. Hemagglutinin processing (Fig. 5) and hemagglutinin yields in larvae were similar to those in Sf9 cells. Hemagglutinin showed its typical biological activities (Fig. 7) and induced immune protection in chickens. These studies showed that functional hemagglutinin cannot only be produced in insect cell cultures, but can also be obtained from living caterpillars (5). 4. Notes 1. Heliothis virescens larvae can be obtained from the Forest Pest Management Institute, Saulte Ste. Marie, Ontario, Canada, the Entomological Society of America, P.O. Box 177, Hyattsville, MD, or the AAB Entomology Group HI Wellesbourne, Wellesbourne, Warwick, UK CV35 9E7. Their diet can be purchased from Bio-Serv (Vineland, NJ).
Fig. 6. Comparison of the oligosaccharide patterns of hemagglutinin derived from vertebrate and insect cells. Left panel: Glycosylation sites and structures of the individual carbohydrate side chains as determined for the hemagglutiniu of A/FPV/Rostock/34 (H7Nl) grown in chick embryo cells (17). Right panel: Oligosaccharide structures at individual glycosylation sites of FPV hemagglutinin derived from Sf9 cells: W, N-Acetylglucosamine; 0, mamrose; 0, galactose; a, fucose. Sequence positions of asparagine residues serving as carbohydrate attachment sites are numbered. Shaded areas indicate antigenic sites A, B, C, and D (14).
334
Kretzschmar,
Kuroda, and Klenk
Fig. 7. Exposure of hemagglutinin at the surface of fat body cells of Heliothis virescens larvae. Four days after parenteral infection with recombinant AcMNPV, larvae were dissected, washed once with PBS, and assayed for hemadsorption. The scanning electron micrograph (magnification, x400) shows numerous erythrocytes attached to the surfaces of fat body cells. Inset: Thin section showing the site of close attachment between a chicken erythrocyte (E) and a fat cell (F). Arrowheads indicate the adjoining lipid bilayers. Bar, 0.2 pm (5). 2. Isotopes can be purchased from Amersham, and they include: [9, 10(n)-3H] palmitic acid (40-60 Wmmol) in ethanol, [2-3H]-mannose (10-20 Ci/ mmole) in ethanol:water (l:l), and [35S]-methionine (1000 Ci/mmol), labeling grade. 3. Chemical crosslinkers such as DSP (dithiobis[succinimidylpropionate]) can be purchased from Pierce. The cross-linking agent is dissolved in DMSO just prior to the experiment, but it can be stored at -20°C for up to 6 mo.
Hemagglutinin
Expression
335
4. Recombinant baculovirus can be isolated by any of the procedures used in Chapters 6-10 of this book. We routinely use the dot-blot hybridization method described by Pen et al. (14). 5. Chicken erythrocytes may be purchased from Advance Biotechnology (Columbia, MD), Crane Laboratories (Syracuse, NY), or Pel-Freez Biologicals (Rogers, AR) as a suspension in Alsevier’s solution. They may be stored at 4°C for up to 1 mo. The red blood cells (0.5 mL) are washed twice with 50 mL PBS though low-speed centrifugation (1000 rpm in a bench-top centrifuge) prior to experiments. 6. Glycosidase activity is measured in nkat or pkat units where kat is an enzyme unit of catalysis. Enzymes can be purchased from Boehringer Mannheim.
References 1. Luckow, V. A. and Summers, M. D. (1988) Trends in the development of baculovirus expression vectors. Bio/Technology 6,47-55. 2. Kang, C. Y. (1988) Baculovirus vectors for expression of foreign genes. Adv. Virus Res. 35, 177-192.
3. Kuroda, K., Hauser, C., Rott, R., Klenk, H.-D., and Doerfler, W. (1986) Expression of the influenza virus haemagglutinin in insect cells by a baculovirus vector. EMBO J. 5,1359-1365.
4. Possee, R. D. (1986) Cell surface expression of influenza virus hemagglutinin in insect cells using a baculovirus vector. Virus Res. 5,43-59. 5. Kuroda, K., Groner, A., Frese, K., Drenckhahn, D., Hauser, C., Rott, R., Doerfler, W., and Klenk, H.-D. (1989) Synthesis of biologically active influenza virus hemagglutinin in insect larvae. J. Viral. 63,1677-1685. 6 Wiley, D. C. and Skehel, J. J. (1987) The structure and function of the hemagglutinin membrane glycoprotein of influenza virus. Ann. Rev. Biochem. 56,365-394. 7. Klenk, H.-D. and Rott, R. (1988) The molecular biology of influenza virus pathogenicity. Adv. Virus Res. 34,247-28 1. 8. Kuroda, K., Veit, M., and Klenk, H.-D. (1991) Retarded processing of influenza virus hemagglutinin in insect cells. Virofogy 180, 159-165. 9. Ivaldi-Sender, C. (1974) Technics pour une Blevage permanente de la tordeuse orientale graphonita molesta (Lep., Torticidae) sur milieu artificielle. Ann. Zool. Ecol. Anim. 6,337-343.
10. Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D. D., Seidman, J. G., Smith, J. A., and Struhl, K. (1989) Current Protocols in Molecular Biology. John Wiley, New York. 11. Felgner, P. L., Gadek, T. R., Holm, M., Roman, R., Chan, H. W., Wenz, M., Northop, J. P., Ringold, G. M., and Danielsen, M. (1987) Lipofection: A highly efficient, lipid-mediated DNA-transfection procedure. Proc. Natl. Acad. Sci. USA 84,7413-7417. 12. Wood, H. A. (1977) An agar overlay assay method for Autographa califomica
nuclear-polyhedrosisVKUS. J. Znvertebrate Pathal. 29,304-307
Kretzschmar,
Kuroda, and Klenk
13. Kuroda, K., Geyer, H., Geyer, R., Doeffler, W., and Klenk, H.-D. (1990) The oligosaccharides of influenza virus hemagglutinin expressed in insect cells by a baculovirus vector. Virology 174,418-429. 14. Pen, J., Welling, G. W., and Welling-Wester, S. (1989) An efficient procedure for the isolation of recombinant baculovirus. Nucl. Acids Res. 17,45 1. 15. Kuroda, K., Hauser, C., Rott, R., Doerfler, W., and Klenk, H.-D. (1989) Processing of the hemagglutinin of influenza virus expressed in insect cells by a baculovirus vector, in Invertebrate Cell System Applications (Mitsuhashi, J., ed.), CRC Press, Boca Raton, FL, pp. 221-227. 16. Kretzschmar, E., Veit, M., Brunschon, S., Kuroda, K, and Klenk, H.-D. (1992) Secretion of fowl plague virus hemagglutinin from insect cells requires elimination of both hydrophobic domains. J. Gen. Virol. 73,839-848. 17. Keil, W., Geyer, R., Dabrowski, J., Dabrowski, U., Niemann, H., Stirm, S., and Klenk, H.-D. (1985) Carbohydrates of influenza virus: Structural elucidation of the individual glycans of the PPV hemagglutinin by two-dimensional ‘H NMR and methylation analysis. EMBO J. 4,27 1l-2720. 18. Kretzschmar, E., Geyer, R., and Klenk, A.-D. (1994) Baculovirus infection does not alter N-glycosylation in Spodoptera frugiperda cells. Biol. Chem. Hoppe-Seyler 375,323-327
CHAPTER19
Purification of Recombinant Protein Derived fkom the Baculovirus Expression System Using Glutathione Affinity Agarose Eric J. Sorscher
and Mqja A. Sommerfizlt
1. Introduction The baculovirus expression system represents an elegant means whereby high-level expression of foreign genes can be achieved using eukaryotic insect cells (1,2). Viruses from the Baculoviridae subfamily of insect viruses, usually Autographa californica nuclear polyhedrosis virus (AcNPV), are most frequently used because they contain very active late gene promoters for proteins that are expressed late in the virus life cycle and are not essential for virus replication. Replacing a late gene, most commonly polyhedrin or ~10, with a foreign gene of interest results in recombinant virions capable of expressing abundant quantities of the foreign gene of interest by virtue of the highly active late gene promoter. Recombinant virions can only be generated using the less efficient process of homologous recombination, because the genome of baculovirus is so large (120-130 kb) that conventional cloning techniques become impractical. Homologous recombination events occur (at a frequency of 0.1-0.2%), following cotransfection of baculoviral DNA with recombinant transfer vector DNA carrying the gene of interest under the control of a late gene promoter and flanked by viral sequences. Recombinant virions are usually identified using a plaque assay, comparing recombinant plaque phenotype to that of wild-type. This has, however, proven to From: Methods in Molecular Biology, Vol. 39: Waculovirus ExpressIon Protocols Edited by. C. D. Richardson Q 1995 Humana Press Inc., Totowa, NJ
337
338
Sorscher
and Sommerfelt
be time-consuming and tedious, necessitating the development of novel approaches to simplify and accelerate the identification process (3-9). The recombinant protein, once expressed in sufficient quantities, usually needs to be purified from recombinant baculoviral proteins and cell debris. This process can be challenging, particularly if the recombinant protein of interest is cell-associated. Purification of proteins derived from prokaryotic expression systems has recently been facilitated by the development of techniques that allow single-step affinity purification (1046). The recombinant protein is synthesized as a fusion protein with polypeptides that can be purified under mild conditions using a suitable affinity matrix. A similar approach has recently been adapted for the purification of proteins expressed using the baculovirus expression system (I 7). Recombinant baculovirus is generated, which synthesizes the foreign protein of interest as a fusion protein with glutathione-S-transferase (GST). Purification from crude cell lysates is achieved using glutathione-affinity agarose. After affinity isolation, the recombinant protein is retrieved following cleavage at an engineered proteolytic cleavage site. This method is rapid, involving only a single purification procedure, and avoids the use of chemicals or detergents, which may interfere with the biological integrity of the foreign protein. Two transfer vectors have been designed for the generation of fusion proteins with GST, one containing a factor Xa cleavage site (G-BAC-l), and the other a thrombin cleavage site (G-BAC-2). These vectors have been successfully used for the isolation of glutathione-S-transferase as well as for the isolation of polypeptides coding for the predicted soluble domains of the cystic fibrosis transmembrane conductance regulator (CFTR), the full-length CPTR, and the full-length human low-density lipoprotein (LDL) receptor. In each case, the proteins could be isolated in a single step and under mild conditions (17). The baculoviral transfer vectors (G-BAC-1 and G-BAC-2) suitable for the generation of recombinant baculovirus coding for glutathione-Stransferase fusion proteins were engineered as described in Fig. 1. Two oligonucleotide primers (5’ GTC GAC TAG TCA TGT CCC CTA TAC TAG GTT ATT GG 3’ and 5’ GCA TGC TAG CAT GAA TTC CCG GGG ATC CCA CGA 3’) were designed to generate an Spe-1 site at the 5’ end and an Nhe-1 site at the 3’ end of a product corresponding to the GST-factor Xa cleavage site of a cDNA region present in the expression
Purification
of Recombinant
339
Protein
GST .
v PH promotor
-Translectlon with baculowrsl -Homologous recombination \-Plaque pwfy
genamie
seculovlrus coding for GST expressmn
‘PlOteaSe Cleavage SllSS (Xa or Thrombm)
Fig. 1. Generationof baculoviral transfervectors(G-BAC) directing the synthesis of glutathione-S-transferase fusion proteins. vector pGEX-3X (Pharmacia). The region encompassed by these primers was amplified by PCR and cloned into the Nhe-1 cloning site of blue-white selection baculoviral transfer vector, pJVETLZ (gift from Chris Richardson) to generate the transfer vector G-BAC- 1. The vector retained the Nhe- 1 cloning site suitable for the insertion of foreign cDNAs, in frame and downstream of the GST cDNA sequence. A second transfer vector (G-BAC-2) was similarly designed containing cDNA-encoding GST followed by a human thrombin cleavage site. The cDNA template used to generate G-BAC-2 was obtained from the pGEX-2T expression vector. These two vectors were used for the production of GST alone, as well as in conjunction with polypeptides coding for the predicted soluble domains of the CFTR, the full-length CFTR, and the full-length human LDL receptor (17). The affinity yield could be compared with that obtained using the pGEX prokaryotic expression system. 2. Materials 2.1. Cotransfection of Viral and Transfer Vector DNAs 1. Purified AcNPV DNA in TB (10 mM Tris-HCl, pH 8.0, and 1 mM EDTA). 2. Transfer vector DNA (G-BAC-1 or G-BAC-2) containing the recombinant protein of interest, purified on a cesium chloride gradient and resuspendedin TB.
Sorscher and Sommerfelt
340
3. Lipofection (GIBCO/BRL). 4. Double-distilled water. 5. Pipets, tips, T25 tissue-culture flasks (CoStar). 6. Spodopterafrugiperda (Sf9) insect cells. 7. Serum-free Grace’s insect medium (GIBCO/BRL, supplemented with lactalbumin and yeastolate), as well asGrace’s insect medium containing 10% fetal bovine serum (FBS). 8. Sterile 1.5 mL Eppendorf tubes.
2.2, Identification of Recombinant AcNPV Using FZuorescence-Activated Cell Sorting 1. Cells from the cotransfection. 2. Grace’s insect medium containing 10% FBS.
3. ImaGene alkyl derivatized fluorescein di-P-o-galactopyranoside substrate (Molecular Probes, Eugene, OR) is used at a final concentration of 33 l.U4 and is stored at -2OOC. Hydrolysis by B-galactosidase produces a fluorescent product. 4. Fluorescence-activated cell sorter (FACS):In our laboratory, we use a FACStar Plus Instrument (Becton Dickinson, San Jose, CA). The parameters for FACS analysis are 535 emission wavelength and excitation of 488 nm. DF 530/30 filters (Omega Optics, Battleboro, VT) are used. A 70 pm nozzle is used and sheath fluid passed though a 0.45 p,rn filter prior to the addition of cells. 5. Six-well tissue-culture trays. 6. 1-mL sterile pipet tips. 7. 1-mL sterile vials.
2.3. Rapid Purification of Recombinant AcNPV and PZaque Assay 1. Medium containing recombinant virus or cells from the highly fluorescent cell population isolated by FACS. 2. Sf9 cells. 3. Six-well tissue-culture trays. 4. 5-bromo-4-chloro-3-indolyl-P-o-galactopyranoside 100X X-Gal solution (20 mg/mL in DMSO) stored concealed from light at 4°C. 5. Agar overlay (autoclaved 3% Sea Plaque agarose [FMC Bioproducts]) in water mixed in equal volume with Grace’s insect medium containing 10% FBS and supplemented with X-Gal at a final concentration of 0.2 mg/mL. 6. l-, 5-, lo-, and 25mL pipets. 7. Grace’s insect medium containing 10% FBS.
Purification
of Recombinant
Protein
341
2.4. Afinity Enrichment of GST Fusion Proteins from S/9 Cells 1, 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.
Purified recombinant baculovirus stock expressing P-galactosidase. Sf9 insect cells. Grace’s insect medium containing 10% FBS. Low-speed centrifuge. Homogenizer. Homogenization buffer: 150 mMNaC1,16 mM N%HzP04, 4 mit4 Na.HP04, pH 7.3, and 1% Triton X-100. Sonicator (Cole Palmer 4710 ultrasonic homogenizer). PBS: 150 mM NaCl, 16 mM NaH,POd, and 4 mMNa2HP04, pH 7.3. Reduced glutathione agarose (GSH-Ag), 12 pmol reduced glutathione/rnL hydrated agarose (Sigma) in PBS stored at 4OC. Just pnor to use, the agarose is incubated in PBS with 1% Triton X100. Human thrombin (Sigma). Cutting buffer: 50 mM Tris, pH 7.4, 150 mM NaCl, 2.5 mM CaCl,, 20 U human thrombin/mg fusion protein. Protein quantitation system: In our laboratory, we utilized a bicinchoninic acid-based quantitation system (Mini BCA, Pierce) according to the manufacturer’s instructions.
3. Methods 3.1. Cotransfection of Viral and Transfer DNAs Using Lip0 fee tin Spodopteru frugiperdu Sf9 insect cells are transfected with both wildtype AcNPV DNA (purified according to the method of Summers and Smith [18]) and recombinant transfer vector DNA (G-BAC-1 or G-BAC2) DNA. A mixed-virus population will result, the majority being wildtype AcNPV-producing polyhedra with a proportion being recombinant, containing the protein of interest fused with glutathione-S-transferase in place of polyhedrin. 1. Two T25 tissue-culture flasks are seeded with Sf9 cells at a density of 2 x lo6 cells in a volume of 5 mL using Grace’s insect medium containing 10% FBS. 2. The cells are allowed to settle overnight at 27OC. The cells are washed three times in serum-free Grace’s insect medium and overlaid with 1 mL serum-free Grace’s insect medium prior to transfection. 3. Wild-type AcNPV DNA (100 ng) is combined with 1 p.g recombinant transfer vector DNA in a volume of 12 PL in double-distilled sterile water,
342
Sorscher and Sommerfelt
4. Lipofectin (8 l..tL)is mixed with 4 PL sterile double distilled water. 5. The DNA and lipofectin mixtures are combined and allowed to stand at room temperature for 15 min. 6. The DNA/Lipofectin mixture is added to the cells dropwise, and the cells are incubated at 27OC. 7. Five to 24 h later, 1 mL Grace’s insect medium (with 10% serum) is applied, and the cells are allowed to incubate at 27°C for 1 wk.
3.2. Identification and Enrichment of Recombinant AcNPV Using Fluorescence-Activated Cell Sorting (FACS) The cells are harvested and recombinant AcNPV identified according to a protocol developed recently in our laboratory (4). 1. Medium is removed from the transfected cells, and 1 mL fresh serumfree Grace’s insect medium with 3 p,L ImaGene alkyl-derivatized fluorescein di-P-galactopyranoside substrate is added (final concentration 33 lU4). 2. Following incubation at room temperature for 15 min, the cells are scraped from the sides of the flask and subjected to FACS. Hydrolysis of substrate by P-galactosidase yields a fluorescent product, which indicates that the transfected cell contains recombinant virus. 3. The cells with the greatest fluorescence intensity (0.5% total cell population) are isolated and subjected to further purification (see Section 3.3.).
3.3. Rapid Purification of Recombinant Virus and Plaque Assay The cells with the greatest fluorescence intensity can be subjected to either of two procedures that result in the rapid purification of recombinant virus (4). Procedure A is: 1. Forty-five cells from the highly fluorescent cell population are plated in mixed culture with 2 x lo6 uninfected Sf9 cells. Seventy-two hours later, the supernatant is harvested and clarified at 2000 rpm for 10 min to remove debris. 2. Clarified supernatant is then plated in 0.4 mL volumes at dilutions of loo, 10-l, 10s2,and 10m3in Grace’sinsect medium on six-well trays containing 2 x lo6 cells/well for 1 h at 27°C. 3. The inoculum is removed, and the cells overlaid with 4 mL molten agar overlay. When the overlay has solidified, 2 mL Grace’s insect medium containing 10% FBS are added to retain moisture and the plates are incubated at 27°C.
Purification
of Recombinant
Protein
343
4. Seventy-two hours later, blue plaques corresponding to recombinant AcNPV should be clearly visible. Procedure B is: 1. Of the highly fluorescent cells, 20, 200, or 2000 cells, can be cocultured with 2 x lo6 uninfected insect cells in six-well trays. 2. The cells are allowed to settle for 4 h after which time the medium is removed and replaced with 4 mL molten agar overlay. 3. When the agar overlay has solidified, 2 mL Grace’s insect medium supplemented with 10% FBS are added to retain moisture and the plates incubated for 72 h, after which time blue plaques should be clearly visible corresponding to the presence of recombinant AcNPV.
For both procedures A and B, isolated plaques can be picked for further rounds of plaque purification by stabbing the plaque with a sterile pipet tip, and transferring the agarose plug to 1 mL serum-containing or serum-free medium in a sterile vial. Virus is eluted from the agaroseplug during overnight incubation at 4°C. Serial 1:lO dilutions of this virus eluate can be retitrated, and blue plaques picked once more and purified by titration on insect Sf9 cells (18). of Recombinant Virus Recombinant AcNPV can be amplified by inoculating insect cells at a low multiplicity of infection (MOI). The virus stocks can be stored in aliquots at 4°C. 3.4. Amplification
1. Sf9 cells are seeded at a density of 10’ cells in a T75 tissue-culture flask. 2. The medium is removed and the cells inoculated with recombinant AcNPV at an MO1 of 0.01 in a volume of 2 mL. 3. After incubation for 1 h at 27OC, 6 mL Grace’s insect medium supplemented with 10% FBS are added, and the cells incubated for 7 d. 4. The medium is harvested and cell debris removed by centrifugation at 2000 rpm for 10 min and the supernatant is aliquotted and stored at 4°C. The medium is also titrated to determine the titer of recombinant virus present. 3.5. Analysis of Recombinant Protein Using SDS-Polyacrylamide Gel Electrophoresis
Recombinant virus must be checked for the expression of the desired recombinant protein. 1. Cells are infected with recombinant virus at an MO1 of 5 or 10 for 1 h at 27OC.Grace’s insect medium is added, and the cells incubated at 27OC.
Sorscher
344
and Sommerfelt
9796#3321 -
A
B
C
Fig. 2. Expression and purification of recombinant glutathione-S-transferase. Recombinant baculovirus expressing glutathione-S-transferase (28 kDa) was used to infect’Sf9 cells (Section 3.6.) and a soluble lysate prepared (Lane A). Elution of affinity-isolated glutathione-S-transferase was possible using either Laemmli buffer (Lane B) or 5 n&I glutathione (Lane C). 2. Three days later, the cells are scraped and lysed in Laemmli lysis buffer. 3. Ten-microliter aliquots are run on a 12% polyacrylamide gel and stained using Coomassie brilliant blue. Wild-type AcNPV is used as a control and should reveal the overexpression of polyhedrin (~31). 3.6. Aflnity Purification Proteins from
of GST Fusion
S/9 Cells
Recombinant proteins that have been expressed as GST fusion proteins can be easily purified using glutathione agarose. Fig. 2 shows the affinity purification of glutathione-S-transferase from a total cell lysate using either Laemmli sample buffer or competitive elution from the agarose using 5 rnM reduced glutathione. Figure 3A shows the affinity purification of a fusion protein of glutathione-S-transferase with a region of the CFTR protein that incorporates the nucleotide binding domain (NBD-1) from a cell lysate (lanes A-C). This CFTR polypeptide can be separated from glutathione-S-transferase by cleavage using thrombin
Purification
of Recombinant
345
Protein
- 97 -66 -44
-33
- 21
Fig. 3A. Enrichment of a 65kDa fusion peptide (upper arrow) containing glutathione-9transferase with a portion of the CFIR encompassing the nucleotide binding domain -1 (NBD-1) (19). Lanes A-C show the affinity isolation of the fusion peptide. Following cleavage of this fusion peptide with thombin, a 37-kDa polypeptide corresponding to the CFTR polypeptide (middle arrow) and a 2%kDa polypeptide corresponding to glutathione-S-transferase (lower arrow) are clearly evident. (Fig. 3A lanes D-F) to give a 37-kDa CFTR polypeptide and a 2%kDa GST polypeptide. Purification of a 44-kDa CFTR polypeptide using glutathione-S-transferase is shown in Fig. 3B. The protein obtained is rec-
ognized by a polyclonal antiserum raised against the same domain produced in bacteria (Fig. 3B, Lane C) and is not recognized by a normal rabbit preimmune serum as negative control (Fig. 3B, lane D). 1. Sf9 cells are inoculated with purified recombinant AcNPV at a high MO1 of 5-10 and harvested 3 d postinfection (Section 3.5.). 2. The cells are scraped, centrifuged at 1OOOgfor 5 min at 4OC, and then homogenized (sonicated) in 1 mL homogenization buffer/4 x 10’ cells. 3. The cell suspensionsare sonicated at an output control of 5 (125 W) using 5 x l-s pulses followed by 2 continuous pulses of 10 s each on ice with a 3-mm microtip probe.
346
Sorscher
and Sommerfel
t
B
A
B
C
D
Fig. 3B. Enrichment of a 44-kDa CFIR polypeptide. Lane A shows an insect cell lysate expressing a fusion protein of glutathione-S-transferase with the CFT’R polypeptide. Lane B shows affinity isolation using glutathione-S-transferase. Lane C shows Western blotting of the peptide using rabbit polyclonal antibody raised independently against a prokaryotic gene product from the pGEX system(16). Lane D is a negative control using preimmune rabbit serum. 4. The insoluble fraction is removed by centrifugation at 10,OOOgfor 5 min at 4OC. 5. The soluble fraction of Sf9 cell homogenate is added to GSH-Ag (50 mL agaroselml Sf9 cell lysate) and rotated at 4OCovernight. 6. The GSHAg is then washed 5 times with 15 bed vol of PBS, and recovered after each wash by centrifugation at 750g for 10 s. GST fusion proteins are eluted by incubation of washed GSH-Ag with 5 bed vol of 50 mM Tris-HCl, pH 8.0, containing 5 rnM reduced glutathione overnight on a rotator at 4°C. 7. Protein can also be eluted using Laemmli sample buffer, loaded directly on a polyacrylamide gel, and stained with Coomassie brilliant blue. 8. The fusion protein can be released by cleavage using thrombin in cutting buffer with 20 U of human thombin/mg fusion protein at 25OC for 3 h (Fig. 3A). 9. Quantitation of protein is performed using a bicinchoninic acidbased protocol.
Purification
of Recombinant
Protein
347
Affinity yields of recombinant fusion proteins obtained by this technique were 25 pg/106 cells (GST), 5 pg/106 cells (65 kDa GST-CFTR, Fig. 3A), and 1.5 pg/106 cells (44 kDa GST-CFTR, Fig. 3B). In addition, one-step affinity yield of a GST fusion with the full-length human LDL receptor was 3 p,g/106cells, and yield of a fusion of GST with the fulllength CFTR (approx 190 kDa) was cl ng/106 cells (not shown). In at least one case, the affinity yield of a GST fusion protein was substantially improved by expression in insect cells as opposed to bacteria. When the same fusion protein isolated as shown in Fig. 3B was produced in E. coli using PGEX-2T (Pharrnacia), the product was completely insoluble and could not be affinity-purified to a measurable extent (19). 4. Notes 1. It is important to use cesium chloride-purified DNA in lipotransfection experiments, sinceminiprep DNA will give lower transfectionefficiency. The transfection procedureusing lipofection must be serum-free for the first 5-24 h, after which time, medium
containing
serum is added.
2. It is important to pick isolated plaques, since contaminating wild-type plaques will causeproblems in future experiments. It is also helpful to use l-n& pipet tips wherethetips havebeencut away to widen the aperture, so that au entire plaque can be incorporated in a single agar plug. 3. Wild-type AcNPV-infected and uninfected Sf9 cell lysates should also be used
as controls when runningprotein gels in orderto confirm the expressionof foreign protein and P-galactosidase, and absence of polyhedrin in infected insect cells. It is also important to harvest infected cells for protein gel analysis over 24-h intervals to show that the overexpression of foreign protein corresponds to the time at which late genes are produced (48-72 h postinfection). 4. Always check blue plaques by light microscopy to ensure that they have a polyhedrin-negative phenotype.
References 1, Kang, C. Y. (1988) Baculovirus vectors for expression of foreign genes. Adv. Virus. Rex 35, 177-192. 2. Fraser, M. J. (1992) The baculovirus-infected cell as a eukatyotic gene expression system. Curr. Top. Microbial. Immunol. 158,131-172. 3. Vialard, J., Lalumiere, M., Vernet, T., Briedis, D., Alkbatib, G., Henning, D., Levin, D., and Richardson, C. (1990) Synthesis of the membrane fusion and hemagglutinin proteins of the measles virus using a novel baculovirus vector containing the l%galactosidase gene. J. Virol. 64,37-50. 4. Peng, S., Sommerfelt, M. A., Berta, G., Berry, A. K., Kirk, K. L., Hunter, E., and Sorscher, E. J. (1993) Rapid purification of recombinant baculovirus using fluorescence-activated cell sorting. Biotechniques 14,274-277.
348
Sorscher and Sommerfelt
5. Patel, G., Nasmyth, K., and Jones, N. (1991) A new method for the isolation of recombinant baculovirus. Nucl. Acids Res. 20,97-104. 6. Domingo, D. L. and Trowbridge, L. S. (1988) Characterization of the human transferrin receptor produced in a baculovirus expression system. J. Biol. Chem. 263, 13,386-13,392. 7. Kitts, P. A. and Possee, R. D. (1993) A method for producing recombinant baculovirus expression vectors at high frequency. Biofechniques 14,81817. 8. Kitts, P. A., Ayres, N. D., and Possee, R. D. (1990) Linearization of baculovirus DNA enhances the recovery of recombinant virus expression vectors. Nucl. Acids Res. 18, 5667-5672.
9. Vlak, J. M., Schouten, A., Usmany, M., Belsham, S. J., Klinge-Roode, E. C., Maule, A. J., van Lent, J. W. M., and Zuidema, D. (1990) Expression of cauliflower mosaic virus gene I using a baculovirus vector based upon the ~10 gene and a novel selection method. Virology 179,3 12-320. 10. Marston, F. A. 0. (1986) The purification of eukaryotic polypeptides synthesized in Escherichia coli. Biochem. J. 240, 1-12. 11. Nilsson, B., Abrahamsen, L., and Mathias, V. (1985) Immobilization and purification of enzymes with staphylococcal protein A fusion vectors. EMBO J. 4, 1075-1080. 12. Guan, C., Li, P., Riggs, P. D., and Inouye, H. (1987) Vectors that facilitate the expression and purification of foreign peptides in Escherichia coli. Gene 67,21-30. 13. Hochuli, E., Bannwarth, W., Dobeli, H., Gentz, R., and Stuber, D. (1988) Genetic approach to facilitate purification of recombinant proteins with a novel metal chelate absorbent. Bio/Technology 6, 1321-1325. 14. Smith, D. B. and Johnson, K. S. (1988) Single step purification of polypeptides expressed in Escherichia coli as fusions with glutathione-S-transferase. Gene 67,3 l-40. 15. Johnson, K. S., Harrison, G. B. L, Lightowlers, M. W., O’Hoy, K. L., Cougle, W. G., Dempster, R. P., Lawrence, S. B., Vinton, J. G., Heltb, D. D., and Richard, W. D. (1989) Vaccination against ovine cysticercosis using defined recombinant antigen. Nature 338,585-587. 16. Hartman, J., Daram, P., Frrzzell, R. A, Rado, T., Benos, D. J., and Sorscher, E. J. (1992) Affinity purification of insoluble recombinant fusion proteins containing glutatbione-S-transferase. Biotechnol. Bioeng. 39,828-832. 17. Peng, S., Sommerfelt, M. A., Logan, J., Huang, Z., Jilling, T., Kirk, K., Hunter, E., and Sorscher, E (1993) One-step affinity isolation of recombinant protein using the baculovirus/insect cell expression system. Protein Expression and Purification
4,95-100.
18. Summers, M. D. and Smith, G. E. (1987) A manual of methods for baculovirus vectors and insect cell culture procedures. Bulletin No. 1555, Texas Agricultural Experiment Station, Texas, A & M University, College Station, TX. 19. Hartman J., Huang, Z., Rado, T. A., Peng, S., Jillmg, T., Muccro, D. D., and Sorscher, E. (1992) Recombinant synthesis, purification and nucleotide binding characteristics of the first nucleotide domain of the cystic fibrosis gene product. J. Biol. Chem. 267,6455-6458.
CHAPTER20 Purification of the Extracellular Domain of the Epidermal Growth Factor Receptor Produced by Recombinant BaculovirusInfected Insect Cells in a 10-L Reactor Maria
l! Debanne, Maria C. Pacheco-OZiver, and Maureen D. O’Connor-McCourt
1. Introduction The binding of ligand to the extracellular domain of the epidermal growth factor (EGF) receptor is the event that initiates the activation of the tyrosine kinase activity present on the cytoplasmic domain. This activation leads to the phosphorylation of the receptor and other cellular substrates, and results ultimately in stimulation of mitogenesis (1,2). The EGF receptor system is implicated in both the uncontrolled growth of certain carcinomas and the controlled growth that occurs during wound healing (3). In order to develop EGF receptor antagonists that may be used to treat cancer and agonists that may promote wound healing, an understanding of the structure-function relationships of the ligandreceptor complex is essential. By studying the soluble extracellular domain of the receptor, the use of detergents is avoided, and biochemical and biophysical analyses are facilitated. We have expressed the extracellular domain of the EGF receptor (EGFR-ED) by infecting insect cells with a recombinant baculovirus that carries a DNA insert encoding for the entire extracellular domain, including the N-terminal signal sequence. Two stop codons were introduced immediately following the last extracellular domain codon (ser 62 1) to prevent further translation. Milligram From:
Methods In Molecular Biology, Vol. 39. Baculovirus Expression Protocols Edlted by: C. D RIchardson
0 1995
349
Humana
Press Inc., Totowe,
NJ
350
Debanne, Pacheco-Oliver,
and O’Connor-McCourt
quantities of active EGFR-ED are secreted by the infected insect cells in a serum-free medium in a 10 L helical ribbon impeller bioreactor (Chapter 12 in this book). In this chapter we report a mild, nonaffinity method for purification of the EGFR-ED which was developed with the intention of avoiding exposure of the EGFR-ED to the relatively harsh conditions normally used for elution from affinity columns. The method is based on one anion-exchange column being used twice at two different pHs. The first pH was selected to be below the isoelectric point of the EGFR-ED, so that the EGFR-ED is not retained on the column, whereas the majority
of the contaminants are retained. The flowthrough is reloaded on the same column after being adjusted to a pH above the isoelectric point of the EGFR-ED. The EGFR-ED is then eluted by an NaCl gradient at -80% purity. Final purification is achieved by ammonium sulfate precipitation. The basic strategies described in this chapter for media concentration and diafiltration can be used for any protein that is secreted into large volumes of media. The anion-exchange purification method should be adaptable to any secreted baculovirus-expressed protein with an isoelectric point above 7.0, since the majority of insect cell-derived contaminants will be removed by retention on the anion-exchange column at pH 6.0.
2. Materials 2.1. Clarification of Culture Medium, Product Concentration, and Diafiltration 1, 2. 3. 4. 5.
Phenylmethyl-sulfonyl fluoride (PMSP) stock solution: 100 mM in ethanol. O.lOum hollow-fiber membrane H5MPOl-43 (Amicon). SlOY30 spiral membrane cartridge, 30,000 mol-wt cutoff (Amicon). Sly30 spiral membrane cartridge, 30,000 mol-wt cutoff (Amicon). DClOL high capacity filtration system with reservoir and equipped with a positive displacement gear pump (Amicon), interchangeable use of Amicon spiral and hollow fiber membrane (SlO and H5), addition of a peristaltic pump XX80 000 00 (Millipore) with a 701821 pump head and 6411-18 silicon tubing to facilitate reservoir filling and continuous diafiltration, and stainless-steel heat exchanger (WHE) installed to the inside of the reservoir to permit temperature control of the process. 6. RM20 Lauda MgW water bath (Brinkman). 7. Temperature probe of Accumet Model 805 pH meter (Fisher). 8. CH2PRS ultrafiltration system (Amicon) with a small polypropylene Tee connector inserted in between the pump and the spiral entrance to facilitate sampling and emptying of the system.
Production
and Purification
of EGF Receptor
351
9. Conductivity meter. 10. Diafiltration buffer: 20-25 L of 0.02M Tris-HCl, pH 7.4,0.1 n-&f PMSF. 11. Membrane cleaning solutions: Hollow-fiber membrane: 5 L of dHzO with cont. nitric acid added to approx pH 2.1% Ultrasil-10 membrane cleaning solution (Henkel). S 10 membrane: 5 L of Ultrasil-53 membrane cleaning solution (Henkel). S 1 membrane: 1 L of O.lN NaOH. 12. Membrane storing solutions: Hollow-fiber membrane: 3% (v/v) glycerol/0.2% (w/v) NaNs. Spiral membranes: O.O2MTris-HCI, pH 8.2,0.5MNaCl, 10 mMEDTA, and 0.1 rniV PMSF.
2.2. Receptor 1. 2. 3. 4.
5. 6. 7. 8.
Purification
Filter units, 500 mL, 0.2- pm membrane. Hiload 26/10 Q-Sepharose high-performance column (Pharmacia). FPLC System (Pharmacia). Purification buffers (degased and kept at 4°C pH adjusted at 22°C): Buffer A. 1.OL of 0.02M Tris-HCI pH 8.2. Buffer B. 0.5 L of 0.02M Tris-HCl pH 8.2,0.3M NaCl. Buffer C. 0.5 L of 0.02M Tris-HCl pH 6.2 150 mL of 2M NaCl. DEAE-Sephacel (Pharmacia). Glass wool. Ammonium sulfate (molecular biology reagent) (Sigma).
2.3. Receptor
Analysis
1. 96-well microtiter plates. 2. Protein adsorption buffer: 10 mM borate buffer, pH 8.2, containing 150 mM NaCl. 3. Blocking buffer: 10 mM borate buffer pH 8.2, containing 150 n-u?4NaCl, 1% (w/v) bovine serum albumin (BSA), and 0.1% (w/v) NaN,. 4. Assay buffer: 10 mMHEPES-HCl pH 7.4, containing 150 mZt4NaCl, 0.1% (w/v) NaN,, 0.1% (w/v) BSA, and 0.02% (w/v) Tween 20. 5. 1 mg/mL goat-antimouse IgG2b. 6. 0.1 mg/mL mouse mAb anti-EGF receptor (RPN.513) (Amersham). 7. 0.1 mg/mL mouse mAb anti-EGF receptor (Ab-1) (Oncogene Science). 8. 1 mg/mL horse radish peroxidase (HRP) conjugated goat antimouse IgG2a. 9. Peroxidase substrate: OPD (0-phenylendiamine s 2 HCl) tablets and diluent for OPD (Abbot Laboratories). 10. 4N H2S04. 11. Microtiter plate reader,
352
Debanne,
Pacheco-Oliver,
and O%onnor-McCourt
12. EGF receptor standard: Full-length EGF receptor from A-43 1 cell vesicles, which was solubilized with 0.1% (w/v) Triton X-100 and calibrated by Scatchard analysis. A-431 cells are supplied by ATCC (Rockville, MD). 13. Phastsystem separation, control and development units (Pharmacia). 14. PhastGel (gradient lO-15% and homogeneous 7.5%) (Pharmacia). 15. PhastGel SDS and Native buffer strips (Pharmacia). 16. Mouse EGF, receptor-grade (Collaborative Research): Dialyzed against ammonium bicarbonate and dried in Eppendorf tubes in quantities of 20 p.g/tube by speed-vacuum drying. 17. Microdialyzer System 100 (Pierce). 18. Protein standard: 1 mg/mL (w/v) bovine serum albumin. 19. 0.025% (w/v) copper-EDTA, and 2% (w/v) sodium carbonate in O.lN sodium hydroxide for protein assays. 20. 10% (w/v) SDS. 21, Folin-Ciocalteu’s phenol reagent.
3, Methods Immediately after harvesting, the medium is cooled down and 1 mL/L of PMSF stock solution is added. The entire purification procedure takes 3 d. If it is inconvenient to proceed immediately after harvesting, glycerol is added to the medium to a final concentration of 5% (w/v), and it is stored at -2OOC. The clarification, first concentration, and diafiltration steps are done at 8°C. The final concentration and purification steps are done at 4OC. At each step, volumes are recorded and aliquots taken from receptor enriched and remnant samples for studies on recovery and yield. 3.1. Clarification of Culture Medium, Product Concentration, and Diafiltration 3.1.1. Clarification of Culture Medium The samples that will be concentrated and diafiltrated on spiral membranes (Section 3.1.2.) must be clear of cells and cell debris. These particles could cause irreversible clogging of the spiral membranes, and the inlet pressure would exceed the manufacturer’s recornrnendation. A clarification step using a hollow-fiber membrane is required to make sure that only clear solutions are used with the spiral membranes. 1. Rinse the hollow fiber membrane with 4 x 5 L dH,O to remove storing solution. 2. Using a tared beaker, determine the dHzO flow rate by measuring by weight the filtrate volume for 30 s with the pump speed at position 40 (see Notes
Production
3. 4. 5. 6.
7. 8.
9. 10. 11. 12.
and Purification
of EGF Receptor
353
1 and 2). Using the hollow fiber’s surface area of 0.45 m2, convert this measurement to a flow rate measured in L/h/m2. This measurement is the filtrate flux prior to processing. Note the pressure (inlet and outlet) and the temperature reading measured by the immersed probe. Fill the DClOL reservoir with the medium using the peristaltic pump. Install a clean reservoir to collect the filtrate. The product will be in the filtrate stream. Slowly start the gear pump on the DClOL system, and set the pump to 40 (see Note 3). Monitor the filtrate flow rate as described in step two for 1 min. Note the pump speed, the pressure (inlet and outlet), and the running temperature. The water bath should be set to 4°C to ensure that the product temperature does not go above 8OC. Continue monitoring the flow rate for every 2 L of filtrate collected. The drop in flux will give the user an indication of the fouling rate of the membrane with time. Continue filtration until approx 2 L of medium remains in the DClOL reservoir. Slow the pump rate, and continue filtermg until the beginning of a vortex. The product will slowly start foaming because of the introduction of air. Stop the pump at this point. Empty the reservoir by opening the collection valve. This retentate contains concentrated cellular debris and some receptor. Store the filtrate at 4°C. Centrifuge the retentate at 8OOOgfor 20 min. Discard the pellet, and add the supernatant to the filtrate obtained during clarification. Rinse the DClOL with dH,O. Determine the flow rate with clean water and the pump speed at 40. Note the pressure and the temperature. This low rate will give an indication of the extent of membrane fouling. If time permits, the membrane can be cleaned by recirculating for 10 mm 5 L of a solution of dH20 and concentrated nitric acid (pH 2). For recirculation, the filtrate line is returned to the reservoir. After emptying the reservoir, the system is rinsed several times with dH20, and the filtrate checked with pH paper. The rinsing is stopped when the pH is neutral. Recirculate for 30 min 5 L of a 1% Henkel-10 solution. Empty the reservoir, and rinse until the filtrate is neutral. Determine the filtrate flow rate at pump speed 40. Note pressure and temperature. This measurement is an indication of whether the membrane has been restored after cleaning. Restoration of the flux to a value of 75% or more of the initial rate is acceptable for this polysulfone membrane. Store the membrane in the appropriate storing solution. If time is lacking, the membrane can be removed after the first system rinse to determine fouling and stored in dH20 untrl cleaning is possible,
354
1.
2. 3. 4. 5. 6. 7. 8.
9. 10. 11.
12.
13.
Debanne, Pacheco-Oliver,
and O’Connor-McCourt
3.1.2. Product Concentration and Diafiltration Rinse the SlOY30 membrane with dHzO to remove the storage solution. Determine the clean water flux as mentioned previously (Section 3.1 .l ., step 2) with the pump at position 40 (the surface area is 0.94 m2). Note the pressure (inlet and outlet) as well as the temperature. Condition the membrane with 5 L of the diafiltration buffer by recirculation for 30 min. Measure the conductivity of the clarified media and the buffer. Fill the DClOL reservoir with the clarified media using the peristaltic pump. Install a clean reservoir for filtrate collection. Slowly start the gear pump, and set the pump speed to 40. Monitor the filtrate flow rate as mentioned previously. When approx 3 L of retentate remains in the DClOL reservoir, add the diafiltration buffer by using the peristaltic pump. Adjust the peristaltic pump’s rate so as to maintain a constant volume in the DClOL reservoir (3 L). This enables the user to perform continuous diafiltration, thereby minimizing the amount of buffer required to condition the concentrated receptor sample prior to ion-exchange chromatography. Five to seven diafiltration volumes may be required to condition the sample. After every diafiltration volume (3 L), the filtrate flow rate should be measured and a sample taken for conductivity measurement. When the conductivity of the filtrate is the same as the buffer, the diafiltration can be stopped. Continue concentration until 2 L of retentate remain in the DClOL reservoir. Do not allow a vortex to occur, since foaming causesprotein denaturation. Collect the retentate by opening the collection valve. If the retentate is turbid owing to the presence of insoluble matter, the solution should be centrifuged and the pellet discarded. The supernatant is further concentrated using the SlY30 membrane and the CH2PRS system. If the retentate appears clear, no centrifugation is required. Rinse the SlOY30 membrane and the DC 1OLsystem. Determine the extent of membrane fouling by measuring the flow rate. If time allows, recirculate for 30 min 5 L of 0.5% Henkel-53 solution. This cleaning solution contains proteolytic enzymes. Rinse well with dH20, and determine membrane restoration by measuring the flow rate with dH20. Store membrane in PMSF containing storage solution. Install the Sly30 membrane on the CH2PRS unit. Rinse the membrane with dH,O, and determine the clean water flux (the surface area is 0.094 m2). Set the pump speed setting to 5 and the pressure (inlet and outlet) to 10 psi with the system at room temperature. Condition the membrane with 1 L of the diafiltration buffer for 30 mm.
Production
and Purification
of EGF Receptor
355
14. Empty the reservoir and place the system at 4OC.Fill the reservoir with the retentate obtained in step 11, and slowly start the pump. Set the pressure to 10 psi using the back-pressure valve. Collect the filtrate using a clean reservoir, Monitor the filtrate flux initially, after the collection of 100 mL of filtrate and then after each 500 mL. Continue concentration until 250 mL of retentate remain in the reservoir (see Note 4). Make sure a vortex does not occur. Stop the pump and open the back-pressure valve (0 psi). Open the clamp of the Tee connector, and collect the final concentrated sample in a clean tared bottle by slowly starting the pump (setting 2). Close the clamp, and add 100 mL of diafiltration buffer to the reservoir. Recirculate the buffer for a few seconds. Empty the contents of the reservoir into a clean tared bottle. This membrane rinsing step allows maximization of the receptor recovery. 15. Place the CH2PRS system at room temperature. Rinse the system with dHzO, and determine membrane fouling by measuring the flow rate. Recirculate a solution of O.lN NaOH with the pump speed at position 5 with no applied pressure for 10 min. Empty the reservoir, and rinse with dHaO. Remove the reservoir lid, and allow the water to empty via the tube in the lid. This allows the membrane to be rinsed. Check the pH of the water until the pH is neutral. Fill the reservoir with more water, and apply 10 psi pressure. Check the pH of the filtrate. Stop recirculation when the pH of the filtrate is neutral. Empty the reservoir, and store the membrane in the appropriate storing solution, 3.2. Receptor Purification Based on the fact that the isoelectric point of the EGFR-ED produced by the insect cells is 7.1, we rationalized that a simple purification procedure could be based on one anion-exchange column being used twice at two different pHs. We established that, as expected, at pH 6.2, the EGFRED is not retained on a Q-Sepharose colurrm, whereas most of the contaminant proteins are; when the flowthrough is reloaded on the same column at pH 8.2, the EGFR-ED is retained and can be eluted as a relatively pure protein by a continuous NaCl gradient (see Fig. 1). A TrisHCl buffer is used for both pHs to avoid intermediate dialysis with the consequent loss of material and time, i.e., when the pH of the EGFR-ED containing flowthrough sample from the first step is simply increased to 8.2, the sample is ready for the second step, 1. Adjust the pH of the 250-300 mL concentrated and diafiltrated media sample to 6.2 (at 22OC),and centrifuge at 15,000g for 20 min. Centrifugation is necessary since the sample usually contains a considerable amount
356
Debanne, Pacheco-Oliver,
and O’Connor-McCourt
Fig. 1. Elution of the EGFR-ED from a Q-Sepharose column at pH 8.2 by an NaCl gradient. Shown is the second ion-exchange chromatography step. The EGFR-ED containing flowthrough from the first ion-exchange step was reloaded on the Q-Sepharose column at pH 8.2, and proteins were eluted with a shallow NaCl gradient. An analysis of the fractions by silver-stained native PAGE (inset) shows that the EGFR-ED elutes between 56 and 86 min as an -80% pure protein. of insoluble material, which increases when the pH is lowered. Collect supernatant. 2. Test protein concentration as described in Section 3.3.1. 3. If the total amount of protein in the sample is lower than 1.3 g (column capacity is 1.6 g), go to step 4. If it exceeds 1.3 g, do a preadsorption step with DEAE-Sephacel as follows: add 60 mL of settled DEAE matrix, which has been washed with 5 vol of buffer C on a Biichner funnel. Stir for 5 min, and filter through glass-wool. Collect the filtered sample, and discard the glass-wool with retained matrix. 4. Filter the sample through a 0.2~pm membrane, and inject it onto the Q-Sepharose column, which has been equilibrated with buffer C and connected to an FPLC system (see Note 5). Adjust the flow rate to 2.5 mL/ min. Collect the flowthrough in bulk.
Production
and Purification
of EGF Receptor
5. Wash the column with 2 bed vol of buffer C collecting the first 5&70 mL of wash in the same container as the flowthrough sample. 6. Adjust the pH of the combined flowthrough sample to 8.2. 7. Condition the column for the next step by washing it with 1 bed vol 2M NaCl followed by 5 bed vol of buffer A at a flow rate of 5 mL/min. 8. Inject the sample as before (step 4) onto the reconditioned column. 9. Wash with 2 vol of buffer A. 10. Elute the EGFR-ED using a shallow 400~mL gradient of 0-0.12M NaCl (O-40% buffer B). Adjust the elution flow rate to 2.5 mL/min, and collect fractions every 2 min (fraction size of 5 mL). 11. Recondition the column as described in step 7, substituting buffer A for buffer C. The column is ready for the next separation. 12. Analyze the peak that elutes between 50 and 90 min (125 and 225 mL) after the beginning of the gradient using silver-stained lo-15% SDS PhastGels (Fig. 1). 13. Select and pool the receptor containing tubes. 14. Bring the pH of the pooled sample to 7 and precipitate the protein by slowly adding an amount of ammonium sulfate that will result in 60% saturation (see Note 6). 15. Stir the sample slowly for 1 h, and collect the precipitated protein by sedimentation at 50,OOOgfor 5 min. 16. Solubilize the pellet in 1 mL of dH20. 17. Centrifuge the solubilized sample at 100,OOOgfor 10 min. 18. Analyze the supernatant for protein concentration, receptor activity, and purity (Section 3.3.). 19. Store the purified EGFR-ED sample in a 60% saturated ammonium sulfate solution at 4°C.
3.3. Receptor Analysis 3.3.1. Protein Concentration Any protein determination method can be used except for samples from the early steps of purification that contain pluronic F-68 (an ingredi-
ent of the insect cell-culture media). Pluronic F-68 interferes with several protein-determination techniques. Either a modified Lowry assay for detergent-containing samples (4) can be used for these samples, as described below, or an assaythat uses bicinchoninic acid as a Cu+l detection reagent and is marketed as the BCA protein assay (Pierce) (5). 1. To 200 pL of diluted sample, add 1 mL of alkaline copper solution. 2. Incubate at room temperature for 15 min. 3. Add 1 mL of 10% SDS and mix well.
358
Debanne,
Pacheco-Oliver,
and O’Connor-McCourt
4. Add 100 PL of phenol reagent diluted 1:2. 5. Incubate at room temperature for 30 min. 6. Read absorbance at 600 nm (or at 750 if protein concentration is low) against a sample buffer blank. 7. Use 1 mg/mL BSA as the standard. 3.3.2. Active Receptor Concentration The EGFR-ED is quantitated using an immunoenzymetric assay (6). The technique uses two mAbs of different IgG subclass directed against two different conformation-dependant epitopes on the EGFR-ED. Accordingly, the assay quantifies only active, folded receptor. 1. Coat wells of a 96-well microtiter plate with 50 pL of goat antimouse
2. 3. 4. 5. 6.
7. 8.
IgG2b diluted 1:400 (2.5 pg/rnL) in protein adsorption buffer for 2 h at room temperature. Wash with same buffer, and add to each well 100 ltL of blocking buffer. Incubate overnight at 4°C. (Plates that are well sealed can be kept at 4°C for long periods of time at this stage.) Wash wells three times with 100 ~.LLassay buffer. Add 50 pL/well of mouse mAb anti-EGF receptor (RPN-513) diluted 1: 100 (1 pg/mL) in assaybuffer. Incubate at 37°C for 1 h. Wash as in step 3. Add to different wells 40 l.tL of 0.25, 0.5, 0.75, and 1 nM dilutions of receptor standard and a range of dilutions of the sample to be tested. Add 10 l.tL of mouse mAb anti-EGF receptor (Ab-1) diluted 1:20 (5 pL/mL) m assay buffer, and incubate at 37OCfor 2 h. Wash as in step 3. Add 50 PL of goat antimouse IgG2b-HRP conjugate diluted 1:400 (2.5 l,tg/ mL) in assay buffer, and incubate at 37’C for 1 h.
9. Wash as in step 3. 10. Develop color by adding 50 @ of substrate solution prepared as indicated by the manufacturers.
11. Quench color development by adding 50 p,L of 4N H,S04. 12. Read at 492 nm in a microtiter plate reader, 3.3.3. Receptor Activity Since the isoelectric point of EGF is lower than that of EGFR-ED, the EGFR-ED-EGF complex runs faster on native-PAGE than uncomplexed EGFR-ED. This band shift method can be used to analyze the percentage of EGFR-ED that is able to complex with EGF, i.e., the percentage of active receptor (see Fig. 2).
Production
and Purification
of EGF Receptor
B
359
A
Fig. 2. Analysis of the activity of the EGFR-ED by band shift on native PAGE. Shown in the inset is the migration of the EGFR-ED alone (A) or the EGFR-ED/ EGF complex on native-PAGE (B). By scanning the gel on a densitometer, we are able to determine that the EGFR-ED is -97% pure and that -85% of the EGFR-ED is active, i.e., it is shifted on native PAGE in the presence of ligand. 1. Dialyze 10 pL of the final purified EGFR-ED sample against PBS in a microdialyzer overnight. 2. Add 10 pL of the dialyzed sample to an Eppendorf tube containing dried EGF. Incubate 5-10 min at room temperature. 3. Run the sample on a native PhastGel (homogeneous 7.5%). 4. Stain with Coomassie blue. 5. Read gel on a densitometer. 6. Calculate percent of receptor band shifted. This percent equals percent of active receptor in the sample. 3.3.4. Receptor Purity
The final purity of the sample is established by densitometric analysis of SDS PhastGels (gradient lo-15%) stained with Coomassie blue. 1. 2. 3. 4.
Run lo-15% gradient SDS PhastGel. Stain with Coomassie blue. Read gel on a densitometer. Calculate percent of the receptor band.
360
Debanne, Pacheco-Oliver,
and O’Connor-McCourt
4. Notes 1. Prior to using a new membrane, nominal flux rate must be determined. This is the flow rate of the membrane with clean water at a given pump rate, pressure, and temperature. This nominal rate represents 100% flux for the membrane. Once the membrane has been used, the water flux at the same running conditions after processing is an indication of the membrane fouling. After cleaning, the measurement allows the user to determine if cleaning was efficient. For the hollow fiber, polysulfone membrane, restoration above 75% is acceptable. With the spiral cartridges, which are made of regenerated cellulose, 90% restoration is achievable. If the membranes are heavily fouled and restoration is not adequate, for the polysulfone membrane, repeat the cleaning procedure and back-flush the fibers. For the SlOY30 cartridge, if the membrane is heavily fouled, increase the Ultrasil-53 to 1%. 2. Range of flux during the process for the membranes utilized: 0. l-pm hollow-fiber membrane: 20-30 L/h/m* SlOY30 membrane: 17-27 L/h/m* S lY30 membrane: 20-40 L/h/m* 3. During the clarification step, it is very important not to apply pressure. Pressure will increase as the fibers become clogged. Note the pressure carefully!! Do not let the inlet pressure go above 5 psi. This is the maximum pressure allowed for the membrane. High pressure will break the fibers, making the membrane nonreusable. With proper care, the hollow-fiber membranes can be used for several years without changes in separation and flux rates. 4. If during the Sly30 concentration step the solution becomes turbid, stop the concentration, empty the reservoir, and centrifuge the retentate. Discard the pellet, and continue concentration of the supernatant. 5. It is converuent to inject the sample directly into the column without using a loop. For that purpose, use the calibration block set on TIME mode instead of volume mode, and readjust the program of the method-file block also to TIME mode. Reconfigure Valve 7 as follows: Port 1 connected to the column, Port 2 with the incoming tube from the sample through the peristaltic pump-l, Ports 3 and 6 closed, Ports 4 and 5, to waste as in the original configuration, and Port 7 also as originally set with the tube transportmg buffer from pumps A and B. 6. Milligrams of ammomum sulfate to be added to 1 mL of solution (7): mg = [533 (S,-S,) ] / [ 100 - .3(S,) ] S1= initial % of saturation and S, = final % of saturation.
Production
and Purification
of EGF Receptor
361
References 1. Carpenter. G. and Cohen, S. (1990) Epidermal growth factor. J. Biol. Chem, 265, 7709-77 12. 2. Ullrich, A. and Schlessinger, J. (1990) Signal transduction by receptors with tyrosine kinase activity. Cell 61,203-212. 3. Aaronson, S. A. (1991) Growth factors and cancer. Science 254,1146-l 153. 4. Wang, C. and Smith, R. L. (1975) Lowry determination of protein in the presence of Triton X-100. Anal. Biochem. 63,414-417. 5. Smith, P. K., Krohn, R. I., Hermanson, G. T., Mallia, A. K., Gartner, F. H., Provenzano, M. D., Fujimoto, E. K, Goeke, N. M., Olson, B. J., and Klenk, D. C. (1985) Measurement of protein using bicinchoninic acid. Anal. Biochem. 150,76-85. 6. Grimaux, M., Lain&Bidron, C., and Magdelenat, H. (1989) Immunoenzymetric assay of epidermal growth factor receptor. Tumor Biol. 10,215-224. 7. Scopes, R. K. (1987) Separation by precipitation, in Protein Purification. Principles and Practice, 2nd ed., Springer-Verlag, New York, p. 5 1.
CHAPTER21
Expression and Secretion of a Soluble Form of Myelin-Associated Glycoprotein John Attia,
Saqjoy
(MAG)
Gupta, and Robert J. Dunn
1. Introduction Myelin-associated glycoprotein (MAG) was first identified as the major glycoprotein in the central nervous system (CNS) through 3Hfucose-labeling experiments (1). It is expressed on the surface of glial cells of both the CNS and peripheral nervous system (PNS). During development, it is localized throughout the wraps of loose myelin; after compaction, it is restricted to Schmidt-Lanterman incisures, paranodal loops, outer mesaxon, and in particular to the glial-axon interface (2,3). The protein is thought to play a role in maintaining the periaxonal space. MAG has also been postulated to be involved in glial-neuron adhesion during myelination. This is based on the finding that anti-MAG antibodies inhibit oligodendrocyte-neuron adhesion by 25% (4), and that MAG liposomes bind a variety of neuronal cells (5). Observations from in vitro myelination cultures also support a role for MAG in Schwann cell adhesion, migration, and elongation along neurites (6,7). Cloning and sequencing of the MAG gene (8,9,lO) show that it encodes a protein having five immunoglobulin folds in its extracellular domain, a hydrophobic transmembrane sequence, and two possible cytoplasmic tails, generated by alternative splicing. There are eight N-linked glycosylation sites in the polypeptide. For MAG, as for many other proteins, biochemical analysis has been limited by an inability to prepare From
Methods in Molecular B/ology, Vol 39 Baculovirus Expression Protocols E&ted by C. D. RIchardson Q 1995 Humana Press Inc , Totowa, NJ
363
364
Attia, Gupta, and Dunn
sufficient quantities of pure protein; MAG constitutes c 1% of all myelin proteins. We consequently studied the potential of baculovirus expression vectors for synthesizing a soluble form of MAG. In this chapter, we summarize our work with baculovirus vectors expressing MAG. We describe our constructs, the optimal culture conditions, which includes choice of media (Grace’s vs EXCELL400), cell lines (Sf9 vs Sf21), our problems with proteolysis, and our purification protocol. We also describe the use of the MAG signal sequence to direct cell secretion of another protein expressed with the baculovirus expression system-ciliary neurotrophic factor (CNTF). CNTF, along with certain other secreted factors, such as acidic fibroblast growth factor (II) and interleukin 1b (12), lacks the normal signal peptide and appears to be secreted from mammalian cells by routes not involving the signal recognition pathway, However, previous experiments involving expression of CNTF in HeLa cells resulted in cytoplasmic accumulation with no detectable secretion (13). We detail the construction of a generally applicable “export” vector used to fuse the MAG signal sequenceto the amino terminus of CNTF; restriction sites were engineered on to the ends of the CNTF open reading frame, which allowed easy removal of the cassette from the shuttle vector permitting its insertion into the baculovirus expression vector pJVP10. 2. Materials
Most of the materials used in the following experiments have been described extensively in the preceding chapters. However, the following basic supplies were employed in the following studies. 1. Spodopteru frugiperda (Sf9) cells were provided by M. D. Summers (Department of Entomology, Texas A & M University) while St21 cells were purchasedfrom Invitrogen (SanDiego, CA). 2. Grace’s medium, fetal calf serum (FCS), fungizone (amphotericin), and gentamycin were purchased from GIBCO/BRL (Gaithersburg, MD). Yeastolateand lactalbumin hydrolysatecamefrom Difco (Detroit, MI). 3. EXCELL400 serum-free medium was purchasedfrom J. R. Scientific (Lenexa, KS). 4. 20-Hydroxyecdysone (20-HE) cell-growth supplement was supplied by Sigma (St. Louis, MO), and a stock was preparedin 95% ethanol. 5. The pAc373 vector was suppliedby M. D. Summers(Departmentof Entomology, Texas A & M University); the pJVPl0 vector was provided by C. D. Richardson (Ontario Cancer Institute, Toronto) and is described
sMAG Expression and Secretion
6. 7. 8. 9. 10. 11, 12.
365
further in Chapter 9 of this book. Analogous vectors, such as BlueBac I (pJVETL) and BlueBac II (pETL), can be purchased from Invitrogen (San Diego, CA). Lipofectin transfection agent came from GIBCWBRL. SeaPlaque agarose came from PMC Bioproducts (Rockland, ME). Bluo-gal was purchased from GIBCO/BRL and was made 50 mg/mL in N,N-dimethylformamide. Circular or linearized AcNPV DNA was supplied by Invitrogen. Monoclonal antibodies directed against MAG were obtained from N. Latov (Columbia University, NY). ImmuIon multititer plates for ELISA assayscame from Dynatech (Alexandria, VA). Dynatech 5000 ELISA reader equipped with the ANOVA costat program for data analysis.
3. Methods 3.1. Cloning of the sMAG Gene into the Transfer Vectors 1. The pAc373 construct containing a truncated version of MAG was described by Johnson et al. (14). In summary, it consists of the pAc373 vector (IS) with the sMAG gene inserted at the unique BamHI site (pAcsMAG). In this construct, the polyhedrin promoter is intact up to position -8 nucleotide, followed by the sMAG gene, including 86 nucleotide of 5’ noncoding region before the initiation codon. 2. The pJVPl0 vector obtained from C. Richardson is a modified version of the pJVNhe1 vector (16) described in Chapter 9 of this book. It is similar to pVL941 (17), but incorporates the P-galactosidase gene under the p10 promoter to allow easy identification of recombinant viruses; it accommodates insertion at an WeI site. Into this vector, we have inserted a modified sMAG gene lacking the 5’ noncoding region. 3. The modified sMAG gene without its 5’ noncoding region was constructed using PCR to synthesize the first half of the sMAG gene, from nucleotide +1 to the EcoRI site at position +735. The 5’ oligonucleotide consisted of: 5’ GCGAATTCGGATCCTATAAAT~ATATI’CCTTACCACCC 3’ EcoR BamHl start sMAG gene codon The TATAAAT sequence was inserted between the BumHI site and the start codon to mimic the sequencepreceding the wild-type polyhedrin start codon. The 3’ oligonucleotide was: 5’ GCCCTCAATGGCCTCCACAGAGG
3’
366
Attia, Gupta, and Dunn
which is the sequence immediately downstream (3’) of the EcoRI site mentioned above. PCR was carried out for 30 cycles, each consisting of: 30 s, 94OC;30 s, 55OC;60 s, 71°C. This yielded a PCR product of 782 bp, which could be subcloned into pBSsMAG (sMAG inserted at the BamHI polylinker site of PBS) using EcoRI, thereby replacing the first half of the previous sMAG gene; this was called rsMAG, 4. Sequencing was carried out to verify that no Tuqpolymerase incorporation errors were present. Digestion with BumHI excised the whole gene as an rsMAG cassette. The ends were rendered blunt-ended with DNA polymerase (large fragment) as described by Maniatis et al. (18), and ligated into NheI-digested, blunt-ended pJVPl0 vector. In summary, this yielded a construct consisting of the complete polyhedrin leader sequence up to +l nucleotide, a mutated polyhedrin start codon (AlT) and 40 nucleotides of polyhedrin sequence followed by the sMAG gene (pJVrsMAG in Fig. 1). 5. In both these vectors, the 3’ ends of the sMAG gene are identical. As described in Johnson et al. (14), the MAG gene has been truncated so that the polypeptide chain for which it codes terminates three amino acids before the transmembrane segment. The protein is therefore secreted into the medium.
3.2. Isolation
of Recombinant
Baculovirus
1. Baculovirus manipulations and isolation of recombinant viruses were performed according to Summers and Smith (19), with the exception that lipofectin reagent was used instead of calcium phosphate for the cotransfection of genomic DNA (2 pg) and transfer vectors (4 pg) into insect cells; DNAs in a total volume of 50 pL of sterile water were mixed with 50 pL of lipofectin reagent, added to a 25-cm* monolayer of Sf9 cells in serum-free medium, and left overnight. 2. Complete medium was added and the incubation continued for 5 d. Recombinant viruses were isolated after three rounds of plaque purification. 3. After amplification, titers of the pAc373-based virus stocks were assayed using the agar-overlay method, and the occlusion body-negative plaques were counted after 5 d (20). Titers of the pJVPlO-based virus were performed by including 150 /.tg/mL Bluo-gal in the agar overlay and counting blue plaques after 3-4 d as described in Chapter 9 of this book.
3.3. Cell-Culture
Procedures
1. Grace’s medium was supplemented with 3.4 g/L yeastolate and lactalburnin hydrolysate and 10% FCS. All media contained 10 pg/mL gentamycin. 2. For comparison of cell lines, Sf9 cells or Sf21 cells were seededat a density of 3-4 x lo6 tells/80-cm* flask and infected with the baculovirus JVrsMAG
~04’ A SMAG ccdmg sequence
I
1
200 bp
/------
<
pACsMAG PH 5’ ul sq”mc8
,--
(50 MS)
//-~taaaaaaafc -20
MAO 5’ ut (86 nts)
sMAG cndmg sequerca
cgagatccgcggatccccgactggccact-//--agctcgcccacttgctggacaag An;
ATA TTC CIT --met 1le phe 1eu
syntheilc
knker
-to
PJVrsMAG PH 5’ td sequence
&
AcNPV
//-aaataaaaaaacctataaata JO -2'0
PH 5’ ul sequence < t-+ataaaaaaaictataaata -*o
PH coding sequence
(Se 111s)
(+I to +40)
synthetttlc knker
a~ccggattattcataccgtcccaccatcgggcgtgct bats)
(55 nts)
aggatcctataaat
/
sMAG coding sequerce /
Al% Pm. TIC cl7 met rle phe leu
---
PH cot6ng sequence / flT&
-10
CCC CAT T A T TCA TAC CGT ccc
met pro
asp
tyr
ser
tyr
arg
pro
KC
ATC c7.x --
thr
11e gly
Fig. 1. The mRNA 5’ noncoding leader sequences of the two baculovirns constructs. The schematic at the top shows the complete sMAG expression gene. mRNA transcription is initiated from the polyhedrin promoter at -58 nucleotide, and continues into the sMAG gene and the remainin g 3’ section of the polyhedrin gene until the polyadenylation signal. The pAcsMAG vector mcludes the polyhedrin 5’ noncoding sequence up to position -8 nucleotide, a synthetic linker, and the sMAG gene, including 86 nucleotides of 5’ noncoding region. In the pNrsMAG vector, the entire polyhedrin 5’ noncoding region is preserved, the polyhedrin start codon is mutated to ATT, and the sMAG gene, with the 5’ noncoding region removed, inserted at posrtion +40 nucleotide. For reference, the wild-type polyhedrm sequence is shown at the bottom. Noncoding sequences are in lower-case letters, and the coding sequence in upper case. The AT-rich sequence preceding the ATG codon is in bold type.
368
Attia, Gupta, and Dunn
at an MO1 of 10 m 10 mL of Grace’s medium. Aliquots of 0.5 mL were taken at day 1 through day 5 after infection; cells were centrifuged and the supernatant stored at -20°C until the last day of the comparisons, when ELISA assayswere performed (see Note 1). 3. For the comparison of expression levels produced by viruses derived from the vectors, pAc373 and pJVPl0, i.e., baculoviruses AcsMAG vs JVrsMAG, 250 mL of Sf9 cells were grown in each of two spinner flasks at a density of 1.5 x lo6 cells/ml and infected at an MO1 of 2 in Grace’s medium, Aliquots of 30 mL were taken at each day, the cells being used for RNA extraction and the supematants for the ELISA. 4. For the comparison of media, four flasks were seeded with 34 x lo6 Sf21 cells and infected with JVrsMAG at an MO1 of 10. The first was grown in Grace’s for all 5 d; the second was grown m EXCELL-400 for 5 d; the third was also grown in EXCELL400, but all the medium was harvested after the second d and fresh medium added for the remainder of the time; the fourth was also grown m EXCELL-400, but all the medium was harvested after the second d and fresh medium containing 2 mg/mL 20-HE added for the remainder of the time. The yields for the latter two flasks at day 3,4, and 5 represent a cumulative amount, i.e., the amount of sMAG in the harvested medium at day 2 plus the amount in the supematantat day 3,4, or 5. 5. During the first of three rounds of plaque purification carried out to obtain a pure recombinant baculovirus, a number of plaques were randomly picked and independently carried through to pure stocks.For the comparison between these isolates, four flasks seeded with 3-4 x lo6 Sf21 cells were infected with JVrsMAG (isolate #2) and three other isolates (#6, 8, 10) at an MO1 of 10. 3.4. ELISA Methods for Quantitation of Expression 1. Immulon plates were coated with 2 mg/mL sMAG protein in O.lM carbonate buffer, pH 9.6, overnight at 4”C, and blocked wrth PBS-l% BSA for 2 h at 25OC. Dilution series (1:2) of media blanks, standards, and samples were prepared in PBS-0.25% BSA-0.05% Tween, and preincubated with a fixed, titrated amount of GenS3, an anti-MAG monoclonal antibody (generously provided by N. Latov, Columbia University). 2. After 30 min, thesedilutions treated with Gen S3 were added to the nucrotiter wells. After 1h, wells were washed three times with PBS-0.25% BSA-0.05% Tween, and a fixed amount of goat-antimouse IgG-alkaline phosphataseconjugated secondary Ab was added. After 1 h, wells were washed three times with 150 mM NaCl-0.05% Tween and developed with 100 p.L of 10 mg/mLp-nitrophenyl phosphate m 1M diethanolarnme buffer, pH 9.8. After 10 min the reaction was stopped with 25 p,L of 2M NaOH and read at 405 nm.
sMAG Expression
and Secretion
369
3. All samples were done in duplicate and sMAG standards were run on each plate. Each set of comparisons was based on two independent experiments. Readings were expressed as percent inhibition (R) relative to the appropriate blank (i.e., medium), and a K-value (log ccl-R>/R>) was calculated. A logit plot, using K values as the Y-axis, and log [sMAG] as X-axis, was used to translate the ELISA readings into protein yields, and these results were compared using a two-way randomized complete block ANOVA (CoStat program). 3.5. RNA Preparation Methods 1. Cell pellets obtained as described in Section 3.3. were stored at -70°C until the last aliquot was taken. 2. RNA extraction was performed essentially as described by Chirgwin et al. (21). RNA concentration was determined spectrophotometrically. 3. Seven micrograms of each sample were denatured and 2 pg loaded/lane on a 1.5~mm thick vertical formaldehyde agarose (1.5%) gel. After electrophoretic separation, the gel was denatured in 50 rniWNaOH, 150 mM NaCl for 10 min, neutralized in 100 m&f Tris-HCl buffer, pH 7.5,150 mM NaCl, and transferred by capillary action onto nitrocellulose (Stratagene) overnight using 6X SSC. The blot was rinsed in 2X SSC and UV crosslinked at 254 nm, with 0.12 J/sq.cm irradiant energy using a Stratalinker apparatus (Stratagene, San Diego, CA). The EcoRI fragment of sMAG was labeled using a random hexanucleotide kit (Pharmacia, Uppsala, Sweden) and used to probe the blot. 3.6. Protein Purification and N-Terminal Sequencing 1. Cells were spun down at 98OOgfor 10 min, and the supernatant subjected to a 60% ammonium sulfate cut. The precipitated protein was resuspended in 10-20 mL of buffer B (10 mM Tris, pH 7.5/50 mM NaCl/0.8% [w/v] CHAPS/l % [w/v] betaine/l mM D’M’) and dialyzed against 1 L of 10 m&f Tris, pH 7.5/.025% CHAPS/.OS% betaine, with three changes of buffer (see Note 2). 2. Particulate matter was centrifuged and the supernatant loaded onto an HR lo/30 Fast Q FPLC column (Pharmacia) in buffer A (20 mZt4Tris, pH 7.5, 1% betaine, 1 mM DTT). After a wash with 10 mL of buffer B, a number of step elutions were carried out with buffer AIlMNaCl: O-15,15-35, and 35-lOO%, all at a flow rate of 2.5 mL/min and fraction sizes of 5 mL. 3. The 15-35% fractions containing sMAG were loaded onto an HR 16/10 lentil lectin-sepharose 4B column (Pharmacia) and washed with buffer A/ 0.5M NaCl. A step elution was performed with buffer A/0.5M NaCl/O.SM a-methyl mannoside (Aldrich) at a flow rate of 1 mL/min, with collection
370
Attia, Gupta, and Dunn
of 1 ml-fractions. The eluate was diluted 3- to 5-fold with distilled, deionized water and loaded onto an HR 5/20 Fast Q column. The column was washed with buffer C (50 mMMES buffer, pH 5.5/l% betaine) and eluted with a step from 10 to 25% with buffer CIlM NaCl. 4. The eluate vol, usually 8-10 mL, was reduced to approx 1 mL using a SpeedVac, and this was dialyzed against 10 mM sodium phosphate buffer, pH 7.4 using microcollodion bags (Sartorius, Goettingen, Germany). 5. Purified sMAG samples were subjected to gas-phase sequencing in a PI 209OE Integrated Micro-Sequencing System (Porton Instruments Inc.). Edman degradation was performed according to standard procedure as recommended by the manufacturer. On-line phenylthiohydantoin-amino acid analysis was carried out using a gradient elution (solvent A: triethylaminel acetic acid/tetrahydrofuran, pH 4.0; solvent B: acetomtrile) from a reversephase Aminoquant (200 x 2.1 mM) column (Hewlett Packard, Palo Alto, CA) with a flow rate of 0.2 mL/min at 42OC.
3.7. Directed Cell Secretion of Baculovirus-Expressed Proteins A schematic diagram of the secretion shuttle vector is shown in Fig. 2A. To construct pSHONEX 1.1, an EcoRI-Sac1 fragment of pBSrsMAG (nucleotides -19 to 288, numbered according to ref. 8) was subcloned into the EcoRI/&zcI sites of pSP64 (Promega, Madison, WI), and the resulting plasmid named pSHONEX 1.0. To introduce conFig. 2. (opposite page) A: A schematic representation of the transfer vector, pSHONEXl.1. ORI, lacZ, and Ap denote the E. coli origin of replication, the P-galactosidase a-fragment and the P-lactamase genes found in the pUC series of vectors (22). Arrows on the plasmid map indicate the direction of transcription. The stuffer fragment denotes the region to be replaced with the cDNA of interest during subcloning procedures. The sequence of constructs in pSHONEXl.1 may be easily determined using the Ml3 forward and reverse primers (the forward primer anneals 5’ to the signal sequence, whereas the reverse primer anneals 3’ to the 3’ polylinker site shown in the diagram). B: Sequence of the 5’ primer used to engineer the CNTF cDNA for subcloning into pSHONEXl.1. The primer used to modify the 5’ end of the CNTF cDNA contains a restriction site for XhoI, and nucleotide sequence encoding three additional amino acids of the MAG signal sequence, in frame with the initiation codon for CNTF. At the 5’ end of the primer, three extra nucleotides were introduced upstream to the XhoI site to ensure proper digestion with the restriction enzyme.
A
I I I GAATKGCTAGCTCTAGAGATCCTATAAAT
start ate of MAO slgnal sequence v I ATGATAnCC~ACCACCCTGCCCTG~GGATAATGATTTCAGCnCTCGAGGGGGGCAC Met Ile
r
Phe Leu Thr Thr Leu Pro Leu PheTtp
Stuffer
fragment
TGGGGTGCCTGGATGCCCTCGTCCATCTCT~AT~T~~~C~~~T~
B Xhol
Sequence of CNTF cDNA I 1 GCT T:T CGA GGG GGA CAC AT0 GCT TTC GCA GAG CA Ala Ser Arg Giy Gly HIS Met Ser Phe Ser Glu
Ile Met Ile Ser Ala
SerArg
Gly Gly HIS
I, I,
Sac1 I
Ad
I,
Smal
Xbal I
-GATCTCTAGAGCTAGCGAAlTC
Nhel I
EcoRl I
I
372
Attia, Gupta, and Dunn
venient NheI and XbaI sites 5’ and 3’ of the signal peptide, a BamHI
fragment of pSHONEX 1.Ocontaining the signal peptide sequence was excised, and blunt ended, with T4 DNA polymerase. Adaptor-linkers (5’ AATTCGCTAGCTCTAGA) and (5’ TCTAGAGCTAGCG) were ligated onto this fragment, creating a 204 bp insert with overhangs corresponding to an EcoRI restriction site. Meanwhile, pUC8 (22) was digested with SmaI and Hi&III, blunt-ended and recircularized to form a derivative pUCSDSH, containing only EcoRI as the unique restriction site of the original polylinker. The EcoRI-compatible fragment was subcloned into the EcoRI site of pUC8DSH to create pSHONEX 1.1 with the following features: 1. pSHONEX1.1 contains a mammalian signal sequence from MAG that is recognized by insect cells. The primary cleavages of the signal recognition sequencewere found to take place at Ser 17 and also at Gly 20 (23). 2. The cDNA of interest may be subcloned, using the XhoI site at the 3’ end of the signal sequence (Fig. 2A). PCR was employed to engineer a compatible 5’ XhoI site and additional nucleotide sequence encoding three amino acids, in frame with the start codon of the cDNA. The 3’ end of the cDNA (containing a termination codon) may be inserted into the 3’ polylinker site.
3. After preparation of the appropriate construct and verification of the sequence (using Ml3 forward and reverse primers), the fragment encompassing the signal sequenceand the fused cDNA may then be excised using the flanking NheI or X&I sites. Cohesive ligation of this insert into the NheI site of pJVPl0 1sfollowed by verifying the orientation of the insert for expression. 4. For example, to prepare CNTF as a secreted protein using the baculovirus expression system, the CNTF cDNA was first modified at the 5’ end using PCR. An XhoI site and nucleotide sequence encoding three additional amino acids were introduced upstream of the initiation codon of the CNTF cDNA (see Fig. 2B). Three extra nucleotides were introduced 5’ to the recognition sequence for XfzoI to ensure proper digestion of the PCR product with the restriction enzyme. The 3’ primer for PCR corresponded to the 3’ end of the cDNA containing a termination codon. The PCR product was blunt-ended, restricted with XhaI, and then subcloned into pSHONEXl.1 using the unique XhoI and SmaI sites in the vector. The construct was verified by sequencing and subsequently restricted with X&I (NheI was not used, since the CNTF cDNA contains an internal NheI site). The ensuing fragment was ligated Into pJV10. Proper orientation of the insert in pJVl0 was also verified by restriction analysis (see Note 4).
sMAG Expression and Secretion 3.8. Effect
of the 5’ Leader
373 Sequence
1. Mammalian mRNAs contain stretches of untranslated leader sequence positioned between the 5’ cap structure and the translational initiator AUG codon. These sequences may reduce the level of expression in heterologous systems, such as insect cells, through effects on translational initiation or mRNA stability. With this possibility in mind, we have removed the mammalian 5’ leader sequence of our original sMAG expression construct as shown in Fig. 1. In the original construct, derived from pAc373 and called pACsMAG, the 5’ leader is 154 nucleotides in length, consisting of the polyhedrin 5’ leader from position -58 to -8, 18 nucleotides of synthetic linker, and 86 nucleotides of MAG 5’ leader sequences. 2. The new construct, pJVrsMAG, was derived from the pJVPl0 vector, a modified form of pJVNhe 1(16), in which the normal polyhedrin ATG has been modified to ATI’ and a unique Nhe I site placed 40 nucleotides into the coding region. The 5’ leader sequence of the new construct contains the complete polyhedrin leader (58 nucleotides), 40 nucleotides of polyhedrin coding sequence rendered inactive by the ATT mutation, and a short synthetic linker to fuse the sMAG ATG to the polyhedrin gene. This short linker contains the AT-rich segment that precedes the polyhedrin ATG in the intact viral gene (Fig. 1). The viruses derived from these two vectors by homologous recombination, AcsMAG and JVrsMAG, should produce mRNA transcripts of 2619 and 2587 nucleotides, respectively.
3.9. Comparison of Protein and mRNA Expression from AcsMAG and JVrsMAG 1. We wished to directly quantitate the amount of sMAG protein secreted into the medium by various infected cell cultures. For this purpose, a competitive ELBA was developed; sMAG in the medium sample to be tested competed with sMAG coated on the microtiter well for binding to a titrated amount of GenS3, an anti-MAG monoclonal antibody (provided by N. Latov, Columbia University). A standard curve was included for reference, as well as the appropriate media blanks (see Note 5). 2. Four Spinner flasks of Sf9 cells were infected, two with each virus, AcsMAG and JVrsMAG. ELISAs were performed on the supernatants of the five daily aliquots and RNA was extracted from the corresponding cell pellets. Fig. 3A shows the time-course of protein expression with both viruses. AcsMAG makes slightly more protein than JVrsMAG during the first 3 d but the latter then overtakes to reach a level almost 3.5 times that of the former at day 5.
374
Attia, Gupta, and Dunn
C
1
2
3 day8
4
5
Fig. 3. Levels of sMAG expression by the two different baculovirus constructs are plotted as a function of d in culture after infection (day 0). (A) Levels of sMAG protein, as determined by ELISA, produced from the two viruses are significantly different, ***j~ < .OOl.(B) Levels of sMAG mRNA, asdetermined by densitometry of the sMAG band on the Northern blot (Fig. 3). (C) Levels of
sMAG Expression
and Secretion
375
Fig. 4. Levels of sMAG mRNA in AcsMAG and JVrsMAG infected Sf9 cells grown in Grace’s medium. A northern blot with equal amounts of total RNA from each sample was hybridized with a probe corresponding to the first half of the sMAG gene. Lower-mol-wt bands represent degradation products. The position of the 18s r-RNA is shown for reference. 3. Fig. 4 shows a Northern blot of the corresponding RNA samples for each time-point. The sMAG transcripts are -2600 bp long and are migrating more slowly than the 18s rRNA, which corresponds to -1500 bp. The lower mol wt bands seen are degradation products of the sMAG transeript probably owing to the cell lysis occurring as the infection proceeds. The sMAG transcripts follow the same pattern as the protein expression, with levels of transcript from AcsMAG being higher at day 2 and 3, and then JVrsMAG overtaking at day 4 and 5. The relative densitometric values are graphed in Fig. 3B. At day 2, AcsMAG makes 4.5 times the amount of mRNA; the levels become almost equal at day 3, and by day 5, JVrsMAG makes almost 2.5 times the level of mRNA. sMAG protein production by four different isolates of JVrsMAG. Virus isolate #lo produces slightly less protein than the other three isolates, *p < .05. Each time-point is the average of two independent experiments. Error bars represent local SEM.
376
Attia,
Gupta, and Dunn
4. A number of studies have extensively characterized the 5’ mRNA leader sequencesneeded for maximal expression of foreign genes in baculovirus vectors (17,24,25). From these studies, we can surmise that one reason for the increase seen here may be that the sMAG mRNA produced from pJVrsMAG has the complete polyhedrin 5’ leader sequence, whereas that in pAcsMAG is truncated at -8 nucleotides. This indicates that the TATAAAT sequenceimmediately preceding the initiator AUG codon may be important and its inclusion in the synthetic linker before the sMAG gene of JVrsMAG may contribute to the increased expression. Another explanation may be that in the pJVrsMAG construct, we have removed the long stretch of mammalian 5’ noncoding DNA preceding the sMAG start codon. A similar effect was noted by Forstova et al. (26); there was a lo-fold increase in expression using a pVL941 vector with no sequence upstream of the start codon of the inserted gene vs a pAC373 vector with 18 nucleotides of 5’ noncoding region. 5. The Northern blot studies in Fig. 4 indicate that the increased expression from JVrsMAG is the result of the accumulation of higher levels of sMAG mRNA rather than more efficient translation of sMAG mRNAs. The pattern of sMAG mRNA levels from both viruses parallel those obtained at the protein level. This increase in mRNA arises from the changes in the 5’ mRNA leader sequence,because the promoter, coding sequences,and 3’ regions of the expression constructs are identical (Fig. 1). The increased steady-state levels of mRNA could result from increased transcription from the polyhedrin promoter, increased stability of the sMAG mRNA, or a combination of the two factors. Our experiments do not allow a distinction to be made between thesepossibilities. A similar increase in expression and corresponding mRNA transcription and/or stability was seenby Luckow and Summers (17) when comparing CAT mRNA levels in a pAc373-based virus versus a pVL941-based virus. pJVPl0 is almost identical to pVL941 in the region of the start codon, except that it usesan ZVheIcloning site instead of BumHI.
3.10. Comparison
of the Virus Isolates
1. During the isolation and amplification of JVrsMAG, a number of virus isolates, presumably (but not necessarily) arising from independent homologous recombination events, were carried through to pure stocks. To see if there was variability between these recombination events, three of these isolates were randomly chosen and tested for levels of sMAG expression against the isolate that was selected for final amplification. Tissue-culture flasks seededwith the same number of Sf21 cells were infected with JVrsMAG (isolate #2) and three other isolates (#6, #8, #lo). Aliquots of media were taken at each day and read by ELISA for sMAG levels.
sMAG Expression
and Secretion
377
2. Fig. 3C shows that all four isolates follow approx the same ttme-course of protein expression, reaching a peak at day 3, with slight increases or decreases at days 4 and 5. Using the nonsignificant ranges generated by ANOVA, it appears that isolate #lO makes significantly less protein than the other three isolates though the differences are relatively small. 3.11. Comparison of Cell Lines 1. Unpublished results indicate that although recombinant baculoviruses can direct the synthesis of large amounts of soluble protein, much of the foreign product remains in the cell probably because of overwhelming the secretory capacity of the insect cells. Other cell lines, e.g., Sf21, and Mb 0503, have been reported to be more efficient m this respect than the original Sf9 line used by Smith et al. (IS). Flasks were seeded with either Sf9 or Sf21 cells and infected with JVrsMAG. 2. Aliquots were taken on days 1 through day 5 following infection; cells were centrifuged and the supernatants quantitated for sMAG with the competitive ELISA. The same experiment was repeated independently a second time, and the results combined to give the time-course of expression shown in Fig. 5A. It appears that both cell lines produce the same amount of protein during the first 2 d, after which Sf9 expression stays constant, but Sf21 expression climbs to a peak of approx three times the Sf9 level at day 4. This effect is statistically significant, with p < .005. There may be some sMAG catabolism occurring at day 5 since the protein level drops. 3. Another point is that in the previous comparison of AcsMAG and JVrsMAG, the time-course of protein expression (Fig. 3A) is somewhat delayed from that seen in this comparison and the next (Fig. 5A and B). This is owing to the fact that the viral comparison in Fig. 3 was carried out in Spinner flasks where, because of practical considerations of the amount of virus stock needed to infect these flasks, we routinely use an MO1 of 2. All other comparisons were carried out in monolayer cultures at an MO1 of 10, leading to an earlier peak of protein expression. 3.12. Comparison of Media 1. Recently a number of serum-free media for insect cell lines (e.g., EXCELL-400, SF-900) have been marketed, and we have tested the relative efficacy of EXCELL-400 for the production of sMAG protein. In addition, it has been reported that addition of 20-hydroxyecdysone (20HE), a moulting hormone, increases the levels of expressed protein (27). 2. To address these questions, tissue-culture flasks were seeded with Sf21 cells and infected with JVrsMAG at an MO1 of 10.
Attia, Gupta, and Dunn
378
0 1
2
3 days
4
5
1
2
3 days
4
5
0
Fig. 5. Effects of cell lines and media on protein expression. (A) sMAG protein levels produced by Sf9 and St21 cells grown in Grace’s medium are plotted as a function of day in culture after infection (day 0). The levels of protein made by the cell lines are significantly different, **p < .005. (B) sMAG protein levels produced by Sf9 cells grown m Grace’s and EXCELL-400 (see Sections 3.11. and 3.12.). The three conditions with EXCELL400 are similar to each other, but significantly different from that with Grace’s, **p < .005. Each timepoint represents the average of two independent experiments. Error bars represent local SEM. 3. Cells were grown in Grace’s medium, EXCELL400, EXCELL400 with a complete change of medium at day 2, and EXCELL-400 with addition of 20-HE at day 2, as described in Section 3.3.4. The same comparison was performed on a second series of four flasks, and the combined time-course is shown in Fig. 5B. The three serum-free media conditions all grve very similar yields that are approximately a quarter of the levels obtained usmg
&TAG Expression
4.
5.
6.
7.
8.
9.
and Secretion
379
serum-containing Grace’s. The levels of sMAG obtained with the change of medium at day 2 is the same regardless of the addition of 20-HE. This level appears to be higher than that observed without any change of medium, but this difference is not statistically significant. In all cases,protein yield decreases during day 4 and 5. Although Grace’s does give a higher yield of protein, the presence of 10% FCS renders the purification much more laborious; the albumin in particular is very sticky and has proven difficult to separate from the sMAG even using Sepharose-Blue. It was for this reason that we had started using the serum-free media. N-terminal sequencing of sMAG purified from cells grown in EXCELL400, however, gave a number of signals. Many of these appeared to be from a number of low-mol-wt contaminants; this made assignment of a sequence difficult. Nevertheless, a major signal was tentatively designated as leugl-leu-gly-asp-leu, indicating that cleavage had occurred between arggo and leugl. We surmised that the absence of serum, and the inhibitors it contains, allowed proteases to act on the secreted sMAG. The site of cleavage indicated a trypsin or papain-like activity. This should have given rise to a peptide approx 8 kDa smaller than the expected full-length sMAG; this was not confirmed on SDS-PAGE (Fig. 6). Nevertheless, this possibility was not completely dismissed, since other groups expressing secreted proteins reported having papain-like proteolysis problems as well (C. Richardson, personal communication). We returned to using Grace’s medium, but with the FCS content reduced to 3%; N-terminal sequencing of protein purified from cells grown in this medium gave much cleaner results. There were three major signals: sMAG with signal peptide cleavage at the predicted glyzO,as reported for mammalian cells (28), fetuin (or an N-terminal fragment thereof), and sMAG with signal peptide cleavage at ser17.The estimated ratios from the HPLC traces were 1: 1.5:2.7, respectively. Fitting the algorithm of von Heijne (29) to predict signal peptide cleavage sites, it was found that the ser17site was preferred over the glyzO;this is consistent with the ratio of -3:l (ser:gly) estimated above. What was not consistent was the indication that almost one-third of the protein was fetuin; SDS-PAGE indicated that the majority of the purified protein was sMAG, which was accompanied by a minor contaminant of -63 kDa; the purified sMAG and the contaminant were present at a ratio of -10: 1. To check if this -63-kDa contaminant was fetuin (-48 kDa), commercially available protein was purchased from Sigma and run on SDS-PAGE in wells adjacent to purified sMAG. The contaminant migrated closely with fetuin, but we could not be certain of the assignment. There are a number of pos-
Attia, Gupta, and Dunn
380
ABC
Fig. 6. Comparison of sMAG isolated from Sf9 cells grown in Grace’s medium/3% FCS (lane 1) and in EXCELL400 medium (lanes 2 and 3, two separate purifications). sMAG isolated under both conditions comigrate on SDS-PAGE gels, indicating a lack of proteolysis; -325 ng of protein was loaded in each lane. sibilities to explain the variable ratios of contaminant seenwith both methods: a. The fetuin contaminant stains very poorly compared to sMAG. b. Fetuin and sMAG bind tightly and comigrate even on denaturing and reducing SDS-PAGE. c. Coupling of fetuin during the sequencing procedure is much more efficient than sMAG. d. The minor 63 kDa component on SDS-PAGE is not fetuin and the fetuin signal in sequencing comes from a low-mol-wt fragment not seen on the gels. Judging from the signal strengths of the HPLC sequencing runs, it does not appear that a majority of the sMAG is blocked at the N-terminus. We believe the latter two possibilities are the most likely. 20. In a previous report (27), it was stated that 20-HE increased the yield of protein. Apparently, during the first 2 d of infection, an extracellular enzyme was produced that catabolized this steroid, and so a protocol was set up such that a change of medium was performed after the second day, adding either simple medium or medium supplemented with 2 mg/mL 20-
sMAG Expression
and Secretion
381
HE. In our hands, changing the medium at day 2 appears to increase the yield at the later time-points but this does not prove to be statistically significant, regardless of the addition of the hormone. (see Note 5)
3.13. Secretion
of CNTF from Baculovirus-Infected
Cells
1. The CNTF construct in pJVPl0 was transfected into Sf9 cells as described in Section 3.2. Isolation of recombinant virus was followed by four rounds of plaque purification. Infections of Sf9 cells with recombinant virus was carried out as described in Section 3.3. 2. To assay for production and secretion of CNTF, lOO+.L aliquots of the culture medium were assayed for biological activity on successive d after infection (30). Dissociated neuronal cells from embryonic day 8 chick ciliary ganglia were plated in the presence of serial dilutions of the insect cellculture medium (1:103 to 1:106). 3. After 24 h of culture, neurons with processes greater than two cell diameters were counted. One trophic unit was defined as the amount of material yielding half-maximal biological activity when diluted in 1 mL (31). After 3 d of infection, the biological activity of the insect culture medium was 1.0-1.5 x 105trophic U/n&, which translates to approx 2.5 mg/mL or 2.5 mg/L culture medium.
4. Notes 1. After the results of this study, we are now using ,921 cells routinely grown in spinner flasks in Grace’s medium with 3% FCS and infecting at an MO1 of 2. We now obtain expression in the range of l-2 mg of purified protein/ L cells. Unfortunately, St’21cells are not as adherent in monolayer cultures and they grow half as quickly as Sf9 in suspension cultures. This drawback can be eliminated by maintaining the cells at a minimum density of 0.5 x lo6 cells/mL. Apart from this caveat, these cells appear to be as hardy as the Sf9 line and have been growing quite well in our facility. 2. In the absence of a reliable functional assay, the conformation of the baculovirus-produced sMAG is difficult to test; nevertheless, this sMAG does react with a conformation-specific monoclonal antibody, 513 (4), on Western blots of nonreducing SDS-PAGE gels, 3. In agreement with Kuroda et al. (32), not only do we find high mannose and trimmed back carbohydrates on sMAG, but we also find indications of fucosylation (unpublished observations). 4. We have found that ligations involving XhoI-digested fragments tend to be troublesome. This was solved by testing XhoI obtained from different companies or treating the digested DNA fragments with proteinase K
before ligation.
382
Attia, Gupta, and Dunn
5. As described previously, all the comparisons were done in duplicate, and although the trends were the same in all cases, the absolute amount of protein varied quite widely in some experiments. We have noticed this even on the preparative scale; there is variability between one infection and the next even when the same procedure and virus stock are used. For this reason, we have restricted ourselves to making conclusions only in terms of relative increases in protein and not absolute amounts.
References 1. Quarles, R. H., Everly, J. L., and Brady, R. 0. (1973) Evidence for the close association of a glycoprotein with myelin m rat brain. J. Neurochem. 21, 1177-l 191. 2. Sternberger, N. H., Quarles, R. H., Itoyama, Y., and Webster, H. D. (1979) Myelinassociated glycoprotein demonstrated immunocytochemically in myelin and myelm-forming cells of developing rat. Proc. N&l. Acad. Sci. USA 76,1510-1514. 3 Trapp, B. D. and Quarles, R. H. (1982) Presence of the myelin-associated glycoprotein correlates with alterations in the periodicity of peripheral myelin. J. Cell Biol. 92,877-882.
4. Poltorak, M., Sadoul, R., Keilhauer, G., Landa, C., Fahrig, T., and Schachner, M. (1987) Myelin-associated glycoprotein, a member of the L2/HNK-1 family of neural cell adhesion molecules, is involved in neuron-oligodendrocyte and oligodendrocyte-oligodendrocyte interaction. J. Cell Biol. 105, 1893-1899. 5. Sadoul, R., Fahrig, T., Bartsch, U., and Schachner, M. (1990) Binding properties of liposomes containing the myelin-associated glycoprotein MAG to neural cell cultures. J. Neurosci. Res. 25, 1-13. 6. Owens, G. C. and Bunge, R. P. (1989) Evidence for an early role for myelin-associated glycoprotein in the process of myelination. Gliu 2,119-128. 7. Owens, G. C. and Bunge, R. P. (1991) Schwann cells infected with a recombinant retrovirus expressing myelin-associated glycoprotein antisense RNA do not form myelin. Neuron 7,565-575. 8. Arquint, M., Roder, J. C., Chia, L.-S., Down, J., Wilkinson, D., Bayley, H., Braun, P., and Dunn, R. J. (1987) Molecular cloning and primary structure of myelinassociated glycoprotein. Proc. Natl. Acad. Sci. USA 84,600-604. 9. Salzer, J. L., Holmes, W. P., and Colman, D. R. (1987) The amino acid sequences of the myelin-associated glycoprotems: homology to the lmmunoglobulin gene superfamdy. J. Cell Biol. 104,957-965. 10. Lai, C., Brow, M. A., Nave, K.-A., Noronha, A. B., Quarles, R. H., Bloom, F. E., Milner, R. J., and Sutcliffe, J. G. (1987) Two forms of lB236/myelin-associated glycoprotein, a cell adhesion molecule for postnatal neural development, are produced by alternative splicing. Proc. Natl. Acad. Sci. USA 84,4337-4341. 11. Kandel, J., Bossy-Wetzel, E , Radvani, F., Klagsburn, M., Folkman, J., and Hanahan, D. (1991) Neovascularization is associated with a switch to the export of bFGF in the multistep development of fibrosarcoma. Cell 66,1095-l 104. 12. Rubartelli, A., Cozzolmo, F., Talio, M., and Sitia, R. (1990) A novel secretory pathway for interleukin lb, a protein lacking a signal sequence. EMBO J. 9,1503-1510.
sMAG Expression
and Secretion
383
13. Stockli, K. A., Lottspeich, F., Sendtner, M., Masiakowski, P., Carroll, P., Gotz, R., Lindholm, D., and Thoenen, H. (1989) Molecular cloning, expression and regional distribution of rat ciliary neurotrophic factor. Nature 342,920-923. 14. Johnson, P. W., Attia, J., Richardson, C. D., Roder, J. C., and Dunn, R. J. (1989) Synthesis of soluble myelin-associated glycoprotein in insect and mammalian cells. Gene 77,287296. 15. Smith, G. E., Summers, M. D., and Fraser, M. J. (1983) Production of human beta interferon in insect cells infected with a baculovirus expression vector. Mol. Cell. Biol. 3,21X-2165.
16. Vialard, J., Lalumiere, M., Vernet, T., Briedis, D., Alkhatib, G., Henning, D., Levin, D., and Richardson, C. (1990) Synthesis of the membrane fusion and hemagglutinin proteins of measles virus, using a novel baculovirus vector containing the beta-galactosidase gene. J. Virol. 64,37-50. 17. Luckow, V. A. and Summers, M. D. (1989) High level expression of nonfused foreign genes with Autographa californica nuclear polyhedrosis virus expression vectors. Virology 170,3 l-39. 18. Maniatis, T., Fritsch, E. F., and Sambrook, J. (1982). Molecular Cloning: A Luboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. 19. Summers, M. D. and Smith, 6. E. (1987) A manual of methods for baculovirus vectors and insect cell culture procedures. Texas Agricultural Experimental Station, College Station, Bull. 1555. 20. Wood, H. A. (1977) An agar overlay plaque assay method for Autographa californica nuclear polyhedrosis virus. J. Znver. Pathol. 29,304-307. 21. Chirgwin, J. M., Pryzybyla, A. E., MacDonald, R. J., and Rutter, W. J. (1979) Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry l&5294-5300. 22. Yanisch-Perron, C., Vieira, J., and Messing, J. (1985) Improved Ml3 phage cloning vectors and host strains: nucleotide sequences of the Ml3mp18 and pUC19 vectors. Gene 33,103-l 19. 23. Attia, J., Hicks, L., Oikawa, K., Kay, C. M., and Dunn, R. D. (1993) Structural properties of the myelin-associated glycoprotein ectodomain. J. Neurochem. 61, 718-726. 24. Matsuura, Y., Possee, R. D., Overton, H. A., and Bishop, D. H. L. (1987) Baculovirus expression vectors: the requirements for high level expression of proteins, including glycoproteins. J. Gen. Viral. 68, 1233-1250. 25. Luckow, V. A., and Summers, M. D. (1988) Signals important for high-level expression of foreign genes in Autographa californica nuclear polyhedrosis virus expression vectors. Virology 167,56-7 1. 26. Forstova, J., Krauzewicz, N., and Griffin, B. E. (1989) Expression of biologically active middle T antigen of polyoma virus from recombinant baculoviruses. Nucl. Acids Res. 17, 1427-1443.
27. Sarvari, M., Csikos, G., Sass, M., Gal, P., Schumaker, V., and Zavodszky, P. (1990) Ecdysteroids increase the yield of recombinant protein produced in baculovirus insect cell expression system. Biochem. Biophys. Res Commun. 167, 1154-1161.
384
Attia, Gupta, and Dunn
28. Pedraza, L., Owens, G. C., Green, L. A. D., and Salzer, J. L. (1990) The myelinassociated glycoproteins: membrane disposition, evidence of a novel disulfide linkage between immmunoglobulin-like domains, and posttranslational palmitylation. J. Cell Biol. 111,265 1-2661. 29. von Heijne, G. (1986) A new method for predicting signal sequence cleavage sites. Nucl. Acids Res. 14,4683-4690.
30. Gupta, S. K., Altares, M., Benoit, R., Riopelle, R. J., Dunn, R. J., and Richardson, P. M. (1992) Preparation and biological properties of native and recombinant ciliary neurotrophic factor. J. Neurobiol. 23,48 l-490. 31. Barbin, G., Manthorpe, M., and Varon, S. (1984) Purification of the chick eye ciliary neurotrophic factor. J. Neurochem. 43,1468-1478. 32. Kuroda, K., Geyer, H., Geyer, R., Doerfler, W., and Klenk, H.-D. (1990) The oligosaccharides of influenza virus hemagglutinin expressed in insect cells by a baculovirus vector. Virology 174,418-429.
CHAPTER22 Comparison of Different Cell Lines for the Production of Recombinant Baculovirus Proteins Thomas J. Wickham, Glen R. Nemerow, H. AZan Wood, and M. L. Shuler 1. Introduction 1.1. Why Investigate New Insect
Cell Lines?
The levels of many recombinant proteins produced using the baculovirus expression vector (BEV) system are extraordinarily high (1,2), in many cases approaching 1000 mg/L, the level of polyhedrin expression. Unfortunately, other proteins that have been expressed in the BEV system are produced at much lower levels, which are often <5 mg/L (Table 1). Certainly, vector construction plays a major role in the level of protein expression, and many of the early reports of lower level expression using BEV can probably be traced to the use of suboptimal vectors that lacked sequencesimportant to high level transcription from the polyhedrin promoter (3). However, more recent data on baculovirus expression levels using improved transfer vectors continue to reveal proteins that are expressed at levels below 5 mgL (4-6). The questions that remain concern the reasons for such large differences and whether these reasons are in some way related to the insect cell line that is used. There are at least two compelling arguments that justify the investigation of cell lines with improved recombinant protein production capabilities. The first reason is practical: the cell line derived from Spodopteru frugiperdu, Sf9, has been the only cell line used in approx 95% of all From. Methods m Molecular Biology, Vol 39. Baculovlrus Expression Protocols EctM.i by: C D. Richardson Q 1995 Humana Press Inc., Totowa, NJ
385
386
Wickham
et al.
Table 1 Reported Expression Levels of Some Polyhedrin Promoter-Driven Protems in Sf9 Cells Recombinant protein Secretory Glycoprotein9 Tissue factor Tissue plasminogen activator Complement receptor Erythropoietin Interleukin Epidermal growth factor receptor Epstein-Barr virus gp350/220 Cytoplasmic Proteinsb Polyhedrin P-galactosidase Adenovirus penton base Chloramphenicol acetyltransferase
Expression level
Reference
OS-2 mg/L 2-3 mg/L 20.5-2 mg/L 10 mg/L 21-2 mg/L 1 mg/L 10 mg/L
(5,22) (814) (14) (11) (12) (6)
(4)
>500 mg/L x500 mg/L
>200 mg/L >lOO mg/L
aTranslocated, secretedprotems. bProtems made of free ribosomes in cytoplasm. CUnpublished observations
BEV studies to date. Most of the remaining studies have used the closely related cell line, Sf21, from which Sf9 is derived. Only recently have new cell lines been investigated. Second, the identification of limitations in the processing machinery of the Sf9 cell line (3,4) implies that other cell lines may be capable of superior production. In other expression systems, the choice of a mammalian cell line for the synthesis of certain proteins can be crucial to the levels of protein produced (7), which suggests that a similar situation may exist when using different insect cell lines in the BEV system. An additional reason to study recombinant protein expression from new cell lines pertains to posttranslational processing of proteins in insect cells. Sf9 cells generally synthesize N-linked glycoproteins that are lower in molecular weight than the same mammalian-produced proteins (7,8). Sf9 cells do not synthesize complex oligosaccharide structures in the expression of influenza virus hemagglutinin (9), although evidence exists that they are capable of adding complex sugars to some proteins (1Q11). Just as in mammalian cells, differences in the processing of N-linked oligosaccharides in insect cell
Different
Cell Lines for Expression
lines have been documented, indicating that other insect cell lines may be more efficient in modifying glycoproteins (12). Thus, there is strong incentive to identify insect cell lines that possess superior abilities to synthesize and process recombinant proteins. 1.2. Limitations ofsecreted
in the Production GZycoproteins
Secreted glycoproteins have represented a major proportion of recombinant proteins produced using the BEV system. Subunit vaccines for enveloped viruses and many pharmacologically important mammalian proteins are often glycosylated and secreted.Unfortunately, it is this class of protein that appears to be consistently produced at lower levels in Sf9 cells (Table 1). Recent evidence has accumulated relating to secretion and/or posttranslational processing. In a controlled study where three different proteins were inserted into the same transfer vector, the levels of recombinant protein production were much lower for secretedglycoproteins than proteins synthesized on free ribosomes in the cytoplasm (13). A number of other secreted glycoproteins were also produced at reduced levels compared to polyhedrin and other recombinant proteins translated in the cytosol (Table 1). Another study demonstrated that synthesis of tissue plasminogen activator (TPA) was dramatically increased when it was produced as a fusion protein that contained the first 15 amino acids of polyhedrin instead of the normal 35 amino acid signal peptide of TPA (2,14). One explanation for the increase is vector-related-the expression construct facilitated transcription/translation of TPA. However, an alternate or additional factor is that without the signal sequence,the fusion protein was translated in the cytoplasm and the rate-limiting steps involved in glycosylation and secretion were avoided. Limitations in the secretion and processing of glycoproteins in St9 cells during baculovirus infection have been identified. Rate-limiting steps have been observed during the secretion of papain from Sf9 cells (15), and also subsequent to translocation and core glycosylation of tissue plasminogen activator (14). Furthermore, the integrity of the secretion/processing machinery becomes compromised during stages of baculovirus infection, This breakdown in the processing machinery should be anticipated, since baculovirus causes a shutdown of host cell
protein synthesis shortly after infection.
Wickham et al. Other proteins may experience additional limitations during their production in insect cells. Even though Sf9 cells are capable of producing recombinant proteins at levels of 40% of the total cellular protein during infection, there is evidence suggesting that this does not represent an upper limit. Cell lines have been identified that can produce up to seven times more unmodified cytoplasmic proteins than Sf9 or Sf21 cells (16). Thus, Sf9 and Sf21 cells may yield suboptimal levels of transcription/ translation, as well as reduced secretion. 2. Comparison of Protein Expression in Other Insect Cell Lines The fact that insect cell lines differ in their ability to synthesize recombinant proteins has recently been demonstrated. A number of cell lines have been identified that surpass Sf9 cells in the production of recombinant proteins (12,16-19). King et al. (17) found that ah4umestra brussicae cell line (of unspecified tissue origin) could produce P-galactosidase and three different recombinant proteins at twice the level as Sf9 cells. Also, variations in the way recombinant glycoproteins were posttranslationally processed was reflected by molecular weight differences in the cell lines. Studies in our laboratory have compared seven cell lines to Sf9 cells with regard to the production of P-galactosidase and secretory alkaline phosphatase (18-20). These lines were originally derived from Spodopteru frugiperdu, Trichoplusiu ni, Mumestru brussicue, and Estigmene ucreu larvae. Five of the seven cell lines produced four to sevenfold more /3-galactosidasethan St-9cells on a per-cell basis (18) (Table 2). In several instances, quantities of recombinant protein synthesized approached 80% of the total cellular protein. The same five cell lines also produced 2.5 to 26-fold more secreted human placental alkaline phosphatase than Sf9 cells (16,19) (Table 2). In the above studies, the cell line Tn-SBl-4 was shown to surpass most cells for expression, and it is currently marketed by Invitrogen as “High 5” cells. These cells synthesize 7- and 26fold higher levels of both P-galactosidase and secreted alkaline phosphataseon a per-cell basis. The Tn-SB l-4 and MB0503 cell lines were able to process the high-mannose forms of glycoproteins much more efficiently (60-79% of intracellular protein was resistant to endo H) than to Sf9 and Sf21 cells (2144% of intracellular protein was resistant to endo H) (21). Insect cells contain glycosidase IA1 and mannosidase I activity, and are capable of processing the
Different
Cell Lines for Expression
389
Table 2 Production of P-Galactosidase and Secretory Alkaline Phosphatase (seAP) from Different Insect Cell Lines” (Optimal Cell Densities for Recombinant Protein Productionb)
Cell LinetiE Tn-SB 1-4 (Tn 5, High 5) Tn MG- 1 St21 Sf!l Ea 88 Tn 368
Mb0503 Tn Ap2
Protein/lo6 306 215 215 204 413 450 469 236
Cells P-Galactosidase IU/106cellsc tells/2.0 cm2b 1090 (5.0 x 104) 873 (5.0 x 104) 177 (1.0 x 105) 156 (3.0 x 105) 302 (1.0 x 105) 717 (1.0 x 105) 1155 (5.0 x 104) 678 (1.0 x 105)
Alk phosphatase IWO6 cell& tells/2.0 cm%
%seAP in Media 77
(l.E,,
81 (l.6::05) 74 (l.6:2105) 59 (6.0 k’los, 81 (7.9YlO4) 75 (1.9505) 85 (1.7%05) 65 (2.8 ‘;‘105,
aAdapated from Wickham et al. (18) and Davis et al. (20). %ell densities (cellsl2.0 cm2 well) at which maximum was reached are shown in parentheses. %J /3-galactosidase= 1 ug protein qU alkaline phosphatase= 1.4 p.g protein
high-mannose forms of glycoproteins to varying degrees,but all cell lines tested lack the terminal sialic acid and galactose residues found on complex carbohydrates (21). Tn-SBl-4 cells have also been shown to produce 6- and 28-fold higher levels of an Epstein-Barr virus envelope protein and a human soluble tissue factor (sTF), respectively, than Sf9 cells (22). Finally, when Tn-SBl-4 cells were compared to Sf9 cells in regard to the production of two cytoplasmic adenovirus structural proteins, Tn 5-5B1-4 cells synthesized at least three times more protein per cell than the Sf9 cells (Fig. 1). These proteins represented 60-80% of the total infected cell protein. In most cases this cell line appearsto be superior, but again the level of expression varies from protein to protein. The preceding studies demonstrate that many insect cell lines probably exist that possessbetter production and processing capabilities than Sf9 cells,
390
Wickham
et al.
12345
Fig. 1. Comparison between Tn-SBl-4 (lanes 3,5) and Sf9 (lanes 2,4) cell lines for the production of adenovirus2 pentonbase(lanes2,3) andfiber (lanes 4,5) structuralproteins.Samplesof 5 x lo4 cells were evaluatedby SDS-PAGE followed by staining with Coomasieblue. 3. Evaluation of Insect Cell Lines for Recombinant Protein Expression By carefully evaluating total production, protein localization (intracellular and extracellular), cell lysis, and cell density, a great deal of information can be obtained regarding the production and processing of the recombinant protein in the particular cell line being investigated (Table 3). Combinations of these factors can vary between cell lines and are important for optimizing protein production. Time-course studies help determine the optimal harvest time, which again varies with the cell line and the recombinant protein being produced. Localization of recombinant protein inside and outside the cells indicate how well a secreted protein is processed. However, appearance of a nonsecreted protein, such as P-galactosidase, in the extracellular media is directly associated with cell lysis during the late stages of infection. Efficiency of secretion is again cell-line-dependent. In the later stages of infection, cell lysis releases cellular proteins that complicate purification of the recombinant protein. For secreted glycoproteins, cell lysis results in the release of cell-associated proteases and glycosidases that subsequently degrade secreted homogeneous forms of the protein. Lysis also releases unprocessed forms of
Different
Cell Lines for Expression
391
Table 3 Important Factors in Determining Optimal Yield and Homogeneity of Secretory Proteins from Cell Lines Kinetics
Localizatron
Cell lysis
Cell density
Can vary between cell lines Used to determine optimal harvest time after accounting for losses in purity and homogeneity from cell lysis Varies between cell lines Used to determine efficiency of secretion Inefficient secretion may necessitate the use of another cell line Usually begins as production rate peaks Releases proteases and glycosidases, which can degrade secreted protein Unprocessed forms of protein released into media Contamination of recombinant product with cellular proteins Important in protein expression from attached cell lines Diminished expression apparently caused by cell-cell contact
the recombinant glycoprotein that are generally higher in molecular weight than the secreted form. Protease activity, glycosidase activity, and/or release of unprocessed glycoproteins after 3 d postinfection often cause a broadening of extracellular glycoprotein bands on SDS polyacrylamide gels. Cell density is also an important consideration when using attachmentdependent cell lines in the BEV system. Recombinant protein and nonoccluded virus production have been shown to depend on the surface density of cells during infection (16). This density effect appears to be the result of cell-cell contact. Contact-induced inhibition of recombinant protein production is a great problem in suspension culture as long as oxygen and nutrient levels are maintained. However, infections of cells in suspension should always be performed before growth approaches stationary phase. Other important characteristics of a cell line that influence protein production are its ability to withstand shearing forces, its doubling time, its adaptability to serum-free medium, and its susceptibility to infection. 4. Screening Methods for High-Producing Cell Lines Important factors to consider in screening cell lines are the percentage of susceptible cells, the density of cells, and the reporter protein. We have screened insect cell lines for recombinant protein production using
Wickham
et al.
recombinant baculoviruses that express nonsecreted E. coli p-galactosidase (18) and a secreted form of human placental alkaline phosphatase (19,20). The advantage of using recombinant enzymes for screening purposes is that a number of cell lines can be quickly and quantitatively screened using a calorimetric assay for functional activity. An additional advantage of using a secretory enzyme is that it allows the identification of cell lines that favor posttranslational modification and secretion of glycoproteins. Other quantitative calorimetric assays could also be used, such as ELISA, to screen for the production of a particular protein. However, ELISA is also not always a reliable measure of intracellular or secreted protein concentrations because of variability in cell lysis and protein degradation late in infection. Other methods of detecting protein levels, such as immunoprecipitation and Western blotting, are less quantitative. The cell density at the time of infection can have a dramatic effect on the amount of protein produced. Confluent cell monolayers in stationary phase produce much less recombinant protein than dividing cells undergoing exponential growth. Thus, cell densities lower than 1 x lo5 cells/cm2 should be used to screen cell lines so that cell density does not factor into the results. Not all the cells in an established cell line are always susceptible to infection. Thus, in screening insect cell lines for production, it is important to monitor the percentage of cells actually infected by AcMNPV. This can be demonstrated by monitoring the percentage of cells that develop cytopathic effect or polyhedra when challenged with wild-type AcMNPV. If a cell line contains a small percentage of hyperproducing cells, it should be possible to subclone a cell line from the population that is enriched for protein production. It should be noted that the percentage of susceptible cells can change following repeated passaging of the cell line, presumably because different populations of cells are selected during passaging. The cell line BTIEa 88 is reported to be 85% susceptible to AcNPV infection vs 19% susceptibility for the BTI Ea cell line from which the Ea 88 cell line was originally cloned. These cells were initially isolated from the hemocytes of Estigmene acrea larvae. In short, it is advisable to test routinely the proportion of cells that can be infected with recombinant virus.
Different
Cell Lines for Expression
393
5. Large-Scale Growth of Newly Developed Cell Lines Culture methods for the large-scale growth of insect cells in suspension culture are well established (21). Sf9, SfZl, Tn 368, and Mb0507 cells have all been reported to grow well and produce recombinant protein in volumes >l L. However, the growth of some cell lines appears to be anchorage-dependent, and they may not easily adapt to suspension culture. On the other hand, Sf9 cells can be directly inoculated from attached cultures into suspension cultures without any cell-clumping problems (23), whereas Tn-SB l-4 cells are not easily adaptable to growth in suspension culture. Tn-SB l-4 cells tend to form very large aggregates following inoculation into suspension culture, even in the presence of the mild detergent Pluronic 40 (22). Recombinant protein production from suspension cultures of Tn-SBl-4 cells is low owing to its anchorage dependency for growth or inhibition of virus infection through cellcell contract. High recombinant protein production by Tn-SBl-4 cells can be realized by growing them in adherent cultures. The density of the cells on the growth surface has been shown to be critical to obtaining optimal yields of recombinant protein (22). Infecting Tn-SB l-4 cells at a density of lo5 cells/cm2 was found to give maximal levels of recombinant protein production. The density of the cells at the time of infection can be easily approximated by measuring the doubling time of the cells and using this value to predict when the optimal cell density will be reached. Tn-5B l-4 cells, and presumably other insect cells that are anchoragedependent, can be grown using techniques commonly employed in the culture of anchorage-dependent mammalian cells (22,24). Roller bottles coated with DEAE microcarriers (24), suspension cultures of collagencoated microcarriers, and multistory polystyrene tissue-culture plates have been used to produce recombinant protein from Tn-5B l-4 cells at levels very close to the optimal levels produced in tissue culture. In addition, a packed bead bioreactor has been used to produce recombinant protein using these cells (25). It is likely that TN-Bl-4 growth is not strictly anchorage-dependent and that failure of the cells to grow directly in suspension culture simply results from their propensity to aggregate, Thus, these and other cells
Wickham et al. that demonstrate similar clumping problems could be adapted to suspension culture. However, in selecting suspension-competent cells, it is possible that the advantages of high recombinant protein levels could be lost. If some high-producing insect lines like Tn-SB 1-4 cells are, in fact, anchorage-dependent, there are numerous methods for the growth of these cells that have not yet been fully explored. References 1. Luckow, V. A. (1990) Cloning and expression of heterologous genes in insect cells with baculovirus vectors, in Recombinant DNA Technology and Applications (Ho, C., Prokop, A., and Bajpai, R., eds.), McGraw-Hill, New York. 2. Luckow, V. A. and Summers, M. D. (1989) High level expression of nonfused foreign genes with Autographa calijornica nuclear polyhedrosis virus expression vectors. Virology 170,3 l-39. 3. Matsuura, Y., Possee, R. D., and Bishop, D. H. L. (1987) Baculovirus expression vectors: The requirements for high level expression of proteins, including glycoproteins. J. Gen. Virol. 67, 1515-1529. 4. Moore, M. D., Cannon, M. J., Sewall, A., Finlayson, M., Okimoto, M., and Nemerow, G. R. (1991) Inhibition of Epstein-Barr virus infection in vitro and in vivo by soluble CR2 (CD21) containing two short consensus repeats. J. Viral. 65, 3559-3565. 5. Ruf, W., Miles, D. J., Rehumtulla, A., and Edgington, T. S. (1993) Mutational analysis of receptor and cofactor function of tissue factor. Methods Enzymol. 222,209-224. 6. Greenfield, C , Patel, G., Clark, S., Jones, N., and Waterfield, M. D. (1988) Expression of the human EGF receptor with ligand-stimulatable kinase activity in insect cells using a baculovirus vector. EMBO J. 7, 139-146. 7. Whang, Y., Silberklang, M., Morgan, A., Munshi, S., Lenny, A. B., Ellis, R. W., and Kieff, E. (1987) Expression of the Epstein-Barr virus gp350/220 gene in rodent and primate cells. J. Virol. 61, 1796-1807. 8. Steiner, H., Pohl, G., Gunne, H., Hellers, M., Elhammer, A., and Hansson, L. (1988) Human tissue-type plasmogen activator synthesized by using a baculovirus vector in insect cells compared with human plasminogen activator produced in mouse cells. Gene 73,449-457. 9. Kuroda, K., Geyer, H., Geyer, R., Doerfler, W., and Klenk, H.-D. (1990) The oligosaccharides of influenza virus hemaglutinin expressed in Insect cells by a baculovirus vector. Virology 174,4 18-429. 10. Davidson, D. J., Fraser, M. J., and Castellino, F. J. (1990) Oligosaccharide processing in the expression of human plasminogen cDNA by lepidopteran insect (Spodoptera frugiperda) cells. Biochemistry 29,5584-5590. 11. Wojchowslu, D. M., Lorkin, S H., and Sytkowski, A. J. (1987) Active human erythropoietin expressed in insect cells using a baculovirus vector: a role for Nlinked oligosaccharide. Biochim. Biophys. Acta 910,224-232.
Different
Cell Lines for Expression
395
12. Hink, W. F., Thomsen, D. R., Davidson, D. R., Meyer, A. L., and Castellino, F. J. (1991) Expression of three recombinant proteins using baculovirus vectors in 23 insect cell lines. Biotechnol. Prog. 7,9-14. 13. Luckow, V. A. and Summers, M. D. (1988) Signals important for high-level expression of foreign genes in Autographa culifornica nuclear polyhedrosis virus expression vectors. Virology 167,56-71. 14. Jarvis, D. L. and Summers, M. D (1989) Glycosylation and secretion of human tissue plasminogen activator in recombinant baculovirus-infected insect cells. Mol. Cell. Biol. 9,214-223.
15. Vernet, T., Tessier, D. C., Richardson, C., LalibertB, F , Khi, H. E , Bell, A. W., Storer, A. C., and Thomas, D. Y. (1990) Secretion of functional papain precursor from insect cells. 1. Biol . Chem. 265,16661-16666. 16. Wickham, T. J. (1991) Baculovirus-insect cell interactions in producing heterologous proteins: Attachment, infection, and expression in different cell lines. Ph. D. Thesis Dissertation, Cornell University. 17. King, L. A., Mann, S. G., Lawrie, A. M., and Mulshaw, S. H. (1991) Replication of wild-type and recombinant Autographa culifomicu nuclear polyhedrosis virus in a cell line derived from Mumestru brussicue. Virus Res. 19,93-104. 18. Wickham, T. J , Davis, T., Granados, R R., Shuler, M. L , and Wood, H. A (1992) Screening of insect cell lines for the production of recombinant proteins and mfectious virus in the baculovirus expression system. Biotechnol. Prog. 8,39 1-396. 19. Davis, T. R., Wickham, T. J., McKenna, K. A., Granados, R. R., Shuler, M. L., and Wood, H. A. (1993) Comparative recombinant protein production of eight insect cell lines. In Vitro Cell. Dev. Biol. 29A, 388-390. 20. Davis, T. R., Munkenbeck Trotter, K., Granados, R. R., and Wood, H. A. (1992). Baculovnus expression of alkaline phosphatase as a reporter gene for evaluation of production, glycosylation and secretion. BioITechnology 10, 1148-l 150. 21. Davis, T. R., Shuler, M. L., Granados, R. R., and Wood, H. A. (1993) Comparison of oligosaccharide processing among various insect cell lines expressing a secreted glycoprotein. In Vitro Cell. Dev. Biol. 29A:ll, 842-846 22. Wickham, T. J., and Nemerow, G. R. (1993) Optimization of growth methods and recombinant protein production in BTI Tn-SB l-4 insect cells using the baculovirus expression vector. Biotechnol. Prog. 9,25-30. 23. Wu, J., King, G., Daugulis, A. J., Faulkner, P., Bone, D. H., and Goosen, M. R. A. (1989) Engineering aspects of insect cell suspension culture: a review. Appl. Microbial.
Biotechnol. 32,249-255.
24. Lazar, A., Silberstein, L., Reuveny, S., and Mizrahi, A. (1987) Microcarriers as a culturing system of insect cells and insect viruses. Devel. Biol. Standard 64,3 15-323. 25. Shuler, M. L., Cho, T., Wickham, T. J., Ogonah, O., Kool, M., Hammer, D. A., Granados, R. R. and Wood, H. A. (1990) Bioreactor development for production of viral pesticides or heterologous proteins in insect cell cultures. Ann. NY Acud. Sci. 589,399-422.
CHAPTER23 Development of Lepidopteran Cell Lines Sardar
S. Sohi
1. Introduction Ever since the early reports of Goldschmidt (I) and Trager (2) on the in vitro cultivation of lepidopteran tissues, researchers have been interested in using lepidopteran cell cultures for investigations in insect pathology, genetics, and physiology. With the successful development of continuous cell lines by Gaw et al. (3) and Grace (4), interest in the use of lepidopteran cell cultures increased dramatically (5-9). As a result of this increased interest, over 150 lepidopteran continuous cell lines were reported by 1991 (IO-14), and several additional cell lines have been developed from lepidoptera since then (15-17, Sohi unpublished data). With the recent advances in biotechnology and the development of baculovirus expression vectors, lepidopteran cell cultures have become invaluable for basic biological and biotechnological research (18,19). Although over 150 lepidopteran cell lines are currently available, there is a need for developing new cell lines that will not only grow better and faster, but also replicate baculoviruses better (14,20). Moreover, the production of different recombinant proteins with the baculovirus expression vector systems, such as the Autographa californica system, varies tremendously among different lepidopteran cell lines (21,22). Thus, there is an increasing need for developing new lepidopteran cell lines to meet the different requirements. In our laboratory, we have developed 69 continuous cell lines from various tissues of several lepidopterans (23-28, Table l), and several of these lines replicate baculoviruses (29-33, Sohi From: Methods w Molecular B/ology, Vol. 39: Baculovirus Expression Protocols Edited by: C. D Richardson 0 1995 Humana Press Inc., Totowa, NJ
397
Sohi
398 Table 1 Lepidopteran Cell Lines Developed in the Author’s Laboratory at the Forest Pest Management Institute, Canadian Forest Service Isect speciesa C. fumijerana C. occidentalis L. dispar A4. disstria M. sexta 0. leucostigma
Total
Neonate larvae
Embryos
Ovaries
Hemocytes
Midgut
lib
1
10
0
13 2 0 4 7 37
3 1 0 0 0 5
5
0
4 0
1
0
0
3 0 0 19
4 0 0 4
0 0 0 4
Total 26 21 4 7 4 7 69
T. fumlferana = Chorrstoneurafimrferana, C. occrdentalrs = Chorrstoneura occrdentalis, L dispar = Lymantrra dispar, M. drsstria = Malacosoma disstrra, IU. sexta = Manduca sexta, 0. leucostigma = Orgyia leucostigma.
bEach figure represents the number of continuous cell lines developed from the respective tissues of each insect species.
unpublished data). Of all the lepidopteran cell lines reported world wide, the majority have been developed from embryos, ovaries (adult, pupae, larvae), hemocytes and neonate larvae (14). Although in our (Forest Pest Management Institute) laboratory we (Insect Tissue Culture Group) also have developed most of the lepidopteran cell lines from these four tissues, the largest number was from neonate larvae (Table 1). Detailed procedures have been published for developing cell lines from lepidopteran ovaries (34) and lepidopteran embryos (35). Although several continuous lepidopteran cell lines have been developed from hemocytes (14), melanization is a serious problem encountered in culturing hemocytes (35). Although we were successful in developing continuous cell lines from the hemocytes of Mulacosoma disstria (23,24), we were unable to establish cell lines from other lepidopterans because of the toxic effects of melanization. Because of this problem, tissues other than hemocytes have been preferred for initiating lepidopteran cell lines. Since we were successful in developing many lepidopteran cell lines using neonate larvae as the explant source (Table l), I have described in this chapter a method to develop continuous cell lines from neonate larvae of lepidoptera. The procedures and protocols described here are the ones used in my laboratory. Please refer to Goodwin and McCawley (34) and Lynn (35) for procedures and “tricks of the trade” used in some of
Lepidopteran
Cell Line Development
399
the other laboratories, and see Vaughn (20) for a current review of literature on lepidopteran cell culture. At present, several insect tissue-culture media are available as mentioned in Chapters 3 and 4 of this volume. Most of the lepidopteran tissue-culture media are based on Grace’s medium (4) which is a modification of Wyatt’s medium (36) formulated in our institute, then known as the Insect Pathology Research Institute. We routinely use Grace’s medium (4), supplemented with 0.25% (w/v) tryptose broth and lO-15% fetal bovine serum (v/v). 2. Materials 2.1. Insects
Eggs of the insect speciesof interest are used that are close to hatching. 2.2. Reagents 1. Grace’s insect tissue-culture medium (4) (GIBCO-BRL Life Technologies, Inc., Grand Island, NY). 2. Fetal bovine serum (heat-inactivated at 56°C for 30 min) (CANSERA, Rexdale, ON, Canada). 3. Bovine serum albumin (Sigma Chemical Co., St. Louis, MO). 4. Tryptose broth (DIFCO Laboratories, Detroit. MI). 5. 2.5 g/L Trypsin (GIBCO-BRL Life Technologies, Inc.). 6. Gentamicin antibacterial reagent solution (50 mg/mL) (GIBCO-BRL Life Technologies, Inc.). 7. Rinaldini balanced salt solution (37) (GIBCO-BRL Life Technologies, Inc.). 8. Ethyl alcohol, 70%. 9. Sodium hydroxide pellets. 10. Glass-distilled water. 11. Javex commercial bleach. 12. Trypan blue: 0.4% solution in 0.85% saline (GIBCO-BRL Life Technologies, Inc.). 2.3. Equipment and Glassware 1. Laminar flow hood (The Baker Company, Sanford, ME). 2. Autoclave (AMSCO Canada, Division of Ingram and Bell Ltd.,
Mississauga,ON, Canada). 3. Oven, sterilizing (400”) (Despatch Oven Co., Minneapolis, MN). 4. Glass still fed by deionized water (Coming Glass Works, Corning, NY).
5. Clinical centrifuge.
Sohi 6. Incubator (2%30°C range; CO2 and humidity not necessary) (CONVIRON Systems, Winnipeg, MB, Canada). 7. Osmometer, freezing point (Advanced Instruments Inc., Needham Heights, MA). 8. pH meter. 9. Pressure filtration unit for liquids (Cat. # YY 30-142-05 Millipore Corp., Bedford, MA). 10. Compressed nitrogen gas for use with the pressure filtration unit. 11. Suction-type disposable filtration units, assorted (0.22 and 0.45 pm pore, Becton Dickinson Canada Inc., Mississauga, ON, Canada). 12. Magnetic stirrer. 13. Stir bars, TeflonTM-coated. 14. Microscope, Dissecting (Objectives: 1,2, and 4X; eyepieces: 10X; American Optical, Buffalo, NY). 15. Microscope, inverted (Diaphot; phase-contrast objectives: 10,20, and 40X; eyepieces: 10X, Nikon Canada Instruments, Mississauga, ON, Canada). 16. Microscope, research (Optiphot; phase-contrast Objectives: 20, 40, and 100X; bright-field Objectives: 40 and 100X; eyepieces: 10X, Nikon Canada Instruments). 17. Alcohol lamp/Bunsen burner. 18. Cell counter, electronic (Coulter Electronics, Inc., Hialeah, FL). 19. Cell counter, hemocytometer (American Optical). 20. Pipeter, mechanical, such as pipet aid (Drummond Scientific Co., Broomall, PA). 21. Beakers, assorted (50-, loo-, . . . 1000~mL). 22. Erlenmeyer flasks, assorted (50-, lOO-, . . . 1000~mL). 23. Cavity slides (75 x 45 x 7 mm with 36 x 5 mm concavity) (Cat. # 3731, Becton Dickinson Canada Inc.). 24. Glass Petri dishes, lo-cm diameter, to hold (a) filter paper for hatching insect eggs, and (b) cavity slides for mincing neonate larvae (Coming Glass Works). 25. Filter paper (Baxter Diagnostic Corp., Mississauga, ON, Canada). 26. Forceps, microdissecting (Baxter Diagnostic Corp.). 27. Knife, microdissecting (IREX Surgical Instruments, Toronto, ON, Canada). 28. Scissors,microdissecting such asiris scissors(IREX Surgical Instruments). 29. Centrifuge tubes, conical (assorted lo-, 15, and 20-mL). 30. Pasteur pipets. 31. Serological pipets. 32. Syringe, tuberculin (l-cc). 33. Tissue-culture plastic flasks, 25-cm2.
Lepidopteran
Cell Line Development
401
3. Methods 3.1. Work Station Microbial contamination has been one of the greatest threats for growing cells in culture, especially before the availability of antibiotics. A dust-free work area away from laboratory traffic is essential for reducing microbial contaminations in cell cultures. A laminar flow hood is ideal for this work. The laminar flow hood should be located in a less-traveled corner of the laboratory, preferably in a small room that has: 1. 2. 3. 4. 5.
Gas; Vacuum; Nitrogen cylinder; A bench-top clinical centrifuge; and Cupboards for storing items, such as sterile pipets, tissue-culture flasks, centrifuge tubes, and rubber policemen.
This arrangement will keep the traffic in the culture room to a rninimum and will reduce the incidence of microbial contaminations in cell cultures, which is of paramount importance.
3.2. Tissue Culture Medium and Other Solutions 3.2.1. Tissue-Culture Medium 1. Use deionized and glass-distilled water for preparing the culture medium and all other solutions used for cell culture. 2. Prepare or buy Grace’s insect tissue-culture medium (4), For small quantities, the liquid medium is recommended. For larger quantities of the medium, it is more economical to buy it in the powder form or to prepare it using the various ingredients according to the formulation (4). However, it is recommended that those inexperienced in preparing complex culture media purchase the liquid medium from any of the tissue-culture/biotechnology suppliers, such as GIBCO/BRL (see Note 1). 3. Add 0.25% (w/v) tryptose broth to the medium, and mix it thoroughly using a magnetic stirrer and a Teflon-coated stir bar. 4. Most lepidopteran cells grow best in the pH range of 6.0-6.5, and at an osmolality of 300-400 mosM/kg (38). Adjust the pH of the medium to 6.2 using 1.ONNaOH, since the medium is generally on the acidic side when it is prepared. If the pH is higher than 6.2, lower it with l.ON HCl. If the osmolality is in the 330-370 mosM/kg range, it can be used without any adjustment. If, however, it is outside this range, adjust it to 350 mosM/kg by either adding distilled water to lower the osmolality or using 15% NaCl
5. 6. 7. 8.
or 20% mannitol to raise it. Check the osmolality with a freezing point depression osmometer. Filter-sterilize the medium using a 0.22 pm pore filter with a vacuumoperated disposable plastic filtration unit for small volumes and a pressure filtration unit for larger volumes with nitrogen providing the pressure. Incubate the medium at 28°C for 5-7 d to detect contamination, The medium will turn turbid if it is not sterile. Also, test selected bottles of the medium for sterility by setting up cell cultures with it. Store the medium at 4OC in the dark; under these conditions, it is good for 6 mo. Dispense sterile fetal bovine serum (FBS) aseptically into sterile test tubes (5-10 mL/ tube). Heat-treat it at 56°C for 30 min to inactivate toxic factors, and store it in a domestic
freezer (-2O’C).
Test heat-inactivated
FBS
for toxicity and sterility in some cell cultures. 9. Thaw one of the FBS tubes by placing in cold water so that the upper onethird of the tube stays out of water to ensure sterility of contents. 10. Add 10% FBS and gentamicin (50 pg/mL) to Grace’s medium, and store it at 4OC. This supplemented medium will be referred to as the growth medium hereafter. Discontinue the use of gentamicin in the growth medium after cells have been subcultured a few times or, at the latest, after some cells have been freeze-preserved. Prolonged use of antibiotics in many cell cultures is not desirable (39). 3.2.2. Trypsin Solution Prepare a 0.25 and a 0.05% dilution of trypsin (2.5 g/L stock solution) in Rinaldini balanced salt solution (37). Sterilize the trypsin solutions by filtering them through a 0.22~pm membrane filter. Dispense them in test tubes (5-10 mutube), and store in a domestic freezer (-20°C) (see Note 2). 3.3. Disinfection of Eggs 1. Remove eggs from any oviposition material adhering to them. For instance, if eggs are laid m a mass, as in the case of the spruce budworm, Choristoneuru fumiferana, treat the egg masseswith 1% sodium hydroxide (NaOH) for 10 min at 35OC and rinse them with 1% bovine serum albumin (BSA) (40). The BSA rinse protects the separated eggs from subsequent damage by dehydration (see Note 3). 2. In the case of eggs that are covered with hair, as in gypsy moth, Lymuntria dispar, remove the hair by applying vacuum suction over the egg masses or by rubbing the egg masseswith silica sand. If the silica sand is used, separate the eggs from sand and broken hair by sieving them through no. 20 mesh screening.
Lepidopteran
Cell Line Development
403
3. Immerse the separated eggs in 70% ethyl alcohol for 5-10 min, and then rinse them with sterile distilled water three times. 4. Transfer the disinfected eggs to a sterile Petri dish that has been lined with sterile filter paper, and incubate them at room temperature (21OC) for hatching.
3.4. Primary
Cultures
1. Wipe the laminar flow hood with 70% ethyl alcohol, and set up the following in it: a. A dissecting binocular microscope; b. An alcohol lamp or a bunsen burner; c. A beaker with 70% ethyl alcohol and cotton wool in it (alcohol to disinfect instruments and cotton wool to protect their tips when they are placed in the alcohol); and d. A small plastic box open in the front with a raised lip in the opening for holding the disinfected dissecting instruments to allow them to cool. 2. Immerse the blades of iris dissecting scissors and forceps in 70% alcohol for 5-10 min, and then pass them through the flame to ignite the alcohol to disinfect them but do not hold them in the flame too long, since this dulls the cutting edges. After the alcohol is burnt off, place them in the small plastic box on the ledge for cooling so that the blades do not touch the bottom or top of the box. 3. Thaw out a tube each of 0.25% trypsin and FBS, and take out a bottle of growth medium so that they warm up to room temperature. 4. Collect larvae hatched from disinfected eggs during the 24-h period, and transfer 50-200 of them (depending on their size) to a sterile cavity slide containing 1.0 mL of 0.25% trypsin. The number of newly hatched larvae used per culture depends on their size, since they are much bigger in some speciesthan others. It is important to have sufficient tissue material to start a 25-cm2 flask culture. In the case of the spruce budworm, we use 200 larvae for a flask (see Note 4). 5. While examining with the binocular dissecting microscope, cut the larvae into as many small pieces as possible. 6. Transfer the minced larvae into a 10-n& centrifuge tube, and add another 4.0 mL trypsin. 7. Incubate the tissues at 37°C for 10 min. Then add 1.0 mL FBS to stop further action of the trypsin, and mix vigorously with a Pasteur pipet. 8. Centrifuge in a clinical centrifuge at a setting of 3 (135g) for 5 min. Discard the supernatant, resuspend the pellet in 3.0 mL growth medium, and transfer to a 25-cm2 plastic tissue-culture flask. Tip the flask back and forth, and sideways to spread the tissue fragments evenly all over the
9.
10. 11, 12.
13.
bottom of the flask and incubate at 28OC. It does not require CO2 atmosphere or high humidity. Label each culture flask with a code that, by convention, has the initials of your laboratory, the initials of the genus and species of the insect, the first letter of the tissue explanted, and the number of the primary culture or the number of your experiment. All letters used in the code should be Upper Case to avoid confusion with numerals. As an example, the code FPMIMS-L25 will be used for culture number 25 started with the neonate larval (L) tissues of the tobacco hornworm, Munducu sexta (MS), at the Forest Pest Management Institute (FPMI). Examine the cultures from time to time for cell growth and microbial contamination, If a culture is contaminated, it is better to start a new one instead of trying to salvage it. Add 1.OmL growth medium to the culture flask after a week and another 1.Onil, after the secondweek to bring the medium to atotal volume of 5.0 mL. Subsequently, replace 1.5-2.0 mL spent medium in the flask with an equal volume of fresh growth medium weekly. If all the tissues and cells are attached to the substrate, simply remove the spent medium, and add the fresh growth medium. If there are free-floating cells and tissues, centrifuge the spent medium at 135g. Discard the supernatant, resuspend the pellet in the fresh medium to be used, and return it to the culture flask (see Note 5). When there are signs of cell growth, the volume of the spent medium to be replaced weekly can be increased gradually to 3.0.-4.0 ml/flask. Do not replace the entire spent medium in a primary culture with fresh growth medium, because the spent medium has essential growth factors and metabolites produced by the cells in culture (see Note 1). 3.5. Subculturing
and Development
of Cell Lines
1. When a culture is 80-90% confluent, which could take from several weeks to several months, make the first subculture. If the cells are free-floating, suspend the cells thoroughly with a pipet. Transfer 2.5 mL cell suspension to a new culture flask, and leave the remaining cell suspension in the old flask. Add 2.5 mL fresh growth medium to each culture. Using the example of the culture designated as FPMI-MS-L25 in Section 3.4., step 9 above, the first subculture should be identified as lFPMI-MS-L25. The numeral before FPMI indicates the number of times the cells have been passaged in vitro (see Notes 5 and 6). 2. If the cells are attached to the substrate, scrape them with a rubber policeman, and make the first subculture as described in step 1 above.
Lepidopteran
Cell Line Development
405
3. If the cells are so strongly attached to the substrate that scraping with a rubber policeman does not dislodge them, use the trypsinization procedure to loosen them. Thaw out a test tube of 0.05% trypsin, and warm the trypsin solution to room temperature. Remove the spent medium from the culture, and save it in a sterile test tube. Add sufficient fresh growth medium to the spent medium to make 10.0 mL, and use it as conditioned medium. Rinse the monolayer of cells with 3.0-4.0 mL of this trypsin solution, and discard the trypsin. Add 4.0 mL fresh trypsin and incubate at room temperature for 2.0-3.0 min. Examine the culture with a microscope to see if the cells have been loosened. If they are loosened, remove trypsin from the culture flask, and add 5.0 mL conditioned medium. Scrape the monolayer of cells with a rubber policeman, and thoroughly resuspend the cells by vigorous pipeting. Transfer 2.5 mL cell suspension to a new flask, and leave the remainder in the old flask. Add 2.5 mL conditioned medium to each flask (seeNotes 1 and 2). 4. When the first subculture becomes 80-90% confluent, which could take several weeks, make a second subculture by making a two for one split as in steps l-3 above (see Note 6). 5. Continue making 2:l splits until the new subcultures can be subcultured weekly. From that point on, start making 3: 1 splits until subcultures can be subcultured weekly again (see Note 7). 6. As the rate of growth of subcultures increases, increase the number of daughter cultures that you make from a culture (see Note 8). 7. Subsequently, you can count the cell number using an electronic cell counter or a hemocytometer. Also, at this point, you can check the viability of cells using the dye exclusion test. For this test, add 0.1 mL of 0.4% trypan blue solution to 0.9 mL cell suspension. Let stand at room temperature for 10 min, mix with Pasteur pipet, and load a sample in a hemocytometer. Let the cells settle for a minute, and count 100-200 cells while keeping a separate count of stained (dead) and unstained (living) cells. From this differential count of dead and live (viable) cells, the percent of viable cells in a culture can be calculated (see Notes 7 and 8). 8. Subcultures can now be set up with a known number of cells, and their actual growth determined over a period of time by counting the cell number at the time of next subculture. Also, one can establish a growth curve for the cells, and determine their population doubling time by setting up multiple cultures and counting the cell number in 2-3 cultures every 24 h for 7-10 d. 9. According to convention accepted by the Tissue Culture Association, a cell line is considered a continuous or established cell line when it has been subcultured 50 times (see Notes 7-9).
406
Sohi 4. Notes
1. When establishing a new cell line, use the tissue culture medium that has been successfully used for culturing cells of the insect species in which you are interested. Failing that, use the medium that has been used for culturing cells of a closely related species. It might also be helpful to use conditioned medium to initiate new cell lines. To condition the medium, use an existing cell line of the same species. If no such cell line is available, use a line from a closely related species. Let the cells grow for 2 d in their usual growth medium. Then remove and discard the spent medium, and add to the culture an equal volume of the growth medium that you want to condition. Incubate the culture at 28T as usual for 24 h, and then remove the medium. Use centrifugation to remove the medium if the cells are free-floating. Filter the conditioned medium through a 0.45pm membrane filter to make sure that there are no cells of the existing cell line in it, and store at 4OC. Add 3-5% FBS to the conditioned medium when using it for your primary cultures. 2. If trypsin treatment does not suit your cell cultures, try collagenase or hyaluronidase as described by Goodwin and McCawley (34). 3. Some eggs may be sensitive to the harsh treatment with 1% NaOH. In such a case, treat the eggs with a commercial bleach such as Javex or Chlorox, diluted to a final concentration of 0.5% NaOCl, and 1.O% triton X-100 for 5-10 min according to Lynn (35), instead of NaOH. 4. For the successful development of a cell line, it is critical to have as large a quantity of tissue material as possible for any given volume of medium, because the tissues explanted in vitro modify and condition the medium with their metabolites. The higher their quantity, within reasonable limits, the better the chances of cell growth in vitro. If you have only a limited quantity of tissue material, start cultures by explanting the tissue fragments in a standing drop in a 35-mm plastic tissue-culture Petri dish as described by Lynn (35). 5. Lepidopteran cells explanted in vitro can take weeks and months before they start to grow. Be patient; do not try to subculture them too soon or too frequently, and do not terminate the cultures too soon. Grace (41) reported that in his first success in developing a lepidopteran cell line, he subjected his cultures to “organized neglect.” He examined the cultures about every 2 d, and if muscle contractions were regular and there were some healthy cells, half the medium was changed. It took him over a year to obtain cell growth.
Lepidopteran
Cell Line Development
407
6. Embryonic and neonate larval tissue cultures have a large number of cell types originating from the different tissues of the embryos and neonate larvae explanted. Lynn (35) reported colonies of four types of cells in the embryonic cultures of the diamondback moth, Plutelh xylostellu. In our work also, we have observed many cell types in the primary cultures and early subcultures of neonate larval tissue cultures. Colonies of four types of cells observed in the neonate tissue cultures of the tobacco hornworm, Munduca sextu, are illustrated in Figs. l-4. As Lynn (35) pointed out, it should be possible to isolate many cell strains from a small number of such primary cultures. 7. As a cell line starts to grow, preserve some of the cells in deep-freeze from time to time. For instance, store cells from the Sth, lOth, 20th, 35th, and 50th subcultures. By doing this, you will be assured of having cells from some of the earlier subcultures in storage should you happen to lose the current cultures of your cell line through contamination or equipment failure. The cryopreservation procedures are described in Chapters 3 and 4. A simple improvised inexpensive freezing device, like the one reported by Sohi et al. (42), can be used if automated controlled-rate freezing equipment is not available. 8. As the cultures begin to grow fairly well, they can be tested for baculovirus replication. Many of our lepidopteran cell lines developed from different tissues listed in Table 1 replicate baculoviruses. For instance, our IPRI-OL- 12 cells developed from neonate larvae of the white-marked tussock moth, Orgyiu Zeucostigmu, replicate a singly enveloped baculovirus of 0. Zeucostigmu (33), and a multicapsid baculovirus of the Douglas-fir tussock moth, Orgyiu pseudotsugutu (32). Healthy control cells are shown in Fig. 5, and cells infected with the singly enveloped baculovirus of 0. Zeucosigmu in Fig. 6. Hypertrophied nuclei packed with polyhedral inclusion bodies are shown in this figure. 9. After a cell line has been developed, it is a good practice to check its species identity by at least two of the methods used for cell line characterization, because cell lines can become contaminated with cells from other cell lines maintained in your laboratory (2843). If a few cells from an established cell line are accidently introduced into a primary culture, they crowd out the slow-growing primary culture because of their faster growth rate, and eventually completely replace it. To characterize our cell lines, we have used karyological(26), serological (27), and isozyme (28) methods of species identification. See Vaughn (20) for other methods of cell line characterization and for more information on the above three methods we use.
408
Sohi
Figs. l-4. Phase-contrastmicrographs of Manduca sex& cells from neonate larval tissues in early stages of in vitro cultivation. (Fig. 1) A colony of epithelial-like cells in an U-d-old culture of FPMI-MS4 cell line in the second passage in vitro. (Fig. 2) A network of dendrite-like cells in a 103-d-old primary culture. This culture later developed into a cell line designated as FPMI-MS-5, but the dendrite-like cells were crowded out and replaced by other faster-growing cells. (Fig. 3) A field of round cells from a 18%d-old primary culture that developed into a continuous line FPMI-MS-7. (Fig. 4) A colony of long fibroblast-like cells in a 30-d-old culture of FPMI-MS-12 cell line in the second passage in vitro.
Lepidopteran
Cell Line Development
Figs. 5 and 6. (Fig. 5) Phase-contrast micrographs of FPMI-OL- 12 cells from neonate larval tissues of the white-marked tussock moth, Orgyia Zeucostigma. A 3-d-old culture of healthy control cells in 90th subculture. IZ = nucleus. (Fig. 6) Phase-contrast micrographs of FPMI-OL-12 cells from neonate larval tissues of the white-marked tussock moth, Orgyia Zeucostigma. Cells infected with a singly enveloped baculovirus of 0. Zeucostigma. Note hypertrophied nuclei (n) packed with polyhedral inclusion bodies (p), some of which are fractured (pf) (Reprinted from ref. 33 with permission).
410 References 1. Goldschmidt, R. (1915) Some experiments on spermatogenesis in vitro Proc. N&l. Acad. Sci. USA 1,220-222.
2. Trager, W. (1935) Cultivation of the virus of grasserie in silkworm tissue cultures. J. Exptl. Med. 61,501-513.
3. Gaw, Z. Y., Liu, N. T., and Zia, T. U. (1959) Tissue culture methods for the cultivation of virus grasserie. Actu. Virol. (Prague) 3 (Suppl.), 55-60. 4. Grace, T. D. C. (1962) Establishment of four strains of cells from insect tissues grown in vitro. Nature 195, 788-789. 5. Grace, T. D. C. (1969) Insect tissue culture and its use in virus research Adv. Virus Rex 14,201-220.
6. Granados, R. R. (1976) Infection and replication of insect pathogenic viruses in tissue culture. Adv. Virus Res. 20, 189-236. 7 Vaughn, J. L and Dougherty, E. M. (1981) Recent progress in vitro studies of baculoviruses, in Biological Control in Crop Production, BARC Symposium Number 5, (Papavizas, G. C., ed.), Allanheld, Osmun, Totowa, pp. 249-258. 8. Ganados, R. R. and Hashimoto, Y. (1989) Infectivity of baculoviruses to cultured cells, in Invertebrate Cell System Applications, vol II, (Mitsuhashi, J., ed.), CRC, Boca Raton, FL, pp. 3-13. 9. Goodman, C. L. and McIntosh, A. H. (1994) Production of baculoviruses for insect control using cell culture, in Insect Cell Biotechnology (Maramorosch, K. and McIntosh, A. H., eds.), CRC, Boca Raton, FL, in press. 10. Hink, W. F. (1972) A catalog of invertebrate cell lines, in Invertebrate Tissue Culture, vol. II, (Vago, C., ed.), Academic, New York, pp. 363-387. 11 Hink, W. F. (1976) A compilation of invertebrate cell lines and culture media, in Invertebrate Tissue Culture Research Applications (Maramorosch, K., ed.), Academic, New York, pp. 3 19-369. 12. Hink, W. F. (1980) The 1979 compilation of invertebrate cell lines and culture media, in Invertebrate Systems In Vitro (Kurstak, E., Maramorosch, K., and Dubendorfer, A., eds.), Elsevier/North Holland, Amsterdam, pp, 553-578. 13. Hink, W. F. and Hall, R. L (1989) Recently established invertebrate cell lines, in Invertebrate Cell System Applications, vol. II, (Mitsuhashi, J., ed.) CRC, Boca Raton, FL, pp. 269-293. 14. Lynn, D. E. (1991) Establishing invertebrate cells in culture: Continued need for new cell lines, in Proceedings of the Eighth International Conference on Znvertebrute and Fish Tissue Culture (Fraser, M. J., Jr., ed.), Tissue Culture Association, Columbia, MD., pp. l-6. 15. Kouassi, K. N., Lery, X, Fediere, G., and Herder, S. (1992) A new permissive cell culture obtained from Latoiu viridissima (Lepidoptera, Limacodidae). J. Znvertebr Puthol. 59,112-l 13. 16. Lee, S.-H. and Hou, R. F. (1992) Establishment of a cell line derived from embryos of the diamondback moth, Plutella xylostella (L.). J. Znvertebr. Puthol. 59, 174-177. 17. Hara, K., Funakoshi, M., Tsuda, K., and Kawarabata, T. (1993) New Spodopteru exiguu cell lines susceptible to Spodopteru exiguu nuclear polyhedrosis virus. In Vitro Cell. Dev. Biol. 29A, 904-907.
Lepidopteran
Cell Line Development
411
18. Luckow, V. A. and Summers, M. D. (1988) Trends in development of baculovirus expression vectors. Biotechnology 6,47-55. 19. O’Reilly, D. R., Miller, L, and Luckow, V. A. (1992) Buculovirus Expression Vectors: A Laboratory Manual. W. H. Freeman, New York. 20. Vaughn, J. L. (1994) Lepidopteran cell cultures, in Arthropod Cell Culture Systems (Maramorosch, K. and McIntosh, A. H., eds.), CRC, Boca Raton, FL, in press. 21. Hink, W. F., Thomsen, D. R., Davidson, D. J., Meyer, A. L, and Castellino, F. J. (1991) Expression of three recombinant proteins using baculovirus vectors in 23 insect cell lines. Biotechnol. Prog. 7,9-14. 22. Davis, T. R., Wickham, T. J., McKenna, K. A., Granados, R. R., Shuller, M. L., and Wood, H. A. (1993) Comparative recombinant protein production of eight insect cell lines. In Vitro Cell. Dev. Biol. 29A, 388-390. 23. Sohi, S. S. (197 1) In vitro cultivation of hemocytes of Malacosomu dtsstria Hubner (Lepidoptera: Lasiocampidae). Can. J. Zool. 49, 1355-1358. 24. Sohi, S. S. (1973) Establishment of cultures of Mulacosoma disstriu Hubner (Lepidoptera: Lasiocampidae) hemocytes in a hemolymph-free medium, in Proceedings of the 3rd International Colloq. Invertebrate Tissue Culture, Bratislava, Czechoslovakia (1971) (Rehacek, J., Blaskovic, D., and Hink, W. F., eds.), Slovak Acad-
emy of Sciences, Bratislava, pp. 27-39. 25. Sohi, S. S. (1973) In vitro cultivation of larval tissues of Choristoneurufimiferana (Clemens) (Lepidoptera: Tortricidae), in Proceedings of the 3rd International Colloquim of Invertebrate Tissue Culture, Bratislava, Czechoslovakia (197 1) (Rehacek, J., Blaskovic, D., and Hink, W. F., eds.), Slovak Academy of Sciences, Bratislava, pp. 75-92. 26. Ennis, T. J. and Sohi, S. S. (1976) Chromosomal characterization of five lepidopteran cell lines of Malacosoma disstria (Lasiocampidae) and Choristoneura fumiferana (Tortricidae). Can. J. Genet. Cytol. l&471-477. 27. Krywienczyk, J. and Sohi, S. S. (1976) Serologic characterization and identification of four lepidopteran cell lines. Can. J. Zool. 54,1559-1564. 28. Harvey, G. T. and Sohi, S. S. (1985) Isozyme characterization of 28 cell lines from five insect species. Can. J. Zool. 63,2270-2276. 29. Sohi, S. S. and Cunningham, J. C. (1972) Replication of a nuclear polyhedrosis virus in serially transferred insect hemocyte cultures. J. Znvertebr. Pathol. 19,5 1-61. 30. Sohi, S. S. and Bird, F. T. (1976) Replication of a nuclear polyhedrosis virus of Choristoneura fumiferana (Lepidoptera: Tortricidae) in Malacosoma disstria Hubner (Lepidoptera: Lasiocampidae) hemocyte cultures, in Znvertebrate Tissue Culture: Applications in Medicine, Biology, and Agriculture (Kurstak, E. and Maramorosch, K., eds.), Academic, London, pp. 361-367. 31. Arif, B. M., Sohi, S. S., and Krywienczyk, J. (1976) Replication of alkali-released NPV in a cell line of Choristoneura fumijerana, in Proceedings of the 9th Annual Meeting of the Society for Invertebrate Pathology, Kingston, Canada, Printing Department, Queen’s University, Kingston, pp. 108-l 12. 32. Sohi, S. S., Percy, J., Cunningham, J. C., and Arif, B. M. (1981) Replication and serial passage of a multicapsid nuclear polyhedrosis virus of Orgyia pseudotsuguta (Lepidoptera: Lymantriidae) in continuous insect cell lines. Can. J Microbial. 27,1133-l 139.
412 33. Sohi, S. S., Percy, J., Arif, B. M., and Cunningham, J. C (1984) Replication and serial passage of a singly enveloped baculovirus of Orgyia leucostigma in homologous cell lines. Intervirology 21,50-60. 34. Goodwin, R. H. and McCawley, P. E. (1977) Initiating attached cell lines from the lepidoptera (Insecta). Tissue Culture Assoc. (TCA) Manual 3,567-569. 35. Lynn, D. E. (1989) Methods for the development of cell lines from insects. J. Tissue Culture Methods 12,23-29. 36. Wyatt, S. S. (1956) Culture in vitro of tissue from the silkworm Bombyx mori L. J. Gen. Physiol. 39,841-852.
37. Rinaldini, L. M. (1959) An improved method for the isolation and quantitative cultivation of embryonic cells. Exp. Cell Res. 16,477-505. 38. Sohi, S. S. (1980) The effect of pH and osmotic pressure on the growth and survival of three lepidopteran cell lines, in Invertebrate Systems In Vitro (Kurstak, E., Maramorosch, K., and Dubendorfer, A., eds.), Elsevier/North Holland, Amsterdam, pp. 35-43. 39. Coriell, L. L. (1973) Methods of prevention of bacterial, fungal, and other contaminations, in Contamination in Tissue Culture (Fogh, J. ed.), Academic, New York, pp. 29-49. 40. Retnakaran, A. and French, J. (1971) A method for separating and surface sterilizing the eggs of the spruce budworm, Choristoneura fumiferana, (Lepidoptera: Tortricidae). Can. Entomol. 10,7 12-7 16. 41. Grace, T. D. C. (1982) Development of insect cell culture, in Invertebrate Cell Culture Applications (Maramorosch, K. and Mitsuhashi, J., eds.), Academic, New York, pp. l-8. 42. Sohi, S. S., Sullivan, C. R., and Bodley, C. L. (1971) Simple controlled-rate freezing device. Lab. Practice (UK) 20, 127-128. 43. Greene, A. E. and Charney, J. (1971) Characterization and identification of insect cell cultures. Cur. Topics Microbial. Zmmunol. 55,5 1-61.