Advances in Cancer Research Volume 93

Contributors

Numbers in parentheses indicate the pages on which the authors’ contributions begin.

Michael Campoli, Department of Immunology, Roswell Park Cancer Institute, Buffalo, New York 14263 (189) Chien-Chung Chang, Department of Immunology, Roswell Park Cancer Institute, Buffalo, New York 14263 (189) Lisa M. Coussens, Cancer Research Institute, Department of Pathology, and Comprehensive Cancer Center, University of California San Francisco, San Francisco, California 94143 (159) Hanna Mellin Dahlstrand, Department of Oncology-Pathology, Karolinska Institute, Karolinska University Hospital, 171 76, Stockholm, Sweden (59) Tina Dalianis, Department of Oncology-Pathology, Karolinska Institute, Karolinska University Hospital, 171 76, Stockholm, Sweden (59) Armin Ensser, Institut fu¨r Klinische und Molekulare Virologie, FriedrichAlexander-Universita¨t Erlangen-Nu¨rnberg, 91054 Erlangen, Germany (91) Soldano Ferrone, Department of Immunology, Roswell Park Cancer Institute, Buffalo, New York 14263 (189) Bernhard Fleckenstein, Institut fu¨r Klinische und Molekulare Virologie, Friedrich-Alexander-Universita¨t Erlangen-Nu¨rnberg, 91054 Erlangen, Germany (91) Alexander Griekspoor, Division of Tumor Biology, The Netherlands Cancer Institute, 1066 CX Amsterdam, The Netherlands (129) Ingrid Jordens, Division of Tumor Biology, The Netherlands Cancer Institute, 1066 CX Amsterdam, The Netherlands (129) Marije Marsman, Division of Tumor Biology, The Netherlands Cancer Institute, 1066 CX Amsterdam, The Netherlands (129) Jacques Neefjes, Division of Tumor Biology, The Netherlands Cancer Institute, 1066 CX Amsterdam, The Netherlands (129) Stephen C. Robinson, Cancer Research Institute, University of California San Francisco, San Francisco, California 94143 (159) Harry Rubin, Department of Molecular and Cell Biology, Life Sciences Addition, University of California Berkeley, Berkeley, California 94720-3200 (1)

ix

Central Roles of Mg2þ and MgATP2 in the Regulation of Protein Synthesis and Cell Proliferation: Significance for Neoplastic Transformation Harry Rubin Department of Molecular and Cell Biology, Life Sciences Addition, University of California Berkeley, Berkeley, California 94720-3200

I. Specific and Nonspecific Stimulators of Cell Proliferation II. Necessity for Prolonged Stimulation and Increased Protein Synthesis to Induce DNA Synthesis III. Requirement of RNA and Protein Synthesis for the Initiation of DNA Synthesis IV. Effects of Inhibitors of Protein Synthesis on Initiation of DNA Synthesis V. Cyclins and CDKs, Specific Proteins Required for the Transition from the G1 to the S Stages of the Cell Cycle VI. Role of Mg2þ in Growth Regulation VII. Protein and DNA Synthesis in Very Low Ca2þ with Variations in Mg2þ Concentrations VIII. Kinetics of Cellular Responsiveness to Mg2þ Limitation in Physiological Ca2þ IX. Mitogen-Induced Increases in Cytosolic Free Mg2þ X. Mg2þ Effects on Diverse Cellular Responses to Growth Factors XI. Possible Roles of Kþ, Ca2þ, pH, and Naþ in Growth Regulation A. Potassium B. Calcium C. pH and Sodium XII. Regulation of Protein Synthesis by the PI 3-K and mTOR Pathways XIII. Role of Cations in Neoplastic Transformation XIV. Conclusions References

Growth factors are polypeptides that combine with specific membrane receptors on animal cells to stimulate proliferation, but they also stimulate glucose transport, uridine phosphorylation, intermediary metabolism, protein synthesis, and other processes of the coordinate response. There are a variety of nonspecific surface action treatments which stimulate the same set of reactions as the growth factors do, of which protein synthesis is most directly related to the onset of DNA synthesis. Mg2þ is required for a very wide range of cellular reactions, including all phosphoryl transfers, and its deprivation inhibits all components of the coordinate response that have so far been tested. Growth factors raise the level of free Mg2þ closer to the optimum for the initiation of protein synthesis. Advances in CANCER RESEARCH 0065-230X/05 $35.00

1

Copyright 2005, Elsevier Inc. All rights reserved

2

Harry Rubin The resulting increase in protein synthesis accelerates progression through G1 to the onset of DNA synthesis and mitosis. None of the other 3 major cellular cations are similarly involved in growth regulation, although internal pH may play an auxiliary role. Almost 105 externally bound divalent cations are displaced from membranes for every attached insulin molecule, implying a conformational membrane change that releases enough Mg2þ from the internal surface of the plasma membrane to account for the increase in free cytosolic Mg2þ. It is proposed that mTOR, the central control point for protein synthesis of the PI 3-K kinase cascade stimulated by insulin, is regulated by MgATP2 which varies directly with cytosolic Mg2þ. Other elements of the coordinate response to growth factors such as the increased transport of glucose and phosphorylation of uridine are also dependent upon an increase of Mg2þ. Deprivation of Mg2þ in neoplastically transformed cultures normalizes their appearance and growth behavior and raises their abnormally low Ca2þ concentration. Tight packing of the transformed cells at very high saturation density confers the same normalizing effects, which are retained for a few days after subculture at low density. The results suggest that the activity of Mg2þ within the cell is a central regulator of normal cell growth, and the loss of its membrane-mediated control can account for the neoplastic phenotype. ß 2005 Elsevier Inc.

I. SPECIFIC AND NONSPECIFIC STIMULATORS OF CELL PROLIFERATION The regulation of cell proliferation almost inevitably became the subject of research with the development of monolayer culture in which all the cells could be observed microscopically and counted by simple techniques. The field is most conveniently called growth regulation, although its major concern is the rate of increase in cell number, but there is an obvious relationship between growth in mass in the form of protein and the division of cells if they are to retain proliferative capacity. When synthetic media consisting of all the known required micronutrients were developed in the middle of the last century, it was realized that cells in culture needed serum proteins to multiply and these were interacting like polypeptide hormones with receptors on the cell surface. Early passage normal fibroblasts derived from chicken or mouse embryos were initially the most commonly used cells for studying growth regulation, partly because they outgrew the other cell types in culture, but they were later joined by mutant derivative cell lines that could multiply indefinitely in culture while retaining normal, growth-regulating behavior. A prominent feature of that behavior was contact inhibition, a term first applied to the inhibitory effect of contact on cell migration but adopted for the limitation on increase in cell number when cells form a confluent sheet covering the entire surface of the culture dish. That limit, called the saturation density, is proportional to the concentration of serum up to a maximum percentage of the medium that varies with different cells. A prominent feature of growth regulation is the failure of normal cells to sustain multiplication in culture unless they attach to and spread on a solid surface. If they are placed in suspension in a semisolid medium, such as soft

Mg2þ in Cell Growth Regulation and Transformation

3

agar or a viscous liquid medium, normal cells will not undergo more than one or two divisions before they become quiescent. Neoplastic cells violate these features of normal growth regulation to a greater or lesser extent. They require less serum to multiply; they have a higher saturation density, sometimes limited only by the supply of micronutrients rather than serum; and they have little or no requirement for attachment to a solid surface. As a result, neoplastic cells can usually be identified by their ability to continue multiplying when surrounded by a confluent sheet of normal cells that has undergone contact inhibition. The neoplastic cells can therefore be quantified by the formation of discrete dense or multilayered foci arising from single cells against a monolayered confluent sheet of normal cells. A major problem of cell culture is to identify the structures, molecules, and pathways that contribute to growth regulation of normal cells and how that regulation is transcended in neoplastic cells. One method for approaching this aim is to use normal fibroblasts that have been contact-inhibited at confluence in the G1 stage of the cell cycle, sometimes further inhibited by withdrawal of serum, then stimulated with a relatively high concentration of serum. After a period of hours that varies with cell type, the cells progress through the G1 stage into the S period of DNA synthesis. The initiation of DNA synthesis and progression through the S period can be monitored through the incorporation of radioactive thymidine into DNA in confluent cultures. This short-circuits the need to measure cell proliferation over several days at low densities and provides ample material for chemically measuring other quantities. Various hormones, such as insulin in the case of chicken embryo fibroblasts, or combinations of hormones act as growth factors sufficient to induce a round of DNA synthesis, which is generally correlated with an increase in the protein content of the cell population. There are other quantities that are stimulated by growth factors within minutes or even seconds, such as the uptake of glucose measured with radioactive analogs, or of uridine which is quickly phosphorylated and destined for incorporation into RNA. These transport and phosphorylation changes are, however, not coupled to growth or DNA synthesis since the external glucose concentration can be lowered so the amount taken up by the stimulated cells is less than that taken up by the quiescent cells with no effect on the onset of DNA synthesis, and uridine is not required at all for growth. The transport of Kþ, Naþ, and Ca2þ into cells increases quickly after stimulation (Rozengurt, 1986) and their putative role in regulating growth will be discussed in the following text. There are other early responses, such as the stimulation of protein and RNA synthesis and a decrease in the rate of protein degradation, which are connected to the onset of DNA synthesis. The sum of such early reactions was initially called the ‘‘pleiotypic response’’ and related to the stringent response of bacteria to amino acid starvation (Hershko et al., 1971). However, no further evidence

4

Harry Rubin

developed for this relationship, and the stereotyped set of animal cell reactions to growth factors and other external conditions will be referred to as the ‘‘coordinate response’’ or ‘‘coordinate control’’ (Rubin, 1975a). Individual hormonelike protein growth factors undoubtedly interact with specific receptors on the external surface of the cell and some schemes of growth regulation call for the activation of different pathways for stimulation of growth (Rozengurt, 1986; Zetterberg and Engstro¨ m, 1983). However, this does not explain the need for attachment and spreading on a solid surface in order for freshly explanted normal cells to grow (O’Neill et al., 1986; Stoker et al., 1968) (Table I) nor the inhibition of growth at confluence. The latter is most clearly shown by the so-called wound-healing experiment in which a strip of cells is removed from a quiescent confluent layer and the cells on the edge of the strip move into the denuded area and multiply at a high rate until confluence is restored (Gurney, 1969) (Table I). It should be noted, however, that cell migration is correlated with proliferation (Barrandon and Green, 1987; Fischer, 1946) and migration stops when a culture becomes confluent. Migration on a solid surface involves continuous perturbation of the plasma membrane and possibly internal membranes such as endoplasmic reticulum as well, which decreases binding capacity for divalent cations, replacing them with monovalent cations (Dawson and Hauser, 1970). Actually, DNA synthesis in the established line of Swiss mouse 3T3 fibroblasts was stimulated sixfold over that of completely suspended cells by attachment of such a small area of their plasma membrane that there was no increase in surface area (O’Neill et al., 1986), indicating a limited perturbation of the membrane is sufficient to activate many cells. In another example of membrane changes, serum-starved quiescent cells exhibit few microvilli at their surface whereas logarithmically growing cells in serum-containing medium reveal an abundance of such microvilli (Evans et al., 1974). Within 1 hr of insulin treatment, microvilli appear at the surface in numbers and subcellular organization characteristic of exponential growth. Since the microvilli are bounded by plasma membrane, their Table I

Nonspecific Mitogenic Stimulation of Nontransformed Animal Cells Treatment

Reference

Attachment and spreading on a solid surface Migration (wound healing)

O’Neill et al., 1986; Stoker et al., 1968 Barrandon and Green, 1987; Fischer, 1946; Gurney, 1969 Carney et al., 1978; Sefton and Rubin, 1970 Rubin and Koide, 1973; Sanui and Rubin, 1984 Bowen-Pope and Rubin, 1983; Rubin and Sanui, 1977

Trypsin and selected proteases Subtoxic concentrations of heavy metals Ca pyrophosphate and Ca phosphate precipitates

Mg2þ in Cell Growth Regulation and Transformation

5

large numbers after insulin treatment represent a significant increase in surface area and movement, which is also likely to decrease the binding capacity of divalent cations. The nonspecificity of stimulation is further brought out by the addition of nonphysiological stimulants to cells. This first drew my attention with the finding that trypsin and other proteolytic enzymes in amounts too small to even retract, much less detach, cells stimulated growth of contact-inhibited cultures of chicken embryo fibroblasts (CEF) (Sefton and Rubin, 1970), and thrombin did the same for mouse fibroblasts (Carney et al., 1978) (Table I). The proposed nonspecificity of stimulation was bolstered by discovering the stimulation of DNA synthesis and hexose uptake in CEF by subtoxic concentrations of Zn2þ, Cd2þ, Mn2þ, Hg2þ, and Pb2þ (Rubin, 1975b; Rubin and Koide, 1973; Sanui and Rubin, 1984). Further support for nonspecificity came from the stimulation of DNA synthesis in Balb/c 3T3 mouse fibroblasts by sodium pyrophosphate at concentrations just sufficient to form flocculent precipitates with calcium (Rubin and Sanui, 1977). (It should be noted that higher concentrations sufficient to bind the Mg2þ levels present in the medium sharply inhibit DNA synthesis.) The stimulatory concentrations also enhance hexose uptake, and exert all their effect at the cell surface (Bowen-Pope and Rubin, 1983). The variety of nonspecific treatments that elicit the same set of early responses and are followed by DNA synthesis and mitosis indicated that they had in common a relatively simple effect on the cell membrane which initially entrained a shared intracellular response. When that stimulus was maintained over the length of the G1 period, it led to the initiation of DNA synthesis followed by mitosis. The same set of responses could, of course, be evoked by polypeptide hormones binding with their specific receptors at the cell surface. The diversity of intracellular effects induced by perturbing the cell membrane suggested that alterations in cation content of the cytosol might serve as a second message to stimulate the pleiotypic response and coordinate with growth and proliferation.

II. NECESSITY FOR PROLONGED STIMULATION AND INCREASED PROTEIN SYNTHESIS TO INDUCE DNA SYNTHESIS In any analysis of growth regulation, it is necessary to establish the essential parameters of the problem. One such parameter is how long a stimulatory agent has to be applied to quiescent (G1 or G0) cells in order to initiate and sustain DNA synthesis in a population. It is generally understood that serum or hormones have to be applied for several hours, well into

6

Harry Rubin

the G1 period, before DNA synthesis begins in a large fraction of animal cells. Early studies on CEF cultures indicated that removal of serum growth factors in the first few hours of the G1 period of the cell cycle kept most cells from entering the S period of DNA synthesis, but the cells were committed to DNA synthesis about 4 hr before the actual start of DNA synthesis, i.e., removal of serum during this late period, about halfway through G1, did not delay the S period (Temin, 1971). Removal of serum from the medium, however, is not equivalent to its removal from the cell. Serum proteins adsorb to the cell surface and are not entirely removed by repeated washing or even by trypsinization (Hamburger et al., 1963). Lowering the pH of the medium from 7.4–7.5 to 6.8 inhibits the proliferation of CEF at high population densities but not at very low population densities (Rubin, 1971b). A combination of removing serum and lowering pH overnight proved to be a particularly effective method for inducing a reversible quiescence in high population densities of CEF (Rubin and Steiner, 1975). When serum and higher pH were restored, a small fraction of the CEF began synthesizing DNA between 2 and 4 hr, and the fraction then increased steeply up to 7 hr (Fig. 1). If either serum or pH or both were restored to inhibitory levels at 2 hr, there was no increase in DNA synthesis in the cultures. If these inhibitory operations were begun at 4 hr when DNA synthesis had already been initiated in a few cells, the removal of serum alone slowed the ensuing steep entry into S by about two-thirds, but the reduction in pH, and particularly the combined treatment, effectively stopped the progression. When the same operations were instituted at 6 hr in the midst of the steep ascent of cells entering S, the removal of serum alone or reduction of pH decreased the rate of entry into S by about one-half or three-fourths, respectively. The combination treatment stopped further progression into S but allowed continuing DNA synthesis in those cells already in the S phase. In effect, then, the passage from G1 into the S phase required the presence of the full stimulatory conditions throughout the entire G1 period of individual cells, though partial escape into S was permitted by maintaining a lower degree of stimulation (Rubin and Steiner, 1975). A similar conclusion could be derived from the use of a purified growth factor MSA (multiplication stimulating activity), which had weaker activity than serum as the stimulant of quiescent CEF (Bolen and Smith, 1977). Unlike removal of serum at different times after the beginning of the S phase in the CEF population, which gave similar partial inhibitions to those already described for the removal of serum, the removal of MSA stopped further progress just as the combination treatment did. The authors concluded that irreversible commitment to DNA synthesis occurred at or near the G1–S boundary. In light of these observations and similar ones in other systems (Prescott, 1968), it is apparent that an early triggering event that irreversibly sets off other processes leading to DNA synthesis cannot account

Mg2þ in Cell Growth Regulation and Transformation

7

Fig. 1 DNA synthesis after growth stimulation of high-density quiescent CEF cultures, and reversal of the stimulation at later intervals. DNA synthesis was turned off by overnight removal of serum and lowering of pH to 6.8. Serum was then restored and pH raised to 7.5 and the cells labeled with 3H-thymidine at 0, 2, 4, 6, and 7 hr. — . At 2, 4, and 6 hr, several cultures were changed to no serum at pH 7.5, – –&; serum at pH 6.8, – –!; no serum at pH 6.8, – –~; and were labeled with 3H-thymidine at 7 hr (Rubin and Steiner, 1975).





 



for the onset of the S phase. Evidence to be considered in the following text indicates that the rate of protein synthesis is the key process that determines the rate of progress through G1, and that the transition to S is brought about by synthesis of a particular protein (cyclin?) just before the transition.

III. REQUIREMENT OF RNA AND PROTEIN SYNTHESIS FOR THE INITIATION OF DNA SYNTHESIS Reproduction of the cell requires a doubling in all its constituents. Protein makes up the largest part of the structure of the cell and RNA, particularly that associated with ribosomal structure, makes a significant contribution to the dry mass of the cell. The rate of protein synthesis in a cycling cell

8

Harry Rubin

increases throughout the entire interphase, reaching a level at mitosis twice that at the beginning of G1 (Zetterberg and Killander, 1965). Analysis of growth in single cells suggests that the initiation of DNA synthesis depends on cell mass (Killander and Zetterberg, 1965). A sharp reduction in serum concentration of perfusion-grown normal human fibroblasts decreases the rate of protein synthesis within 2 hr and increases the rate of protein degradation, resulting in cessation of protein accumulation (Castor, 1977). DNA synthesis only begins to decline at 6 hr. Contact inhibition between normal human fibroblasts also reduces the rate of protein, RNA, and DNA synthesis, and these changes are associated with a large reduction in free cytoplasmic polysomes (Levine et al., 1965). A 10-fold reduction in the rate of protein accumulation is accompanied by only a 37% decrease in the rate of protein synthesis, indicating a role for degradation. Contact inhibition in hamster cells is associated with about a two-thirds decrease in the rate of protein synthesis (Stanners and Becker, 1971). The reduction can be accounted for by the observations that (a) the average cell in stationary phase contains about half the total number of ribosomes per cell as the average cell in exponential growth and (b) only two-thirds of the ribosomes are bound in polysomes in stationary phase while virtually all of them are bound in polysomes in the exponential phase. The polysomebound ribosomes of the stationary phase function with the same efficiency as those in the exponential phase, and produce proteins of about the same average length. The results suggest that the higher proportion of free ribosomes in stationary phase is not due to a limitation of messenger RNA but to a decreased probability of attachment of ribosomes to messenger RNA (Stanners and Becker, 1971). Related results were found after treatment of chick embryo epidermis with epidermal growth factor (EGF) (Cohen and Stastny, 1968). Within an hour after addition of EGF to the epidermal cells, there is a conversion of preexisting ribosomal monomers into polysomal structures with an increase in protein synthesis accompanied by synthesis of all classes of cytoplasmic RNA. The conversion of ribosome monomers to polysomes and the stimulation of RNA synthesis do not require synthesis of new protein, nor do they require the increased transport of glucose and amino acids induced by EGF. The ribosome monomers decrease to 70% of their unstimulated value within half an hour and to a stabilized level 30 to 50% of their unstimulated value at 1.5 hr. None of the articles describing the changes in polysome formation with growth state has proposed a mechanism based on observations of the in vitro formation of polysomes. It should be noted, however, that the rate of protein synthesis in vitro, which is based on the formation of polysomes, is acutely dependent on free Mg2þ concentration, estimated in some mammalian cells to be about 1 mM (Rink et al., 1982). However, a wide variety of free Mg2þ concentrations have been reported depending on the method

Mg2þ in Cell Growth Regulation and Transformation

9

used for measurement, the tissue involved, and its metabolic state (Garfinkel and Garfinkel, 1988). Most of the estimates fall below 1 mM, some as low as 0.1 to 0.2 mM, with many in the range of 0.4 to 0.6 mM. The results as a whole show that only a small fraction of the total cell Mg2þ of 10 to 20 mM is free, and that varies with metabolic conditions. Newer estimates of cytosolic Mg2þ fall in the range of 0.25 to 1.0 mM (Grubbs, 2002), where it can exert substantial control on metabolism (Garfinkel and Garfinkel, 1988). Protein synthesis on mammalian polysomes in vitro increases sharply between 0.6 and 2.6 mM Mg2þ and decreases almost as sharply between 2.6 and 5.6 mM Mg2þ in a bell-shaped curve when corrected for the chelating action of ATP and GTP (Schreier and Staehelin, 1973) (Fig. 2). The optimal, uncorrected Mg2þ concentration varies with the source of messenger RNA from 2 to 3 mM Mg2þ concentration (Brendler et al., 1981; Ilan and Ilan, 1978). Protein synthesis in vitro also varies with Kþ concentration over a much broader range of about 40 to 120 mM Kþ. There is not an absolute requirement for Kþ, however, since it can be replaced by NH4þ

Fig. 2 Mg2þ and Kþ concentration dependencies of globin synthesis in vitro by mouse liver ribosome subunits labeled with 14C-leucine. (A) Mg2þ-dependence in 70 mM Kþ. After correction for chelation of Mg2þ by ATP and GTP, the effective Mg2þ concentration is 1.4 mM lower than that indicated. (B) Kþ-dependence in 3 mM Mg2þ – – ; in 4 mM Mg2þ — (Schreier and Staehelin, 1973).

 

 

10

Harry Rubin

in bacterial polysome formation in vitro (Meselson et al., 1964). The initiation of protein synthesis requires a higher Mg2þ concentration than elongation of the peptide chain (Revel and Hiatt, 1965; Schreier and Noll, 1971). Contrary to the general view, the Mg2þ-sensitive step is not the binding of messenger RNA to the small ribosomal subunit but is thought to be the first translocation (Schreier and Noll, 1971). The significance of these findings for the role of Mg2þ in control of protein synthesis in cultured animal cells will be discussed later in the appropriate places.

IV. EFFECTS OF INHIBITORS OF PROTEIN SYNTHESIS ON INITIATION OF DNA SYNTHESIS The foregoing results indicate essential roles for protein and RNA synthesis in the initiation of DNA synthesis. Specific inhibitors of these syntheses can provide a more precise indication of the roles of each of these macromolecular species in growth regulation. Cycloheximide or actinomycin were added to synchronized HeLa cells to inhibit protein synthesis or RNA synthesis, respectively (Kim et al., 1968). Exposure to either drug in the G1 period completely prevented the onset of DNA synthesis. Exposure to a relatively large dose of cycloheximide in the DNA synthetic phase that immediately abolishes protein synthesis also shuts down DNA synthesis, whereas actinomycin D allowed the cells to continue DNA synthesis for 2 hr before steeply reducing its synthesis. The results indicate that DNA synthesis is directly dependent on concurrent synthesis of protein and only after a 2-hr delay on the synthesis of RNA. That delay may be due to the need for exhaustion of messenger RNA molecules before its effect on protein synthesis is felt. The initiation of DNA synthesis is more sensitive to inhibitors than is the continuation of DNA synthesis. The activities of several enzymes necessary for the formation of DNA were not reduced by several hours of treatment during the S period with either drug while the rate of DNA synthesis was markedly reduced. It was proposed that there is a protein with a very short half-life that is necessary for continuing DNA synthesis (Kim et al., 1968). The addition of low concentrations of cycloheximide at the end of the G1 period to synchronized 3T3 cells rapidly reduced the rate at which the cells enter the S phase by an amount proportional to the inhibition of protein synthesis (Brooks, 1977) (Fig. 3). This suggests that the initiation of DNA synthesis depends on the continuous synthesis of a protein with a short halflife. The rate-limiting transition occurs within 2 hr of the start of DNA synthesis. The addition of a very low dose of cycloheximide (33 ng/ml) at the beginning of the G1 period completely suppressed DNA synthesis over a

Mg2þ in Cell Growth Regulation and Transformation

11

Fig. 3 Inhibition of protein synthesis by cycloheximide and its effect on the initiation of DNA synthesis in Swiss 3T3 cells. Quiescent cells were stimulated with serum in the presence of the indicated concentrations of cycloheximide. Cultures were pulse-labeled for protein synthesis with 3H-leucine at 2 hr (d), and continuously labeled for 24 hr with 3H-thymidine ( ). Ordinate: radioactivity incorporated as percentage of control (no cycloheximide) (Brooks, 1977).



24-hr period although protein synthesis itself was reduced by less than 30%, again indicating the acute sensitivity of the initiation of DNA synthesis to the rate of protein synthesis. These results suggested that the transition from G1 to S periods depends on the synthesis of a protein with a short half-life (Brooks, 1977). Additional evidence for a labile protein needed for the initiation of DNA synthesis came from a variety of experimental manipulations of protein synthesis during the G1 period (Rossow et al., 1979; Schneiderman et al., 1971).

V. CYCLINS AND CDKs, SPECIFIC PROTEINS REQUIRED FOR THE TRANSITION FROM THE G1 TO THE S STAGES OF THE CELL CYCLE When cells are maintained in culture under conditions that increase their generation time, they do so by expanding the G1 period (Prescott, 1968). For example, CEF maintained in the absence of serum and/or at low pH can be maintained in healthy condition for 2 days during which very few cells

12

Harry Rubin

enter the S period, but the addition of serum and/or raising pH allows most of the cells to initiate DNA synthesis within 4 to 6 hr (Rubin and Steiner, 1975). Different epithelia of the living mouse have average generation times varying from 16.7 hr in the lower ileum to 181 hr in the esophagus, but the S period remained about 7 hr in all the tissues (Cameron and Greulich, 1963). Although there may be some variation in the length of the G2 period, the greatest flexibility undoubtedly lies in the G1 portion of the interphase. G1 is primarily a period of growth during which the cell has to reach a minimal mass before initiation of DNA synthesis, and the rate of its accomplishment apparently depends on the amount of general metabolic machinery. The regulation of cell proliferation therefore depends on the rate of protein synthesis during G1 (Liskay et al., 1980; Rossow et al., 1979). As has been indicated, however, the requirement of protein synthesis up to the time of DNA synthesis suggested that synthesis of a particular, unstable protein (or proteins) is needed to bring about the transition to the S period (Brooks, 1977; Kim et al., 1968; Liskay et al., 1980; Rossow et al., 1979; Schneiderman et al., 1971). Although existence of an unstable protein that is required for the initiation of DNA synthesis in animal cells has been posited for some time (see Prescott, 1968), one of the first, if not the first, isolation of a protein that apparently had the requisite properties was reported only in 1983 (Croy and Pardee, 1983). At this time, the concept was developing that there were cell division cycle (CDC) mutants in yeast that arrested division at unique stages of the cell cycle regardless of the time they were shifted from permissive to restrictive temperature (Hartwell, 1991). It should be noted, however, that ribonucleotide reductase, the enzyme that converts cytosine monophosphate to dCMP, exhibits a strict parallelism with DNA synthesis (Turner et al., 1968). It was estimated that there were as many as 500 genes with stage-specific functions, or 10% of the yeast genome. Attention turned to mammalian cells and proteins that control the transition from G1 to S. Assuming that vertebrates have the same genomic proportion of stagespecific functions as does yeast, there would be thousands of genes controlling such functions in animal cells. Candidate regulators of the G1/S transition included cyclins and their associated protein kinases (Matsushime et al., 1991). Cyclins were first identified in marine invertebrates as proteins that undergo periodic fluctuations during each cell cycle. Genes required for G1 progression in yeast include cyclins 1, 2, and 3. Because the mutants of the 3 genes are functionally redundant, their combined inactivation is required to induce cycle arrest in G1. Observations that the transcripts of cyclin 1 and 2 increase during G1 and decrease as the cells enter the S phase, and noting the association of cyclin 2 protein with a CDC kinase, reinforce the view that these genes form part of the regulatory apparatus that governs the G1/S transition in yeast. The first cyclinlike (CYL) protein was

Mg2þ in Cell Growth Regulation and Transformation

13

reported in mouse macrophages (Matsushime et al., 1991). Deprivation of a growth factor during G1 leads to degradation of a CYL protein and correlates with failure to initiate DNA synthesis. The timing of this CYL expression, its rapid turnover in absence of the growth factor, and its transient binding to a CDC-related polypeptide suggested that this CYL gene functions during S phase commitment. It was later called cyclin D1 and was recognized as a member of a family that includes two other related genes, cyclins D2 and D3 (Sherr, 1993). The cyclins bind to and activate cyclin-dependent kinases (CDKs) which phosphorylate proteins that control the transition from the G1 to the S phase of the cell cycle. The levels of mRNAs for cyclins D1 and D3 decrease in human diploid fibroblasts upon serum depletion or at high cell densities (Won et al., 1992). Following stimulation of quiescent fibroblasts with serum, the mRNA levels increase gradually to a peak at 12 hr prior to the onset of the S phase and then decline, which suggests a correlation between their gene products and the induction of DNA synthesis. However, induction of these genes is not sufficient for the transition from quiescence into S phase (Won et al., 1992). Cyclin E was maximal in synchronized Hela cells near the G1–S phase boundary as was the cyclin E-associated protein kinase activity (Dulic et al., 1992). The kinase activity declined when cells entered G2. Cyclin E associated with Cdk2 and induced maximal levels of cyclin E-dependent kinase activity at the G1–S transition (Dulic et al., 1992; Koff et al., 1992). It is thought that cyclins D and E might fulfill different cyclinlike functions required for controlling late G1 restriction point control, with cyclin A triggering S phase; alternatively, cyclin E might be necessary for the actual onset of DNA synthesis (Sherr, 1994). Many other elements of cell cycle control have been identified, including ubiquitin ligases for degrading cyclins and proteins that inhibit Cdk activity by dephosphorylation (Alberts et al., 2002). Since there are likely to be thousands of stage-specific proteins in vertebrate cells based on the estimates of the number in yeast (Hartwell, 1991), the level of complexity of checkpoints controlling the progress of cells through G1 and the transition to the S phase will continue to grow as the number of identified proteins increases. The cyclin-controlled checkpoints, however, are sequential obligatory gatekeepers that are dependent on the orderly progression of cells through the cycle. They are not determinants of the rate of progression. They occur in proper order whether cells are growing slowly or rapidly. Our focus here is on the regulation of the rate at which cells cycle, and particularly what moves them from a relatively stationary state with an expanded G1 phase to a rapidly multiplying state with a much shortened G1 phase. That change in state depends on events that begin with a persistent perturbation of the cell membrane by specific growth factors or nonspecific agencies of various kinds and is maintained at least through the G1 period for a single cycle of initiation of DNA synthesis, or throughout the

14

Harry Rubin

cell cycle for a doubling of an entire cell population. Basically, we are interested here in what changes are entrained within the cell by the membrane perturbations that drive the cell through the G1 period at an accelerated rate. The molecular biology of such changes will be considered in a later section.

VI. ROLE OF Mg2þ IN GROWTH REGULATION As noted earlier, there is a direct relationship in cells between the rate of protein synthesis in G1 and the rate of passage through G1 into the S phase of DNA synthesis (Brooks, 1977; Kim et al., 1968; Stanners and Becker, 1971). Protein synthesis itself depends on polysome formation which is acutely sensitive to the concentration of free Mg2þ (Brendler et al., 1981; Schreier and Staehelin, 1973). Also, ribosomal monomers combine with mRNA and are converted into polysomes in the absence of protein synthesis within half an hour of adding growth factor to cells and without the increased glucose or amino acid transport evoked by the growth factor (Cohen and Stastny, 1968). The external stimulus has to be present through most, if not all, of G1 in order to initiate DNA synthesis (Bolen and Smith, 1977; Brooks, 1977; Rubin and Steiner, 1975). These relationships suggested that perturbation of the cell membrane by specific or nonspecific stimulants of growth releases Mg2þ from binding sites on phospholipids and proteins on the internal surface of the plasma membrane (Dawson and Hauser, 1970) or other membrane sites (Sanui, 1970), thereby increasing the free Mg2þ concentration to accelerate the rate of protein synthesis and progress through G1 to S. Changes in Mg2þ pump activity could also contribute to increased cellular Mg2þ (Lostroh and Krahl, 1974). To determine whether altering the cellular content of Mg2þ would influence the rate of DNA synthesis, the author varied the concentration of Mg2þ in the medium downward from physiological levels to almost zero and the effect on DNA synthesis was recorded in sparse and crowded CEF cultures (Rubin, 1975a). There was a sharp decrease in the rate of DNA synthesis in the sparse cultures when measured after 16 hr when the concentration of external Mg2þ was reduced below 0.2 mM. There was a less dramatic decrease of DNA synthesis in the crowded cultures, with less than 0.1 mM Mg2þ. The results obtained at the low levels of Mg2þ were erratic, so a buffering agent was sought that would complex most of the Mg2þ in the medium but maintain the free Mg2þ at a constant level. Phosphorylated compounds like ATP and inorganic pyrophosphate, which bind Mg2þ a little more firmly than Ca2þ, inhibited DNA synthesis at 16 hr when their concentration exceeded that of Mg2þ, but were largely independent of the concentration of Ca2þ. The effect of Mg2þ binding by inorganic pyrophosphate was examined on other elements of the coordinate response to growth factors.

Mg2þ in Cell Growth Regulation and Transformation

15

There were decreases in the rates of 2-deoxyglucose uptake, synthesis of RNA and protein, and lactic acid production. The cells remained in healthy condition, and the effects were fully reversed by adding excess Mg2þ, but not Ca2þ, to the medium after 16 hr of inorganic pyrophosphate treatment. Similar results were obtained in crowded CEF cultures by simply lowering Mg2þ in the medium below 0.1 mM (Kamine and Rubin, 1976). The intracellular concentration of Mg2þ and Kþ decreased almost twofold when external Mg2þ was lowered from 0.8 to 0.016 mM; by contrast, intracellular Naþ and Ca2þ levels increased 2.2- and 1.5-fold, respectively (Sanui and Rubin, 1977). The external Mg2þ effects were greater when the Ca2þ concentration of the medium was decreased from 1.0 mM to 0.2 mM (Fig. 4). The decreases in DNA synthesis paralleled the decrease of intracellular Mg2þ. The results also indicated that the decreased level of Ca2þ in the medium increased the permeability of the cells, thereby allowing freer exchange of Naþ, Kþ, and Mg2þ.

Fig. 4 Mg2þ dependence of DNA synthesis and cation content in CEF. Quiescent cells were stimulated with medium containing serum in 0.22 mM Ca2þ and varying concentrations of Mg2þ. At 16 hr, the cellular cation concentrations were determined in some cultures, while others were labeled with 3H-thymidine for 1 hr (Sanui and Rubin, 1977).

16 Table II

Harry Rubin Intracellular Cation Changes 16 to 17 hr After Adding Mitogen

Cation

Insulin 0.1 unit/ml on CEF (59)

Calf serum 20% on Balb/c 3T3 cells (68)

Mg2þ Kþ Ca2þ Naþ

þ22% þ14% No change Negligible

þ14% þ3% 58% 9%

The significance of these observations for physiological growth control was evaluated by measuring changes of intracellular cation content after adding growth factors to quiescent cell cultures. The addition of insulin in graded concentrations from 0 to 0.1 U/ml to serum-free medium on confluent CEF cultures for 16 hr resulted in a 17-fold increase in DNA synthesis which began to approach a maximum with as little as 0.01 U/ml of insulin (Sanui and Rubin, 1978). Concomitantly, there were graded increases in intracellular Mg2þ (22%) and Kþ (14%) with no change in intracellular Ca2þ (Table II). Naþ rose slightly with 0.01 U/ml insulin but remained at control levels with 0.1 U/ml insulin. The intracellular concentrations of Mg2þ and Kþ increased relative to controls within 10 min after the addition of 0.1 U/ml insulin and remained higher through 16 hr. There was no significant effect of insulin on total Ca2þ or Naþ at any time through 16 hr. It will be shown later that only the increases of intracellular Mg2þ induced by the insulin treatment had significant effects on the rate of protein synthesis, which were later translated into increases in DNA synthesis (Moscatelli et al., 1979). The externally bound cations were determined by washing the cells 5 times in CO2-free 0.25 M sucrose solution to remove unbound cations without displacing those that were bound (Sanui and Rubin, 1978). The surface-bound cations were removed for measurement by a 10-sec wash with carbonated (pH 4) 0.25 M sucrose to exchange Hþ for the cations (Sanui and Rubin, 1979b). Hþ are especially efficient in displacing bound cations because they are bound to membranes about 100 times more tightly than Ca2þ or Mg2þ, which, in turn, are bound about 100 times more tightly than Naþ and Kþ (Carvalho et al., 1963). There was a sharp decrease in externally bound Ca2þ of cells treated with 0.001 units of insulin and a smaller drop in Mg2þ. There was a more gradual decrease in externally bound Ca2þ and Mg2þ with higher doses of insulin up to a maximal decrease in 0.1 units of insulin of 34% externally bound Ca2þ and 45% externally bound Mg2þ. Data for externally bound Kþ and Naþ showed wide variation but the general trend was upward in exchange for the loss of externally bound Ca2þ and Mg2þ. It was estimated that about 105 divalent cations were displaced per bound insulin. This relation differs by 5 orders of magnitude from the stoichiometric

Mg2þ in Cell Growth Regulation and Transformation

17

changes in bound cations produced by cation exchange or by complexing agents like ATP or EDTA (Sanui and Pace, 1967). These results are consistent with the suggestion (Shlatz and Marinetti, 1972) that insulin induces a generalized conformational change in the membrane to which it binds and thereby causes the gross changes in bound cations. While Caþ is the major divalent cation bound to the external surface of intact cells, Mg2þ would be the dominant divalent cation bound to the inner surface of the membranes since there is so much more free Mg2þ than free Ca2þ in the cell. Any conformational change in the insulin-treated membrane would be expected to release enough Mg2þ to significantly raise the free Mg2þ concentration of the cytosol. Cortisol has effects on DNA synthesis and other elements of coordinate control in CEF that are antagonistic to the stimulatory effects of insulin (Fodge and Rubin, 1975a; Rubin, 1977). It inhibits the uptake of 2-deoxyglucose and uridine and the incorporation of uridine and thymidine into acid insoluble material. It has been estimated that 3000 additional divalent cations are bound to purified rat liver membranes for every dihydroxycortisone molecule bound, which indicates that it too produces a conformational membrane change (Shlatz and Marinetti, 1972). This would result in an increase in binding of cytosolic Mg2þ and a lowering of free Mg2þ. Lowering intracellular Mg2þ by depriving cells of the external supply reproduces the same coordinate inhibition of transport and metabolism as does the addition of cortisol or removal of serum (Rubin, 1976) and reinforces the idea that Mg2þ acts as the primary secondary messenger for the hormones. Although insulin is sufficient to stimulate DNA synthesis in CEF in the absence of serum, additional growth factors are needed to stimulate DNA synthesis in the established line of Balb/c 3T3 mouse fibroblasts, which have a much higher serum requirement for growth than CEF. The 3T3 cells were grown to confluence in 10% serum and incubated to a quiescent state for 1 day in 1% serum (Sanui and Rubin, 1982a). They were then shifted to media with 0 to 20% serum for 17 hr when measurements were made of the intracellular concentrations of the 4 cations and the rate of DNA synthesis. The rate of DNA synthesis increased 30-fold between 0 and 20% serum. The intracellular cation concentrations at 20% serum as compared with 0% serum were increased 14% for Mg2þ and 3% for Kþ, but were decreased 58% for Ca2þ and 9% for Naþ (Table II). Hence, the only significant increase in intracellular cation content was Mg2þ, with Ca2þ and Naþ showing large to minimal decreases in concentrations, respectively. The major increase in intracellular Mg2þ occurred between 2 and 5 hr with further increases to 10 hr and a slight increase to 15 hr of serum treatment. The major seruminduced increase in DNA synthesis of 3T3 cells occurred between 10 and 17 hr, unlike insulin-stimulated CEF, which began DNA synthesis in 4 to 5 hr after adding serum (Rubin and Steiner, 1975). The increase in Mg2þ and that

18

Harry Rubin

of protein synthesis precedes that of DNA synthesis by several hours in stimulated CEF and 3T3 cells (Rubin and Sanui, 1979). Externally bound Mg2þ decreased gradually with increasing serum concentration to a level in 20% serum being 45% of that in the absence of serum (Sanui and Rubin, 1982a). Externally bound Ca2þ was much higher than Mg2þ and decreased sharply between 0 and 1% serum with a continued decrease to 20% serum down to 22% of the value of externally bound Ca2þ in the absence of serum. Hence, the use of serum as growth stimulant revealed the same picture with regard to release of externally bound cations as that found in the stimulation of CEF by insulin (Sanui and Rubin, 1978). Although no estimate was made of the number of binding sites for serum growth factors, some of which remain unknown, it is likely that the number of divalent cations released exceeds by far the number of receptor-growth factor interactions as was the case for insulin, and involves a conformational change in the membrane. As in the insulin case, the conformational change would release much more Mg2þ than Ca2þ from the inner side of the membrane and significantly raise the level of free Mg2þ in the cytosol by cation exchange, mainly with Kþ, the major intracellular cation. The increased free Mg2þ would propel increased protein synthesis and quicker transit through G1 to the onset of DNA synthesis. Although the release of Mg2þ from internal membranes would raise the cytosolic free Mg2þ, it would not account for the increase in total intracellular Mg2þ observed in cell stimulation by insulin or serum. That increase must come from increased uptake of extracellular Mg2þ into the cell. The increased total Mg in itself suggests that free Mg2þ also increases. It has, in fact, been reported that free Mg2þ rises disproportionately with increases in total Mg2þ (Corkey et al., 1986). That would imply that free Mg2þ would rise more than 20% after stimulation. Given the steep dependence of protein synthesis on Mg2þ concentration (Schreier and Staehelin, 1973) and of DNA synthesis as a function of protein synthesis, a sustained increase in Mg2þ would be expected to sharply increase DNA synthesis in cells. As will be described later, the increased Mg2þ operates on protein synthesis through an increase in the MgATP2, which activates a key protein kinase in a molecular pathway that regulates the initiation of translation.

VII. PROTEIN AND DNA SYNTHESIS IN VERY LOW Ca2þ WITH VARIATIONS IN Mg2þ CONCENTRATIONS Extreme reduction of Ca2þ in the medium from 1.5 mM to 0.02 mM reduces surface-bound Ca2þ by 75% and intracellular Ca2þ by 40% (Rubin et al., 1978). It increases the passive permeability of the cell as indicated by

Mg2þ in Cell Growth Regulation and Transformation

19

measured uptake of L-glucose (Bowen-Pope and Rubin, 1977), and inhibits the onset of DNA synthesis in crowded but not sparse cultures of 3T3 cells. The inhibition of DNA synthesis does not result from a reduction in total intracellular Mg2þ since it remains constant, but DNA synthesis is reversed by raising Mg2þ in the medium from 1.0 to 19 mM Mg2þ, which raises intracellular Mg2þ by about 40%. In contrast, the inhibition of DNA synthesis induced by drastic reduction of Mg2þ is not reversed by raising Ca2þ. These results suggest that the 40% reduction in intracellular Ca2þ induced by very low Mg2þ in the medium results in an increase of binding of intracellular Mg2þ to negatively charged membrane sites once occupied by Ca2þ, thereby bringing on a decrease in free Mg2þ which is compensated by raising external Mg2þ. The failure of high Ca2þ to reverse the inhibition produced by very low Mg2þ supports the primary role of Mg2þ in growth regulation. Increasing extracellular Mg2þ above 20 mM in very low Ca2þ inhibits DNA synthesis, giving a bell-shaped curve for DNA synthesis as a function of Mg2þ concentration that resembles the relation between Mg2þ concentration and protein synthesis in vitro (Fig. 2). The foregoing results prompted a study of the relation of protein synthesis to DNA synthesis as a function of Mg2þ concentration in very low Ca2þ (Rubin et al., 1979). Protein synthesis increased sharply with increasing external Mg2þ in the presence of 0.02 mM Ca2þ, and then decreased with very high concentrations of Mg2þ when measured at 3 hr after changing the cation concentrations (Fig. 5). The figure shows only the internal Mg2þ concentrations, but the corresponding external concentrations are in the figure legend. The peaks of synthesis of protein at 3 hr and DNA at 17 hr did not always coincide precisely with each other as they do in Fig. 5. However, both functions always exhibited a bell-shaped curve as a function of Mg2þ concentration. External Mg2þ of 35 mM in 0.02 Ca2þ was inhibitory for protein synthesis at 5 hr, but it began to rise at 10 hr and was at a maximum level at 22 hr (Fig. 4 in Rubin et al., 1979). The high level of intracellular Mg2þ at 3 hr in 30 mM external Mg2þ and low Ca2þ decreased in the following hours. These results indicated that the cells extruded their excess inhibitory Mg2þ and thereby increased the rate of protein synthesis, followed in about 5 hr by an increase in DNA synthesis. This behavior further supported a causal consecutive relationship between intracellular Mg2þ concentration, protein synthesis, and DNA synthesis. External Mg2þ of 48 mM in low Ca2þ reduced protein synthesis almost to zero at 3 hr with no recovery at a later time. Intracellular Mg2þ continued to rise with time, indicating the cells were unable to extrude a large excess of inhibitory Mg2þ. These results again confirmed a causal, tandem relationship between intracellular Mg2þ levels, protein synthesis, and DNA synthesis. At very low concentrations of both extracellular Ca2þ and Mg2þ, intracellular Naþ rose sharply and Kþ declined at 3 hr, but at higher extracellular

20

Harry Rubin

Fig. 5 Rates of protein and DNA synthesis in Balb/c 3T3 cells as a function of intracellular Mg2þ concentration in very low external Ca2þ concentration. Quiescent cultures were stimulated with serum in medium with 0.02 mM Ca2þ and 1.0, 19, 30, and 48 mM Mg2þ. At 3 and 17 hr, the intracellular concentration of Mg2þ of each group was measured. At 3 hr, the rate of protein synthesis was determined by labeling with 3H-leucine, ~– –~, and at 17 hr, the rate of DNA synthesis was determined by labeling with 3H-thymidine, — . Controls in 1.7 mM Ca2þ and 1.0 mM Mg2þ were labeled with 3H-leucine m– –; or 3H-thymidine d—. The rates of incorporation are plotted against the intracellular concentrations of Mg2þ (Rubin et al., 1979).

 

Mg2þ beginning at 1.0 mM and extending through 48 mM, the intracellular Naþ and Kþ remained normal. The increased Naþ and reduced Kþ at 3 hr returned to normal values at 17 hr. Since the rates of protein and DNA synthesis remained at a markedly reduced level at 17 hr in the combination of very low Ca2þ and Mg2þ and in the combination of very low Ca2þ and very high Mg2þ, it is apparent that neither Kþ nor Naþ influenced the major macromolecular syntheses. Since intracellular Ca2þ remained constant under all the conditions, it is also evident that intracellular Mg2þ was the only cation whose concentration was correlated with protein synthesis, which later determined DNA synthesis. Uridine uptake, which depends on the rate of its phosphorylation, is stimulated within a few minutes of the addition to quiescent cells of 10%

Mg2þ in Cell Growth Regulation and Transformation

21

serum in physiological Ca2þ and Mg2þ (Rozengurt and Stein, 1977). The high serum fails to stimulate uridine uptake in low Ca2þ with low Mg2þ but does so if the Mg2þ level is raised to 10 or even to 40 mM (Bowen-Pope et al., 1979). Since 40 mM Mg2þ sharply depresses protein synthesis and DNA synthesis, it raises the question whether any of the early transport responses significantly affect the macromolecular syntheses required for growth.

VIII. KINETICS OF CELLULAR RESPONSIVENESS TO Mg2þ LIMITATION IN PHYSIOLOGICAL Ca2þ The wide variations in Mg2þ content of cells already described and their effects on protein and DNA synthesis were made possible by varying external Mg2þ over a wide range in very low external Ca2þ, which itself alters cellular cation content, although only temporarily. A transformed clone of 3T3 cells was found in which intracellular Mg2þ could be reduced to 50% of normal by severely lowering external Mg2þ for 12 hr in physiological concentrations of Ca2þ, and quickly restored to normal after 2.5 days by adding back Mg2þ (Fig. 6) (Terasaki and Rubin, 1985). This allowed observation of the dynamics of change in protein and DNA synthesis to determine how sensitive these functions are to intracellular Mg2þ in physiological Ca2þ. The deprivation of external Mg2þ for only 3 hr reduced cellular Mg2þ to 84% of control values and to 67% at 12 hr, beyond which it leveled off at 50% of control values (Fig. 6A). The rate of protein synthesis decreased to almost exactly the same extent and with the same kinetics as that of cellular Mg2þ (Fig. 6B). There was a 12-hr delay before the rate of DNA synthesis began to decline, but it reached a level 100-fold less than the control value between 36 and 48 hr (not shown), with only a minor decrease in cellular Kþ within a range known to have no effect on cellular activities (Moscatelli et al., 1979). It is important to note that there was no loss of total protein in the limited concentration of Mg2þ. In fact, there was a slight increase in total protein and the cells appeared in healthy condition after 60 hr with the only morphological change a flattening out so they appeared more like normal than transformed cells. Restoration of physiological Mg2þ to the deprived cells was followed by a coordinated return to normal of cellular Mg2þ and protein synthesis (Fig. 6C, D). Both parameters doubled within 1 hr and remained at that level thereafter. The rate of DNA synthesis showed its first increase between 8 and 12 hr after restoration of Mg2þ and rose rapidly to normal levels at 24 hr with a kinetics paralleling that of serum stimulation. The total protein

22

Harry Rubin

Fig. 6 Parallel decreases with time in Mg2þ content and rates of protein synthesis in Balb/c 3T3 cells with deprivation of Mg2þ and their increase following restoration of Mg2þ. Medium with either 0.02 mM or 1.0 mM Mg2þ was added to Balb/c 3T3 cultures with 1.7 mM Ca2þ and the rate of protein synthesis (A), and intracellular Mg2þ content (B) determined at indicated intervals up to 48 hr. The medium was then changed in the remaining low-Mg2þ cultures to medium with 1.0 or 0.02 mM Mg2þ and rate of protein synthesis (C), and the Mg2þ content (D) determined at intervals up to 12 hr continuously in 1.0 mM Mg2þ d—d; 0.02 mM Mg2þ d– –d; 48 hr in 0.02 mM Mg2þ, then switched to 1.0 mM Mg2þ — (Terasaki and Rubin, 1985).

 

level in the Mg2þ-deprived cultures remained constant throughout while a detectable and steady increase with the restoration of Mg2þ began at 4 hr. The overall results of this experiment showed that the rate of protein synthesis correlated directly and proportionately with changes in cellular Mg2þ. The magnitude of the changes in protein synthesis were within the limits of those reported for the removal or restoration of growth factors in a variety of cell systems (Castor, 1977; Cohen and Stastny, 1968). The role of growth factors in altering rates of protein and DNA synthesis can therefore be fully accounted for by changes in total cellular Mg2þ, which is presumably reflected in free cytosolic Mg2þ.

Mg2þ in Cell Growth Regulation and Transformation

23

IX. MITOGEN-INDUCED INCREASES IN CYTOSOLIC FREE Mg2þ It was presumed in this work that the increases in total cellular Mg observed after treatment of CEF with insulin or of 3T3 cells with serum were associated with increases in free Mg2þ. This presumption was consistent with the observation that hepatocytes from streptozotocin-induced diabetic rats had 22% less total Mg2þ than those from normal rats and their free Mg2þ was 55% less as determined by null point titrations (Corkey et al., 1986). Total Mg2þ decreases in lymphoma cell lines as the cells enter the stationary phase (Hosseini and Elin, 1985). Somewhat surprisingly, the fraction of bound Mg2þ, as determined by specialized methods, increases as the total decreases, implying that free Mg2þ decreases disproportionately. It was reported that total cellular Mg2þ increased sevenfold with time after quiescent, confident mammary epithelia were subcultured at low density in correlation with DNA synthesis and returned to basal level when they became confluent again (Wolf et al., 2004). However, these findings did not establish that an increase in free Mg2þ occurred in cells treated with growth factors. The development of a Mg2þ-sensitive indicator mag-fura-2 (Murphy et al., 1989; Raju et al., 1989) facilitated the measurement of intracellular free Mg2þ. Epidermal growth factor (EGF) produced a 48-fold increase of DNA synthesis in a serum-starved line of muscle cells. Free Mg2þ increased after a 5-min lag period, rising gradually from an initial 0.32 mM to as high as 1.4 mM at 20 min after the addition of EGF to responsive cells and then leveling off (Grubbs, 1991). The dependence of EGF for increase in free Mg2þ was similar to that of DNA synthesis. There were no changes of pH or free Ca2þ during 20 min in the EGF-stimulated cells. The mag-fura-2 indicator was also used to monitor free Mg2þ in quiescent, confluent Swiss 3T3 cells stimulated by insulin or by EGF in combination with insulin (Ishijima et al., 1991). The combination led to a significant increase in free Mg2þ from basal 0.22 mM to 0.29 to 0.35 mM after 30 to 60 min. The free Mg2þ then began to decline, but the measurements became unreliable because the mag-fura-2 leakage after 20 to 30 min caused considerable error. There were also increases in free Ca2þ, but they were rapid and transient in contrast to the slow, long-lived increases in free Mg2þ. Further study revealed, however, that there was a rapid, early increase in free Mg2þ in bombesin-stimulated 3T3 cells (Ishijima and Tatibana, 1994). The free Mg2þ reached peak values in 15 sec in most cells and lasted for only 1 to 2 min. It was not dependent on external Mg2þ, but was partly dependent on external Ca2þ and the action of tyrosine kinase. The physiological role of this transient increase in free Mg2þ is not known, but it is obviously not involved in the long-term exposure through G1 to growth

24

Harry Rubin

factors that are associated with the increase in protein and the onset of the S phase. We can learn more by taking a step away from the multicellular state in the strict sense and growth factors per se. The oocytes of Xenopus laevis are large cells—850 to 950 m in diameter, 350 to 500 g in weight—that can be isolated from their environment by immersion in paraffin oil. The paraffin oil allows the controlled alteration of the intracellular ionic state of the oocyte by injection of ions and measurement of amino acid incorporation in a single cell without a contribution from external cations (Horowitz and Lau, 1988). Exposure of the quiescent oocyte in its follicle to gonadotropin stimulates 3 H-leucine incorporation into protein and increases intraoocytic Kþ activity for at least 10 days (Lau et al., 1988). To mimic the gonadotropin effect on Kþ activity, the oocytes’ Kþ content was raised by injection of KCl into isolated oocytes in paraffin oil and the Kþ activity measured by microelectrodes. This treatment mimicked the influence of gonadotropin on both the rate of protein synthesis and the synthesis of specific polypeptides. The findings suggested that gonadotropin-stimulated oocyte growth is attributable largely to the hormone’s influence of transfollicular Kþ fluxes and supported the hypothesis that the changes in Kþ activity are critical for subsequent increases in protein synthesis and growth. Further analysis of the results, however, showed that there was a greater effect of Kþ on translation in vivo than was expected from its behavior on an in vitro translation system (Horowitz and Tluczek, 1989). It was also found that Naþ influences protein synthesis in the oocyte to the same extent as Kþ does, but Naþ has no significant effect in cell-free translation systems. This indicated that a mechanism exists through which either Kþ or Naþ can influence translation in the intact cell but is missing or degraded in cell-free systems. Experiments showed that both Kþ and Naþ are buffered in the cell, meaning they compete with other cations for high-affinity anions such as those associated with membranes. As a consequence, an increase in Kþ or Naþ will cause other cations to become less associated with fixed ionic sites and therefore increases the activity of the cations. If one cation among those newly displaced is more potent in its influence on translation than is Kþ, this exchange or buffering reaction could be the direct effector linking gonadotropin to its metabolic effects. Injection of Ca2þ into oocytes over a very wide range of concentrations had no significant effect on protein synthesis, nor did injection of EGTA, a chelating agent for Ca2þ, have any effect on the increase of protein synthesis due to Kþ or gonadotropin, indicating that Ca2þ is not the downstream effector for the activational increase in oocyte translation (Horowitz and Tluczek, 1989). In contrast to Ca2þ, however, injection of increasing concentrations of Mg2þ at first greatly stimulated and then at higher concentrations sharply inhibited protein synthesis. The function resembled those

Mg2þ in Cell Growth Regulation and Transformation

25

obtained by injecting Kþ and Na2þ except for the much lower and narrower range of Mg2þ concentrations required to obtain the full stimulatory and inhibitory response. The optimal concentration of free Mg2þ was about 4 mM, which is similar to the uncorrected Mg2þ requirement for in vitro synthesis of protein (Schreier and Staehelin, 1973). Isotherm data suggested that the oocyte contains exchange sites occupied chiefly by Kþ and Mg2þ. As mentioned earlier, the increase in protein synthesis from increasing Kþ is much greater than expected from the behavior of cell-free systems and the response to Mg2þ is more intense than that to Kþ. It was therefore hypothesized that increasing Kþ causes Mg2þ to dissociate from intracellular sites for which both compete, thereby increasing free Mg2þ and first stimulating, then inhibiting, translation with higher concentrations of Mg2þ. The hypothesis was put to the test by injecting EDTA, which chelates both Mg2þ and Ca2þ, to see if it prevents Kþ from stimulating protein synthesis. Injection of 0.3 mM EDTA reduced protein synthesis to the quiescent cell level. Since EGTA, which is a highly specific chelator of Ca2þ, had no effect on protein synthesis, it was concluded that Mg2þ is itself the intracellular effector controlling translation rates. Injection of EDTA also inhibited the gonadotropin-induced increase in protein synthesis. About 3 times more EDTA was required to inhibit the hormone-induced increase in protein synthesis than that induced by Kþ. The reason for the difference was proposed to be that gonadotropin stimulation was done in a salt medium that contained Mg2þ, which can then enter the oocyte, whereas Kþ stimulation was done in paraffin oil in which Mg2þ cannot enter the oocyte. Horowitz and Tluczek (1989) noted that this hypothesis had already been strongly argued as the basis for the regulation of protein synthesis and proliferation of somatic cells of vertebrates (Rubin, 1975a; Rubin et al., 1979; Sanui, 1970; Terasaki and Rubin, 1985). They also remarked that the Mg2þ activity of the dormant oocyte (0.3 mM) is 80 times less than expected at electrochemical equilibrium, assuming a membrane potential of 60 mV, which indicates that the oocyte, like other cells, has a transport system that actively extrudes Mg2þ. Since gonadotropin activation is followed by long-term growth and a continuous high rate of protein synthesis, translational control by Mg2þ implies that the activational increase in its activity must be maintained during postactivational growth. This suggests that the original increase of free Mg2þ brought about largely by Kþ-driven dissociation of Mg2þ from endogenous binding sites is likely to be accompanied by a down-regulation of Mg2þ active transport to maintain postactivational growth. How gonadotropin acts to bring about these changes remains to be determined, but it may have to do with cation exchanges initiated by perturbation of the cell membrane as suggested for stimulation of somatic cells by growth factors (Sanui and Rubin, 1978, 1982b).

26

Harry Rubin

X. Mg2þ EFFECTS ON DIVERSE CELLULAR RESPONSES TO GROWTH FACTORS The addition of serum or other growth stimulatory agents to inhibited cell cultures accelerates a number of early responses (Rubin and Fodge, 1974; Rubin and Koide, 1975). These include the uptake of 2-deoxyglucose, uridine, amino acids, and choline as well as the production of lactic acid. Although the increased uptake of these materials is not required for the onset of DNA synthesis and does not require protein synthesis, they are considered characteristic features of the coordinate response of cells to growth factors. If Mg2þ is the primary intermediary in converting membrane perturbation into a growth response, it might be expected to stimulate the other associated reactions. The deprivation of Mg2þ by its chelation with sufficient inorganic pyrophosphate was found to inhibit the uptake of 2-deoxyglucose and production of lactic acid as well as the synthesis of protein, RNA, and DNA (Rubin, 1975a) and the uptake of uridine (Rubin, 1976). The rate of uptake of glucose analogs in cultured cells depends on the rate of transport rather than the rate of phosphorylation (Bowen-Pope and Rubin, 1977). Removal of serum from the medium or deprivation of Mg2þ reduces the transport of the glucose analogs and does so in both cases by reducing the Vmax but not the Km of the reaction. This is consistent with a serum-induced increase in cellular Mg2þ as the driving force in the increased transport of glucose into cells. The rate of uridine uptake by contrast depends on its phosphorylation by uridine kinase (Plagemann et al., 1978; Rozengurt et al., 1977). The serum stimulation of uridine is mimicked by increasing the cellular concentration of Mg2þ, and is blocked by depleting cells of their Mg2þ (Bowen-Pope and Rubin, 1977; Vidair and Rubin, 1981). Like changing the concentration of serum in the medium, directly altering the concentration of Mg2þ in cells affects the Vmax of the uridine uptake system with little change in the Km (Bowen-Pope and Rubin, 1977). Unlike uridine, however, the uptake of thymidine is unaffected by serum treatment (Vidair and Rubin, 1981). This can be explained by the observation in cell-free extracts that the requirement of thymidine kinase for Mg2þ is less than one-tenth that of uridine kinase. It indicates that the Mg2þ level in quiescent cells is high enough to support maximal activity of thymidine kinase, but uridine kinase requires a substantial increase of Mg2þ to effectively increase the uptake of uridine upon addition of serum (Vidair and Rubin, 1981). Neither Ca2þ, Kþ, or Naþ influences the increased uptake of glucose analogs or uridine by growth factors, which adds to the support of an increase of free Mg2þ as the mechanism of these effects, as well as other reactions of the coordinate response. Perhaps the most convincing demonstration of the role of Mg2þ in

Mg2þ in Cell Growth Regulation and Transformation

27

regulating the phosphorylation of uridine in cells is that raising the intracellular concentration of Mg2þ high enough in the presence of the divalent cation ionophore A23187 and very high extracellular Mg2þ increases uridine uptake in Balb/c 3T3 cells to the same extent as addition of a growth factor (Vidair and Rubin, in press).

XI. POSSIBLE ROLES OF Kþ, Ca2þ, pH, AND Naþ IN GROWTH REGULATION A. Potassium Early changes induced by growth factors have been reported in the uptake and intracellular content of cations other than Mg2þ and their possible involvement in growth regulation. Serum or purified growth factors rapidly stimulate ouabain-sensitive Naþ/Kþ-ATPase activity in Swiss 3T3 cells or CEF as measured by the uptake of 86rubidium, or by enzyme assay (Rozengurt and Heppel, 1975; Smith, 1977). Treatment with ouabain, which inhibits the Naþ/Kþ-ATPase, interferes with the accumulation of Kþ and, if the reduction of Kþ in the cells is severe enough, prevents the onset of DNA synthesis. If intracellular Kþ in human fibroblasts is reduced to 60% of control, the rate of protein synthesis is decreased by no more than 20% from control values (Ledbetter and Lubin, 1975). As Kþ is further reduced to 30% of control, rates of DNA synthesis are maximally reduced without affecting RNA synthesis (Ledbetter and Lubin, 1977). The increase in Kþ uptake after serum stimulation of quiescent 3T3 cells resulted in a 75% increase in cell Kþ on a per mg protein basis, or a 40% increase on a per volume basis (Tupper et al., 1977). This increase peaked at 4 to 5 hr and declined steadily to initial levels at 10 to 14 hr, which is about the time the S period begins in the Balb/c 3T3 cells used. However, my laboratory found a rise of Kþ of only between 3 to 12% on a per mg protein basis in different experiments with serum stimulation of quiescent confluent Balb/c 3T3 cells (Sanui and Rubin, 1982a). In CEF stimulated by insulin, there was only a 14% increase in Kþ from quiescence to maximal response (Sanui and Rubin, 1978). The latter levels of change in Kþ per cell would, according to the earlier reports (Ledbetter and Lubin, 1975, 1977), have produced no change in protein synthesis. To further explore the role of Kþ in regulating protein and DNA synthesis, CEF were incubated in varying concentrations of Kþ, from the physiological level of 5.0 mM down to almost 0.4 mM for 16 hr in the presence of insulin (Moscatelli et al., 1979). An 80% reduction of intracellular Kþ in the CEF produced no change in the rate of DNA synthesis and a 90% reduction of

28

Harry Rubin

intracellular Kþ produced less than a twofold reduction in DNA synthesis. Hence, changes in intracellular Kþ concentrations within the physiological range cannot account for changes in DNA synthesis from quiescence to stimulation by growth factors. Since the reduction in Kþ uptake and cellular content in most of the previous experiments (Ledbetter and Lubin, 1975, 1977; Rozengurt and Heppel, 1975; Smith, 1977; Tupper et al., 1977) was induced by ouabain to inhibit the Naþ/ Kþ pump, but these experiments did not include measurement of divalent cations, a study was undertaken of the effects of ouabain on the major intracellular cations and synthesis of protein and DNA synthesis in Balb/c 3T3 cells (Sanui and Rubin, 1979a). At 1 hr after serum stimulation, there was a large decrease in the ouabain-treated cultures of Kþ and an increase in Naþ, with no change in Mg2þ and no reduction of protein synthesis. Beyond 1 hr, the rate of protein synthesis was significantly depressed in the ouabaintreated cultures, the value at 17 hr being 40% of control. In addition to maintaining the large changes in Kþ and Naþ seen at 1 hr, however, there were significant reductions in Mg2þ, the average at 17 hr being 85% of control, and DNA synthesis was reduced more than 10-fold. Since there was no change in cellular Ca2þ, the implications were that the large reductions in Kþ as seen at 1 hr had no effect on protein synthesis, but the later reduction in Mg2þ did reduce protein synthesis and could account for the later reduction in DNA synthesis. In another laboratory using Balb/c 3T3 cells, serum growth factors stimulated equal rates of entry into the S phase in both the presence and the absence of 100 M ouabain (Frantz et al., 1981). Since there was a decrease of intracellular Kþ in the ouabain-treated cultures, it was concluded that an increase of intracellular Kþ is not required for entry into S phase, and serum growth factors do not regulate cell growth by altering intracellular Kþ. This interpretation was countered by experiments with Swiss 3T3 cells in which intracellular Kþ of quiescent cultures was altered by reducing extracellular Kþ and stimulating the cells with purified peptide growth factors instead of serum (Lopez-Rivas et al., 1982). In that serum-free medium, the intracellular Kþ content was close to the threshold required to allow a mitogenic response, which implies that any reduction in intracellular Kþ would reduce the rate of onset of DNA synthesis. After 20 hr in the medium with peptide growth factors, the rate of DNA synthesis was very low even in those cultures with a physiological concentration of 5 mM Kþ in the medium, at a time when DNA synthesis in serum-stimulated cultures would have reached a maximum (Frantz et al., 1981). By 40 hr, there was about a sixfold increase in DNA synthesis in all cultures in the serum-free medium with the 3 highest concentrations of extracellular Kþ, even though the intracellular Kþ concentrations were unchanged from 20 hr, indicating either that the cells were under stress in the serum-free medium even in a physiological concentration of Kþ, or that the peptide growth factors produced a much

Mg2þ in Cell Growth Regulation and Transformation

29

slower progression into the S period than did serum. One of the coauthors of the paper related that the combined growth factors would not support proliferation of the cells enough to form a colony (E. Adelberg, personal communication). Between 20 and 40 hr, the cells in the lower concentrations of extracellular and therefore lower intracellular Kþ not only increased their rates of DNA synthesis but their intracellular concentrations of Kþ as well. The cellular Kþ at all reduced extracellular concentrations of Kþ continued to increase to 60 hr, indicating that there was a time-based adaptation of the cells to low Kþ medium. Since the rate of protein and DNA synthesis at 40 hr decreased with intracellular Kþ, it was suggested that a small change in cellular Kþ can influence the ability of the cells to initiate DNA synthesis in a serum-free medium. It was also stated, however, that the 3T3 cells stimulated by a combination of serum and platelet-derived growth factor are much less sensitive to a reduction in cellular Kþ than cultures stimulated by the purified polypeptides in a serum-free medium. The implication of this statement as well as the results of the foregoing papers using serum (Frantz et al., 1981; Moscatelli et al., 1979; Sanui and Rubin, 1979a) is that the dependence of protein and DNA synthesis on cellular Kþ is only expressed under suboptimal conditions of growth in which the Kþ level of cells in physiological external Kþ was set close to the threshold for a mitogenic response. This condition may therefore resemble that of Xenopus oocytes, in which the synthesis of protein can be accelerated by the addition of Kþ, which displaces Mg2þ from bound sites such as those on membranes to increase free cytosolic Mg2þ and accelerates the synthesis of proteins (Horowitz and Tluczek, 1989). The rate of protein synthesis is far more sensitive to change in Mg2þ than to Kþ in vitro (Brendler et al., 1981; Ilan and Ilan, 1978; Schreier and Staehelin, 1973) and in vivo (Moscatelli et al., 1979; Rubin et al., 1979; Terasaki and Rubin, 1985). It is noteworthy that no measurement was made of cellular Mg2þ in the growth factor experiment in Swiss 3T3 cells (Lopez-Rivas et al., 1982) so the possibility cannot be ruled out that a reduction in cellular Mg2þ, not in cellular Kþ, accounts for the reductions in protein and DNA synthesis, as was apparently the case in ouabain-treated 3T3 cells (Sanui and Rubin, 1979a).

B. Calcium Very large reductions in extracellular Ca2þ (from 1.5 to less than 0.01 mM) failed to inhibit proliferation of CEF in serum, but complete omission of Ca2þ from the medium did so (Balk et al., 1973). CEF in plasma, which lacks the potent growth factor PDGF, multiply more slowly than in serum and are more sensitive to depletion of Ca2þ in the medium. CEF transformed by Rous sarcoma virus (RSV) were more resistant

30

Harry Rubin

to omission of extracellular Ca2þ in either serum or plasma. Sensitivity to deprivation of extracellular Ca2þ varied greatly in an unspecified line of 3T3 cells grown in serum-containing medium (Boynton et al., 1974). Primary and secondary cultures of rat heart cells cultured in homologous plasma were arrested in the G1 phase in very low (0.02 mM) Ca2þ, but were not arrested in heterologous (calf) serum (Swierenga et al., 1976). From these and related experiments it was concluded that Ca2þ is a major regulator of cell proliferation in vertebrates. No measurement was made of intracellular Ca2þ in these experiments, so there was no evidence of its reduction when the extracellular concentration was drastically reduced. The addition of the specific Ca2þ-chelating agent EGTA in amounts beyond the concentration of Ca2þ in serum-containing medium inhibited the entry of CEF into the S phase of the cell cycle 5- to 10-fold, but actually increased the intracellular concentration of Ca (Moscatelli et al., 1979). The growth inhibitory concentrations of EGTA lowered intracellular Mg2þ about 10% and Kþ about 25% while it increased Naþ almost 300%. Simple decrease of Ca2þ in the medium, without chelation, from 1.7 to 0.01 mM Ca2þ did not significantly inhibit DNA synthesis. Further decreases in external Ca2þ decreased DNA synthesis threefold to fourfold but had no effect on intracellular Ca2þ; Mg2þ and Kþ were decreased and Naþ increased. Decreasing Mg2þ in the Ca2þ-deprived medium further reduced DNA synthesis as well as reducing intracellular Mg2þ and Kþ while further increasing Naþ. Stimulation of quiescent CEF with insulin or Balb/c 3T3 cells with serum slightly lowered intracellular Ca2þ and markedly lowered surface-bound Ca2þ (Sanui and Rubin, 1978, 1982a). As noted earlier, these treatments raised intracellular Mg2þ 22% in CEF and 15% in Balb/c 3T3 cells, but lowered externally bound Mg2þ to a lesser degree than externally bound Ca2þ. In later experiments with CEF in lower serum concentrations, 100-fold reduction of external Ca2þ inhibited cell proliferation but did not reduce DNA synthesis (Rubin and Chu, 1978). More severe Ca2þ depletion did reduce the rate of DNA synthesis in the CEF, but it also caused retraction of the cells from the dish and from each other, leading to a distinctly abnormal, rounded appearance. The results suggest that the effects produced by lowering Ca2þ in the medium are caused by its removal from the external surface of the cell. In contrast, reduction of Mg2þ in the medium below 0.8 mM reduced DNA synthesis and proliferation coordinately in CEF. The cells were flattened out in low Mg2þ resembling the cells in low serum concentration, and were quickly restored to a high rate of proliferation and DNA synthesis by replenishment of Mg2þ (Rubin and Chu, 1978). The strong inhibitory effect on DNA synthesis of reducing external Ca2þ to 0.03 mM in quiescent, confluent Balb/c 3T3 cells is reversed by

Mg2þ in Cell Growth Regulation and Transformation

31

raising external Mg2þ from 1 to 15 mM (Rubin et al., 1978). Raising the Mg2þ concentrations above 20 mM caused a marked inhibition of DNA synthesis in the Ca2þ-deprived cultures, giving a bell-shaped curve for DNA synthesis as a function of Mg2þ similar to that of protein synthesis in vitro (Brendler et al., 1981; Ilan and Ilan, 1978; Schreier and Staehelin, 1973). The inhibition of confluent 3T3 cells in very low Mg2þ was unaffected by large increases of Ca2þ. The results support the thesis that inhibitory effects of Ca2þ deprivation on cells are indirect and are caused by a reduction in the availability of Mg2þ. It is noteworthy that the deprivation of neither Ca2þnor Mg2þ in physiological concentrations of the other inhibits DNA synthesis in sparse cultures of 3T3 cells, which suggests that the free Mg2þ levels in exponentially growing cells is higher than that in quiescent confluent cells. Since DNA synthesis in quiescent confluent 3T3 cultures shows no response to variations in external Ca2þ or Mg2þ for about 10 hr after restoring physiological concentrations of the cations, their effect on DNA synthesis must be considered to be indirect. To determine what the direct influence of the cations on cell function is, a study was made on the uptake of uridine into the acid-soluble pool which responds immediately to the addition of serum growth factors (Bowen-Pope et al., 1979). Combined drastic reduction of both external Ca2þ and Mg2þ lowers uridine uptake to the same extent as the omission of serum from the medium. The rate-limiting step in the uptake of uridine is its phosphorylation and that is controlled by the availability of Mg2þ (Vidair and Rubin, 1981). The requirements for serum and for Mg2þ in uridine uptake are considerably less than are their requirements for the initiation of DNA synthesis. The very low rate of uridine uptake in cells deprived of Ca2þ and Mg2þ for 2 hr (Bowen-Pope et al., 1979) is unaffected by restoration of large amounts of Ca2þ. When hypernormal concentrations of Mg2þ are added, however, uridine uptake is quickly increased to control levels, indicating that Mg2þ is the regulatory agent for this reaction. The role of low Ca2þ in reducing uridine uptake is apparently to increase the permeability of the cells to Mg2þ so that its intracellular concentration is highly responsive to its external concentration. Further evidence of the increase in cell permeability in very low Ca2þ is the switch of intracellular concentrations of Kþ and Naþ, and particularly a marked increase in the rate of uptake of L-glucose which enters the cell only by unmediated diffusion. The increased uptake of L-glucose in low Ca2þ occurs even in physiological concentrations of Mg2þ, but is greatly reinforced by lowering external Mg2þ. In contrast, drastic lowering of Mg2þ in physiological Ca2þ has no effect on the uptake of L-glucose. Hence, Mg2þ acts directly as the regulator of uridine phosphorylation by uridine kinase (Vidair and Rubin, 1981) and, as at higher Mg2þ concentrations, acts indirectly as the regulator of the initiation of

32

Harry Rubin

DNA synthesis. Since onset of DNA synthesis reflects the rate of protein synthesis through the G1 period, it can be assumed that the optimal Mg2þ requirement for the attachment of messenger RNA to ribosomes is much higher than it is for uridine phosphorylation. It might be said that the failure to detect an increase in total cellular Ca2þ in growth-stimulated cells does not rule out an increase in free Ca2þ since the basal level is 4 orders of magnitude lower than total Caþ. And indeed, the addition of serum to human fibroblasts produces an immediate rise of free Ca2þ that peaks at about a twofold increase in 30 sec, then rapidly declines and stabilizes at a new steady level in about 3 min (Moolenaar et al., 1984). However, several different growth factors fail to produce a transient rise of Ca2þ in human fibroblasts, and EGF produces no such rise in myocytes (Grubbs, 1991). In any case, no ‘‘trigger’’ reaction in the literal sense, as presumably represented by the early Ca2þ transient, can account for the requirement of the continuous presence of growth factors during most, if not all, of the G1 period to maintain the increased rate of protein synthesis that underlies a speedup of cell proliferation. Also relevant is the observation that the direct injection of a wide range of Ca2þ concentrations into Xenopus oocytes has no significant effect on protein synthesis, nor does the Ca2þ-chelating agent EGTA prevent the increase in protein synthesis induced by Kþ or hormone treatment (Horowitz and Tluczek, 1989). Therefore, the Ca2þ transient induced by some growth factors is no more essential for protein synthesis or the onset of DNA synthesis than are other early events such as increased transport of glucose or uptake of uridine. Those early events that are essential for growth are, of course, increased protein synthesis and the utilization of glucose in energy metabolism (Fodge and Rubin, 1975b; Rubin and Fodge, 1974). There is one final piece of evidence that has been cited to support a major role of Ca in regulation of DNA synthesis. The addition of hypernormal concentrations of Ca2þ to quiescent cultures of Balb/c 3T3 cells induces DNA replication (Boynton and Whitfield, 1976; Dulbecco and Elkington, 1975). However, the effect depended on the concentration of inorganic orthophosphate in the medium and was associated with the formation of insoluble complexes of Ca2þ with HPO2 (Rubin and 4 Sanui, 1977). It could be simulated by raising the concentration of HPO2 4 in normal concentrations of Ca2þ which resulted in precipitate formation. It could be prevented in hypernormal concentrations of Ca2þ by lowering HPO2 4 to levels that did not form precipitates. A more striking stimulation of DNA synthesis was produced by adding very small amounts of inorganic pyrophosphate to the medium, just enough to form floccules with Ca2þ which interacted with the cell surface (Bowen-Pope and Rubin, 1983; Rubin and Sanui, 1977). Strontium can partly replace Ca2þ in these effects. The increase of DNA synthesis in hypernormal Ca2þ therefore

Mg2þ in Cell Growth Regulation and Transformation

33

has nothing to do with an increase of intracellular Ca2þ but results from the action of Ca3(PO4)2 precipitate at the cell surface. Although intracellular Ca2þ plays no direct role in cell growth regulation, extracellular Ca2þ may do so because it is essential in forming adhesions between cells (Vleminckx and Kemler, 1999). The release of Ca2þ from the external surface of the cells by growth factors (Sanui and Rubin, 1978, 1982a) may therefore contribute to the stimulation of contact-inhibited cells by loosening intercellular contacts.

C. pH and Sodium The rate of cell proliferation in culture depends on the pH of the medium between pH 6.6 and 8.0 for CEF, with an optimum between pH 7.2 and 7.6 (Rubin, 1971b). The optimum pH varies with different cell lines (Ceccarini and Eagle, 1971). Cells are less susceptible to inhibition at pH below 7.0 at low population density than at high population density, and will reach a saturation density at pH 6.5 before they are fully confluent (Rubin, 1971b). It requires about 3 times as high a concentration of protons to inhibit a sparse population of cells as a dense population (Rubin, 1971a). There is estimated to be 2- to 10-fold higher proton concentration at the surface of the plasma membrane than in the bulk medium, and it increases with negativity of the electrostatic surface potential. Since the negative surface potential increases as the distance between cells decreases, the surface proton concentration is higher in a dense population than in a sparse population, and to that extent, lowered pH may contribute to density-dependent or contact inhibition of cells. The growth rate of normal cells is also proportional to their rate of migration which depends on membrane motility, which, in turn, reduces the local proton concentration (Rubin, 1971a,b). Surface proton concentration presumably influences intracellular pH and this, in turn, affects the rate of glycolysis in cells. A major enzymatic point of control of the glycolytic pathway is phosphofructokinase which is activated by increasing pH of the medium or the concentration of serum (Fodge and Rubin, 1973). Phosphofructokinase also exhibits a bell-shaped curve of dependence on Mg2þ in cell-free preparations, with an optimum in the same range as that of protein synthesis (Garner and Rosett, 1973). Metabolic inhibitors of glycolysis strongly and irreversibly inhibit the initiation of DNA synthesis, but the effect on ongoing DNA synthesis is minimal (Fodge and Rubin, 1975b; Rubin and Fodge, 1974). These observations indicate that intracellular pH might play an auxiliary role in the regulation of cell proliferation.

34

Harry Rubin

The addition of a combination of growth factors (insulin, vasopressin, and PDGF) to quiescent Swiss 3T3 cells resulted in an average increase of cytoplasmic pH of 0.16 units (Schuldiner and Rozengurt, 1982). The increase required external Naþ or Liþ and could not be substituted by Kþ or choline. The half maximal effect of Naþ in the medium for the increase in pH was 38 mM and was attributed to Naþ/Hþ antiport in the cells since it was blocked by amiloride, an inhibitor of the Naþ/Hþ ATPase. The influx of Naþ was thought to be sufficient to activate the Naþ/Kþ pump and increase Kþ in the stimulated cells, thereby contributing to mitogenesis. Doubt was raised about this interpretation because amiloride also directly inhibits protein synthesis, which is essential for the onset of DNA synthesis (Lubin et al., 1982). Further investigation revealed that the decrease of Naþinhibited DNA synthesis by different mechanisms depended on the growth factor used to stimulate the cells and the monovalent cation used as an osmotic substitute for Naþ (Burns and Rozengurt, 1984). The decrease in DNA synthesis was correlated with a decrease in intracellular Kþ in some cases, and the blocking of cellular alkalinization in others. The latter effect occurred in the stimulation of early passage human fibroblasts (Moolenaar et al., 1983), and with a variety of mitogens in mouse thymocytes and Swiss 3T3 cells (Hesketh and Moore, 1985). The increase in cytoplasmic pH was, in most cases, preceded by a transient rise in Ca2þ, but the increased pH persisted for more than 25 min (Hesketh and Moore, 1985). It was thought likely that the activation of Naþ/Hþ exchange is necessary to proceed through the cell cycle but by itself is insufficient to ‘‘trigger cell division’’ (Glaser et al., 1984). As already noted, mention of a trigger for cell division raises questions of its own since it is generally agreed that the growth stimulus has to be maintained through most, if not all, of the G1 period in order to hasten DNA synthesis (Bolen and Smith, 1977; Brooks, 1977; Rubin and Steiner, 1975). The notion of necessity of elevated pH for mitogenesis has to be qualified because the substitution of 95% of Naþ in the medium by choline does not inhibit the increase of DNA synthesis in quiescent CEF stimulated by insulin, although substitution by Kþ profoundly suppresses DNA when external Naþ is decreased below 30 mM (Moscatelli et al., 1979). Another caveat is that the stimulation of myocytes by EGF does not raise cytosolic pH (Grubbs, 1991). It should also be borne in mind that decrease in Kþ of the medium suppresses the onset of DNA synthesis in Swiss 3T3 cells much less when they are stimulated by serum than by purified growth factors (Frantz et al., 1981; Lopez-Rivas et al., 1982), which illustrates the contingency of the effects of Kþ, Naþ, and pH on the cell cycle. This contrasts with the role of Mg2þ in mediating growth control since any alteration of intracellular Mg2þ alters the rate of protein synthesis (Terasaki and Rubin,

Mg2þ in Cell Growth Regulation and Transformation

35

1985) and thereby progression through G1 to DNA synthesis (Brooks, 1977; Castor, 1977).

XII. REGULATION OF PROTEIN SYNTHESIS BY THE PI 3-K AND mTOR PATHWAYS Most of what has been presented here defined the basic physiological parameters of growth regulation which were revealed in the period from the 1960s to 1985, with a few important accretions about the role of Mg2þ extending to 1991. This era did not include studies of molecular pathways of growth regulation, which began in earnest in the late 1980s and continues to this day. The most relevant of the molecular studies to stimulation by growth factors involved the response of protein synthesis to polypeptide growth factors, such as insulin, EGF, and PDGF (Alberts et al., 2002). Binding of the growth factors to the extracellular domain of their receptors (Fig. 7A) causes a conformational change which leads to their lateral movement in the membrane to form oligomers which then autophosphorylate on tyrosine in their cytoplasmic domain (Fig. 7B) (Schlessinger, 1988; Ullrich and Schlessinger, 1990). These phosphorylations entrain a cascade of serine/threonine kinases combined with inhibitory proteins in the PI 3-K pathway that lead to phosphorylations of a large protein, the mTOR kinase (Fig. 7C) (Richardson et al., 2004). mTOR phosphorylates S6 kinase which phosphorylates the S6 protein of 40S ribosomal subunit (Fig. 8A) to drive the initiation of protein synthesis on 50 TOP mRNAs (Fig. 8B) (Schmelzle and Hall, 2000). mTOR also phosphorylates the binding protein 4E-BP1 which dissociates from and activates the initiating factor eIF-4E (Fig. 8A) (Beretta et al., 1996). eIF-4E is the rate-limiting factor in formation of a larger complex, eIF–4F, which initiates protein synthesis on mRNA (Fig. 8B) (Richardson et al., 2004). The overall effect of mTOR phosphorylations of S6K and 4E-BP1 is an increase in the efficiency of translation initiation, which is a characteristic effect of the stimulation of protein synthesis by growth factors (Stanners and Becker, 1971), especially for the synthesis of ribosomal proteins (DePhilip et al., 1980) and elongation factors (Gingras et al., 2004; Terada et al., 1994). The previously stated considerations place the mTOR phosphorylations of S6 kinase and 4E-BP1 in a crucial position for the regulation of protein synthesis. The kinase activity of mTOR for these two targets is distinguished from most protein kinases analyzed to date by its high Km (Michaelis content) for ATP, i.e., slightly greater than 1.0 mM (Dennis et al., 2001), versus 10 to 20 M for the other protein kinases (Edelman et al., 1987). The marked reduction of ATP concentration by inhibiting glycolysis with

Fig. 7 Simplified model of the activation of the PI 3-K signaling pathway by binding of insulin to its receptors, accompanied by membrane perturbation and the partial release of membrane-bound Mg2þ. (A, B) The plasma membrane receptor for insulin is a tyrosine kinase. The binding of insulin to the external domain of the receptors causes their conformational change. They then move to form oligomers and phosphorylate each other on their intracellular domains. The conformational change and movement of the receptors perturb the membrane and weaken its binding for Mg2þ, which results in its partial release into the cytosol to increase the concentration of MgATP2. (C) The autophosphorylation of the receptors activates the PI 3-K pathway, which activates a kinase cascade leading to phosphorylation of mTOR (Rubin, in press).

Mg2þ in Cell Growth Regulation and Transformation

37

Fig. 8 Phosphorylation of mTOR and the increase in the initiation of protein synthesis. (A) Phosphorylation of mTOR and the increase in MgATP2 increases the phosphorylation of S6 kinase which activates the S6 protein of the 40S ribosomal subunit. mTOR also phosphorylates 4E-BP1 bound to eIF-4E initiation factor, which releases the latter. (B) The phosphorylated S6 and released eIF-4E (combined with other initiation factors) increase the frequency of initiation of protein synthesis on 50 TOP mRNA (Rubin, in press).

2-deoxyglucose, an analog of glucose, strongly inhibits key phosphorylations by mTOR of S6 kinases and 4E-BP1 (Dennis et al., 2001). It was therefore inferred that ATP is the regulatory factor that determines the initiation rate of ribosomal protein synthesis in cells stimulated by insulin. There are several inconsistencies in assigning such a regulatory role to ATP in the paper that made the proposal (Dennis et al., 2001). The omission of

38

Harry Rubin

stimulation by insulin in control cultures reduced ATP by less than 10% but abolished detectable phosphorylations in S6 kinase and 4E-BP1; a mitochondrial inhibitor reduced ATP 28% but allowed considerable phosphorylation (Fig. 1 in Dennis et al., 2001). Stimulation of CEF with serum, which is a stronger mitogen than insulin, actually decreases ATP concentration (Fodge and Rubin, 1973). However, the true substrate of phosphoryl transfer reaction is MgATP2 rather than ATP itself (Lardy and Parks, 1956; Rose, 1968). The other forms of ATP such as ATP4, KATP3, and HATP3, which constitute a significant fraction of total ATP, are inhibitors of phosphoryl transfers (Achs et al., 1982, 1979). Variations in ratio of Mg2þ and ATP exert profound effects on the velocity of the kinase reactions of carbohydrate metabolism (Garner and Rosett, 1973) and are likely to act in a similar fashion in the mTOR phosphorylations since both have a similar Km for MgATP2. The Km of ATP for the mTOR phosphorylations was determined in vitro in a large excess (10 mM) of Mg2þ (Dennis et al., 2001) and is likely to be the Km for MgATP2. The free Mg2þ concentration in quiescent cultured mammalian cells is 0.2 to 0.4 mM (Grubbs, 1991; Ishijima et al., 1991), which is considerably lower than the 1 mM concentration of ATP (Gribble et al., 2000). Since free Mg2þ increases significantly in stimulated cells (Grubbs, 1991; Ishijima et al., 1991) in contrast to ATP (Fodge and Rubin, 1973), the level of MgATP2 would be determined by free Mg2þ. At the same time, it would also decrease the concentration of the inhibitory forms of ATP. The combined effect would increase mTOR phosphorylation of its two key substrates and thereby increase the frequency of the initiation of protein synthesis. In contrast, the other protein kinases of the PI 3-K pathway that have 50 to 100 times lower Km for MgATP2 would presumably be saturated with it in quiescent as well as stimulated cells, and their activity would be unaffected by changes in free Mg2þ, although stimulated by upstream phosphorylations of the PI 3-K pathway. The more general and more specific mechanisms would have to combine with each other in order to fully open the gates for increased protein synthesis.

XIII. ROLE OF CATIONS IN NEOPLASTIC TRANSFORMATION Efforts were made to determine whether neoplastic transformation changes the regulatory role of Mg2þ or Ca2þ in cells. The concentration of Mg2þ for half-maximal rate of proliferation of transformed human lung fibroblasts was about 20 times lower than the concentration required for normal diploid human lung fibroblasts (McKeehan and Ham, 1978). In contrast, there was no difference in the concentration of Ca2þ required for

Mg2þ in Cell Growth Regulation and Transformation

39

half-maximal rate of proliferation of the transformed and normal cells. The results indicated that for normal cells the role of Mg2þ is more proximal than Ca2þ to the intracellular events that determine maximal growth rate. The role of both cations for half-maximal growth rate of the normal cells was studied as a function of serum concentration and suggested that Ca2þ interacts more directly than Mg2þ with the serum molecules that stimulate proliferation of the normal cells. It was proposed that transformation causes a selective loss of the growth regulatory role of Mg2þ but not Ca2þ, and that the regulatory effect of Ca2þ in normal but not transformed cells is primarily mediated through Mg2þ-dependent processes. Hence, transformation is associated with the loss of ability to regulate to Mg2þ levels in the cell. Since the major process that regulates the onset of DNA synthesis is protein synthesis, and there is no indication that ribosomes of transformed cells differ from those of normal cells in their requirement for Mg2þ, it would appear that the transformed cells have lost their capacity to regulate Mg2þ, which depends, at least to some extent, on the binding capacity of membranes. The cellular content of the 4 major cations was measured in normal epidermis of control mice and in epidermis painted repeatedly with methylcholanthrene, a powerful carcinogen (Carruthers and Suntzeff, 1943). The Mg2þ content of the methylcholanthrene-treated epidermis was 20% higher than that of the controls, with no change in Naþ and Kþ but there was a 60% decrease in Ca2þ. There is an increase in Kþ in many tumors but that is related, at least in part, to an increase in the rate of proliferation as seen in normal tissue (deLong et al., 1950). Squamous cell carcinomas induced in mice by methylcholanthrene had only 20% of the Ca2þ of normal epidermis (Lansing et al., 1948). Three spontaneously transformed clones of Balb/c 3T3 fibroblasts had on average 15 to 20% more total Mg2þ than nontransformed fibroblasts, although there was widespread variation among the clones in degree of transformation and in Mg2þ content (Terasaki, 1983). The Ca2þ content of 12 human intestinal cancers was reduced an average of 44% compared with adjacent normal mucosa (deLong et al., 1950). The Ca2þ content of spontaneously transformed Balb/c 3T3 cells was only onethird that of nontransformed cells (Rubin et al., 1981). It appears then that a wide variety of neoplastic cells have a much lower capacity to retain Ca2þ than non-neoplastic cells. Ca2þ plays a major role in the mutual adhesion of cells to one another mainly through binding to proteins known as cadherins (Vleminckx and Kemler, 1999). The adhesiveness of cells to one another was long ago ascribed to mediation between cell surface proteins by Ca2þ and to a lesser extent by Mg2þ (Lansing et al., 1948; Zeidman, 1947). The adhesiveness of cells from human squamous cell carcinomas and a variety of adenocarcinomas is much lower than that of their normal counterparts (Coman, 1944; McCutcheon et al., 1948). The reduced adhesiveness may

40

Harry Rubin

result from a reduction in a major cadherin in some cases (Shimoyama et al., 1992; Umbas et al., 1992) but in others it could result from other causes like the destabilization of membranes and their loss of Ca2þ in response to oncogenes or other factors from within genetically altered cells. In view of the evidence that transformed cells have a diminished capacity to regulate their free Mg2þ, the effects of Mg2þ deprivation on their behavior were examined. Nontransformed and spontaneously transformed clones of Balb/c 3T3 cells were isolated and their quantitative requirement for Mg2þ to support proliferation was determined (Rubin, 1981). As was the case with normal and transformed human lung cells (McKeehan and Ham, 1978), the transformed Balb/c 3T3 fibroblasts had a much lower requirement of Mg2þ for proliferation than did the nontransformed cells. The transformed cells in physiological concentrations of Mg2þ did not have a true saturation density due to contact inhibition in conventional cultures because they depleted the medium of essential constituents at high population density even with daily medium changes. However, when external Mg2þ was reduced 60-fold, the rate of exponential growth of the cells was reduced slightly and they developed a true saturation density, i.e., the result of intercellular contact rather than depletion of essential medium components. After 3 days at saturation density, the cells resumed exponential growth. If subcultured at lower density before that happened, the cells flattened on the dish and simulated the appearance and regular arrangement of normal cells (Fig. 9). However, when they resumed multiplying, they

Fig. 9 Normalization of transformed cells by deprivation of Mg2þ. Growing cultures of a nontransformed and a transformed clone of Balb/c 3T3 mouse cells were maintained in medium with 10% calf serum and 1.0 mM Mg2þ. The medium was exchanged for fresh medium with (A) 1.0 mM Mg2þ for the nontransformed cells, and (B) 1.0 mM Mg2þ or (C) 0.01 mM Mg2þ for the transformed cells for 2 days. The nontransformed cells in (A) were well spread, flat, and tended to line up with each other. The transformed cells in 1.0 mM Mg2þ of (B) were retracted, thin, and randomly arranged. The transformed cells in 0.01 mM Mg2þ flattened out and lined up with each other in (C), resembling the nontransformed cells in (A) (Rubin, 1981).

Mg2þ in Cell Growth Regulation and Transformation

41

reverted to the thin, spiky appearance of transformed cells. The normal cells at low saturation density in very low Mg2þ also resumed multiplication after a week or so. In both cases, the resumption may have occurred by the adaptation to low Mg2þ through increasing their intracellular Mg2þ content (Rubin et al., 1979). Just as the transformed clone required much less Mg2þ for multiplication than the nontransformed cell, it also required much less serum (Rubin et al., 1981). Deprivation of Mg2þ raised the serum requirement of the transformed clone for DNA synthesis, especially at higher population densities, resulting in a behavior similar to that of nontransformed cells. None of these normalizing effects of lowering Mg2þ concentration could be reproduced by lowering Kþ or Ca2þ concentration, nor could either of the latter restore the normal appearance or arrangement of the cells. Addition of dibutyryl cyclic AMP partly flattened the transformed cells but did not induce either serum or density dependence. The Ca2þ content of the transformed cells was only one-third that of the nontransformed cells, but it was raised to the same level as the nontransformed cells by deprivation of Mg2þ. Another indication that Mg2þ deprivation restores the normalized phenotype of transformed cells is that it is much more effective in inhibiting DNA synthesis in crowded cultures than in sparse cultures (Rubin, 1981, 1982). In contrast, deprivation of Kþ or Ca2þ or treatment with cyclic AMP is more effective in sparse than in crowded, transformed cultures. Taken together, these observations support the view that a defect in Mg2þ regulation is a basic feature of neoplastic transformation and argue against a direct role for a defect in regulation by Kþ, Ca2þ, or by cyclic AMP. Highly transformed cells do not reach a saturation density at confluence under ordinary conditions of culture because they deplete the medium of essential constituents even when the medium is replenished daily. If they are attached to a small coverslip which is then placed in a dish with a large volume of medium, they achieve a multilayered saturation density, depending on serum concentration, that does not deplete the essential components of the medium (Rubin and Chu, 1982). At saturation density in physiological Mg2þ, the transformed cells took on the morphology and orderly arrangement of normal cells. Their rate of DNA synthesis was markedly decreased and total cellular Mg2þ was reduced by 33%. When they were subcultured at low densities, they at first exhibited the flattened appearance of isolated normal cells followed by a lag period of 10 to 13 hr before an increase in DNA synthesis began. The rate of DNA synthesis and percentage of cells labeled with 3H-thymidine were highly dependent on serum concentration, as would be expected of stimulated normal cells. The capacity of the density-inhibited transformed cells to produce colonies when trypsinized and suspended in agar was reduced 10-fold. Within 3 days after subculture, the cells resumed their transformed appearance and underwent a fivefold

42

Harry Rubin

increase of colony formation in agar. The self-normalization of the transformed cells at very high saturation density may be related to the normalization of individual, spontaneously transformed cells by contact with confluent normal cells (Stoker, 1964; Stoker et al., 1966; Weiss, 1970). It suggests that the inhibition of membrane activity in transformed cells at very high population density lowers free Mg2þ activity and restores normal appearance and behavior. It is important to note that the inhibition of protein synthesis in transformed cells by cycloheximide does not reverse the transformed phenotype (Terasaki and Rubin, 1981). Apparently, the fully transformed phenotype, including its characteristic morphology, depends on membrane/ Mg2þ activation of metabolic pathways in addition to protein and DNA synthesis. Those additional pathways would include energy metabolism in the form of glycolysis and the Krebs cycle (Achs et al., 1979, 1982; Garfinkel et al., 1979).

XIV. CONCLUSIONS Several biological features of cell growth regulation have been established that provide guidelines for critically analyzing the mechanism of the process. The first of these has to do with the specificity of the treatments used to stimulate or inhibit cell proliferation. Cells need growth factors, most commonly animal serum or certain polypeptide hormones, in order to stimulate multiplication. A considerable body of evidence shows that these act through combination with specific receptors on the plasma membrane. But there are also a number of nonspecific ways to stimulate cell proliferation that have mainly been demonstrated in confluent, contact-inhibited cultures. The most graphic demonstration of surface action is seen after scraping away part of the confluent, contact-inhibited layer, thereby allowing individual cells to migrate from the confluent region on the denuded surface, which allows them to multiply at a maximal rate. Other nonspecific agencies include treatment of CEF with trypsin and certain other proteases; subtoxic concentrations of heavier metals, i.e., Zn2þ, Cd2þ, Hg2þ, Mn2þ, and Pb2þ, on CEF; and addition to the medium of Balb/c 3T3 cells of 5 mM of Ca2þ, which forms precipitates with inorganic phosphate, or addition of 0.15 mM inorganic pyrophosphate which forms floccules with Ca2þ (Rubin and Sanui, 1977). The precipitates presumably interact with the lipid bilayer of Balb/c 3T3 plasma membranes to stimulate their proliferation (Bowen-Pope and Rubin, 1983) but they do not stimulate CEF. (When the pyrophosphate concentration exceeds that of Mg2þ, it is inhibitory to growth due to drastic lowering of Mg2þ by chelation.) Proliferation of CEF is inhibited by glucocorticoids (Fodge and Rubin, 1975a) or by suspending them in a semi-solid or viscous liquid medium.

Mg2þ in Cell Growth Regulation and Transformation

43

The converse of suspension is the requirement for normal cells to attach and spread on a solid substratum in order to multiply. Accumulated evidence indicates most, if not all, of stimulatory treatments perturb the cell membrane, and the inhibitory treatments stabilize it. A second biological feature of growth stimulation is the crucial relationship between increased protein synthesis and the acceleration of cells through the G1 period into the S period. The increase in protein synthesis begins immediately after the addition of serum or other growth factors and their action must be continued for hours through much or, in some cases, all of the G1 period to initiate DNA synthesis among a large population of cells. Indeed, removal of serum and lowering pH of CEF cultures during the exponential rise of DNA synthesis allows initiated cells to continue DNA synthesis but inhibits cells still in G1 from entering the S phase (Rubin and Steiner, 1975). Furthermore, a strong shutdown of protein synthesis with cycloheximide during the S period will shut down DNA synthesis (Kim et al., 1968). Thus, there is a continuing need for protein synthesis to maintain DNA synthesis. There are many other early responses to growth factors, such as increased uptake of hexoses and other metabolites, and accelerated transport of Kþ, Ca2þ, and Naþ but few, if any, of these are consistently required for accelerated onset of DNA synthesis. Exceptions are blockage of glycolysis or oxidative phosphorylation by metabolic inhibitors which inhibit the initiation but not the continuation of DNA synthesis (Rubin and Fodge, 1974). The most acute relation of an early response to later DNA synthesis is elevation of protein synthesis but, as stated, that must be maintained throughout the G1 phase to be effective. The primary problem of stimulating DNA synthesis then resolves itself into the manner by which perturbation of the cell membrane by growth factors or by nonspecific treatments translates into the stimulation of protein synthesis. It is common knowledge that protein synthesis depends on the availability of Mg2þ. The initiation of protein synthesis by attachment of messenger RNA to the small subunit of ribosomes requires a higher concentration of Mg2þ than elongation of the polypeptide chain (Revel and Hiatt, 1965). The concentration of free Mg2þ in the cell has been estimated as 1.0 mM or less (Rink et al., 1982), with most recent estimates in quiescent cells running between 0.2 and 0.4 mM (Grubbs, 1991; Ishijima et al., 1991). The optimal free Mg2þ concentration for initiation of protein synthesis in vitro is 2 to 4 mM, which ensures that any increase in free Mg2þ in cells will increase the rate of protein synthesis. Inoculation of rabbits with insulin more than doubles the fraction of a phosphoglucomutase in the active Mg2þ form in skeletal muscle and lowers the inactive Zn2þ form, indicating that the action of insulin on muscle cell membrane increases the level of free Mg2þ in the cytosol (Peck and Ray, 1971). Continuous treatment of confluent CEF or Balb/c 3T3 cells with insulin or serum, respectively, sharply increases the rate

44

Harry Rubin

of DNA synthesis after a lag of 4 or 10 hr (Sanui and Rubin, 1978, 1982a) in which the rate of protein synthesis remains at an elevated level. The total Mg2þ content of the cells increases 14 to 22% within an hour or two after treatment with growth factors and is maintained at the elevated level throughout the G1 and S periods (Sanui and Rubin, 1978, 1982a). It was assumed that the increase in total Mg2þ was accompanied by an increase in free Mg2þ. This assumption is supported by the finding of large changes in free Mg2þ with small variations in total cell Mg2þ in the hepatocytes of streptozotocin diabetic rats (Corkey et al., 1986). Direct evidence for a physiological role of increasing Mg2þ in initiating protein synthesis came from injection of Mg2þ into single Xenopus oocytes, which reproduced the physiological action of gonadotropin (Horowitz and Tluczek, 1989). Confirmatory evidence of a prolonged rise in free Mg2þ in somatic mammalian cells treated with growth factors was obtained with the presence in the cells of the Mg2þ-sensitive fluorescent indicator mag-fura 2 (Grubbs, 1991; Ishijima and Tatibana, 1994; Ishijima et al., 1991). An estimate can be made of the intracellular rise in Mg2þ resulting from its release from binding sites on the internal surface of the plasma membrane after insulin stimulation, based on the measurement of externally bound divalent cations released by insulin (Sanui and Rubin, 1978). CEF were repeatedly washed with a 0.25-M sucrose solution. The externally bound cations are displaced by Hþ in a 10-sec rinse with pH 4.0 medium. The results show that treatment of CEF with 0.1 unit of insulin reduces externally bound Mg2þ by 45% and externally bound Ca2þ by 34% (Sanui and Rubin, 1978). The insulin treatment displaces 1.37 nmoles of Mg2þ and 4.02 nmoles of Ca2þ per mg of cell protein. Using values for CEF of about 104 insulin binding sites per cell (Raizada and Perdue, 1975) and 4  106 cells per mg protein, about 5  105 nmoles of insulin can be bound per mg cell protein. From these figures, it is estimated about 105 divalent cations are displaced per insulin bound, which is orders of magnitude higher than the stoichiometric exchange with other cations or with EDTA or ATP (Sanui and Pace, 1967). Such large cation displacements indicate that insulin induces a conformational change of the membrane to which it binds. Shlatz and Marinetti had come to the same conclusion after finding that insulin markedly reduces the binding of Ca2þ to isolated rat liver plasma membranes and, conversely, that one molecule of hydrocortisone leads to the additional binding of 3000 atoms of Ca2þ (Shlatz and Marinetti, 1972). Significantly, insulin and cortisol, an analog of hydrocortisone, have opposite effects on DNA synthesis and other reactions of coordinate control in CEF (Fodge and Rubin, 1975a), and Mg2þ deprivation simulates quantitatively the addition of cortisol to CEF (Rubin, 1976). Accepting that insulin induces a conformational change in the plasma membrane, it should release approximately the same number of divalent

Mg2þ in Cell Growth Regulation and Transformation

45

cations from the inner surface of the plasma membrane as it does from the external surface of the cell. Since the association constant of Mg2þ to anionic sites on membranes is the same as that of Ca2þ (Sanui and Pace, 1967), and free Mg2þ is 3 to 4 orders of magnitude higher in concentration in the cytosol than Ca2þ, most of the divalent cations bound to the internal surface of the plasma membrane would be Mg2þ. Calculating the approximate number of Mg2þ cations released into the cytosol to be about 108 per CEF cell with an average volume of 1000 3 (Rubin and Hatie´ , 1968), there would be an approximate increase of 1.0 mM free Mg2þ per cell. This is within the range of the values reported from measurement of free Mg2þ by mag-fura 2 in cells stimulated by growth factors (Grubbs, 1991; Ishijima et al., 1991), and would account for the increase in protein synthesis needed to drive DNA synthesis and mitosis. The mag-fura 2 indicator was only reliable for 1 hr as it leaked from the cells, but in that time, it recorded a constant elevation in free Mg2þ, so it is plausible that the increase persisted through G1 and into S period in keeping with the increase in total Mg2þ that remained constant for at least 17 hr after the addition of insulin (Sanui and Rubin, 1978). More than 300 enzymes use Mg2þ as a cofactor and their activity displays the same bell-shaped curve for dependence on Mg2þ as protein synthesis does with a similar optimum (Ebel and Gu¨ nther, 1980; Garner and Rosett, 1973). That large catalog of enzymes includes phosphofructokinase which is a major control point in glycolysis and operates as such in serum-stimulated CEF (Fodge and Rubin, 1973). This parallelism implies that energy metabolism is coordinately controlled by Mg2þ to a similar extent as protein synthesis. It, of course, makes sense that the energy generation and synthetic activities of cells rise coordinately in response to growth stimulation to keep up with the increased demand for energy and maintain balanced growth. Similarly, early responses which are not needed for a single round of DNA synthesis may be part of a reserve for repeated rounds. That would be true for increased uptake of hexoses and other metabolites which are regulated by Mg2þ availability (Bowen-Pope and Rubin, 1977; Rubin, 1977), and cation pumps which are driven by Mg2þ-dependent ATPases (Lostroh and Krahl, 1973, 1974). A simplified model for the internal release of Mg2þ from membranes by mitogens and consequent effects on cellular metabolism and growth is shown in Fig. 10. It has been designated the MMM (Membrane, Magnesium, Mitosis) model of cell proliferation control (Rubin, in press). Transformed cells are much less subject to inhibition of growth by deprivation of Mg2þ than are nontransformed cells (McKeehan and Ham, 1978; Rubin, 1981). In very low Mg2þ, the transformed cells take on the appearance and regulatory behavior of normal cells, which includes increased sensitivity to contact inhibition, higher requirement for serum, and reduced capacity to form colonies when suspended in agar (Rubin et al., 1981). When grown to very high density on a coverslip in a large volume of

46

Harry Rubin

Fig. 10 Major features of a model for coordinate control of intermediary metabolism and growth by the availability of Mg2þ. Intracellular cations are bound to the internal surface of the plasma membrane. The relative amount of each ion that is bound depends on its abundance and affinity for fixed anionic groups in phospholipids, like phosphotidyl serine, and to a lesser extent to proteins. The affinity for divalent cations is determined in part by the closeness of the fixed anionic groups to allow two-point attachment of the divalent cation. An external stimulus like insulin that perturbs the membrane (Shlatz and Marinetti, 1972) extends the distance between the anionic groups and reduces affinity for the divalent cations (Dawson and Hauser, 1970), of which Mg2þ is the predominant intracellular species. The Mg2þ can also be displaced by heavy metals, such as Zn2þ, Cd2þ, Hg2þ, or Pb2þ, which stimulate cells at subtoxic doses (Rubin and Koide, 1973; Sanui and Rubin, 1984). The increase in availability of Mg2þ speeds up intermediary metabolism, which increases the supply of substrates, including nucleotides for biosynthetic reactions and other cellular activities, by changing the inhibitory K-ATP to Mg-ATP, which is the substrate for phosphorylation reactions. The increased free Mg2þ also increases the frequency of the initiation step for protein synthesis which requires higher [Mg2þ] than elongation of the polypeptide chain (Revel and Hiatt, 1965). The increase of protein synthesis accelerates progress through G1 to the S phase of the cell cycle and then to mitosis (Rubin and Sanui, 1979).

medium with normal Mg2þ, their intracellular content of Mg2þ is reduced by one-third, and they adopt a normal, flattened morphology which is retained for a few days on subculture at low density (Rubin and Chu, 1982). These observations suggest that phenotypically transformed cells have less control of their free Mg2þ content than do nontransformed cells, and they therefore constitutively maintain a higher activity of Mg2þ. Many transformed cells have higher proteolytic activity at their surface than do normal cells (Rubin, 2001), reduced adhesiveness to the substratum, and more random movement of the cell surface (Abercrombie and Ambrose, 1962). This implies either some defect in the cell membrane, as in the case of APC mutations in human colorectal cancer (Kinzler and Vogelstein, 1996), or metabolic change that perturbs the membrane from the inside just as growth factors do from the outside.

Mg2þ in Cell Growth Regulation and Transformation

47

Transformation by retroviruses or transfection of their oncogenes requires the action of strong promoter-enhancer regions that drive overexpression of the oncogenes (Chakraborty et al., 1991; Hua et al., 1997). Even normal protooncogenes can transform cells if they are coupled to a strong promoter and are heavily overexpressed (Blair et al., 1981; Chakraborty et al., 1991; Chang et al., 1982; Miller et al., 1984; Zhou and Duesberg, 1990). The src gene product of RSV is a 60,000 molecular weight protein pp60v-src that is responsible for the transforming activity of the virus (Brugge and Erikson, 1977). It is a protein kinase (Collett and Erikson, 1978) that specifically phosphorylates tyrosine residues (Hunter and Sefton, 1980). The pp60v-src is found mainly in the plasma membrane of RSV-transformed cells (Courtneidge et al., 1980; Krueger et al., 1980), where its substrates are 115- to 120k-Da proteins (Linder and Burr, 1988; Reynolds et al., 1989). Linkage of the src protein to myristic acid is necessary for its association with membranes and for transformation (Cross et al., 1984; Kamps et al., 1985). Oncogenic P3K retroviruses code for mutated homologs of the 110-kDa catalytic unit of the PI 3-Kinase fused to Gag (G-antigen of oncogenic retroviruses) sequences (Aoki et al., 2000; Aoki and Vogt, 2004). The mutations are not necessary for transformation since fusion of the cellular PI 3-K to Gag causes transformation. The Gag proteins localize the PI 3-kinase at the plasma membrane, and can be substituted by myristylation or farnesylation for this purpose. Both the membrane localization and the kinase activity are required for cell transformation, as is the activity of mTOR (Aoki et al., 2001). Since deprivation of Mg2þ reverses the transformed phenotype (Rubin et al., 1981), it suggests that the extensive phosphorylation of membrane components by overexpressed pp60v-src or PI 3-kinase may sufficiently alter membrane structure to release bound Mg2þ to the cytosol and continuously activate the coordinate response of the cell. In that sense, these oncogenes would be acting on the internal surface of the plasma membrane in a manner similar to that of growth factors on the external surface, only the activation would have to be constitutive and more intense in order to generate the transformed phenotype. It would therefore be of interest to compare the affinity for Mg2þ of the plasma membrane obtained from normal and transformed cells. Alternatively, perturbation of the membrane by heritable changes in the intracellular milieu might greatly influence the binding of Mg2þ (Dawson and Hauser, 1970) but not affect the affinity of isolated membranes. Given the weight of experimental evidence for a primary role of free Mg2þ in regulating the rate of protein synthesis and cell proliferation, much of it more than 20 years old, it might be asked why it has not received more attention in the literature on cell growth regulation. For example, a major review of early signals in the mitogenic response gives prominent attention to Naþ, Kþ, Ca2þ, and Hþ but there is no mention of Mg2þ (Rozengurt,

48

Harry Rubin

1986). More surprisingly, there is no mention in the review of the widely acknowledged, early, and persistent increase in protein synthesis induced by growth factors in normal cells and its essential relation to DNA synthesis (Brooks, 1977; Castor, 1977; Kim et al., 1968; Levine et al., 1965; Stanners and Becker, 1971). Once the role of protein synthesis is acknowledged, its sensitivity to small changes in Mg2þ, as demonstrated in vitro (Brendler et al., 1981; Ilan and Ilan, 1978; Schreier and Staehelin, 1973), would merit its consideration as a candidate for in vivo regulation of protein synthesis. However, a common attitude seems to be that intracellular Mg2þ is very well buffered and that the cell goes to ‘‘some pains to hold [it] steady’’ because it influences so many reactions (Rink et al., 1982). Mg2þ was therefore not considered a likely candidate for cellular regulation. It was contrasted with Ca2þ ‘‘which is well set up as a regulatory ion: translocation of small amounts producing large proportional changes in free concentration’’ (Rink et al., 1982). To bolster this argument, the authors found ‘‘rapid changes in T-cell [Ca2þ]i in response to plant lectins but no measurable change in cell Mg2þ.’’ However, insulin stimulation of large increases in the Mg2þ form of phosphoglucomutase have long been known (Peck and Ray, 1971) and probably reflect an increase in free Mg2þ. It was also known that insulin markedly reduces the binding of Ca2þ to the plasma membrane of rat liver cells (Shlatz and Marinetti, 1972) and that Mg2þ has the same binding properties to membranes as Ca2þ (Carvalho et al., 1963). In addition, my own laboratory had reported that insulin treatment of quiescent CEF, which markedly raised DNA synthesis and led to cell division, consistently raised total Mg2þ and presumably free Mg2þ, but was without effect on total intracellular Ca2þ (Sanui and Rubin, 1978). Of course, when a fluorescent indicator for Mg2þ became available, it was shown that mitogens do induce a sustained increase in free Mg2þ (Grubbs, 1991; Ishijima and Tatibana, 1994; Ishijima et al., 1991). The fact that Mg2þ influences so many cellular processes is an argument for rather than against its primary role in growth regulation because all cellular structures have to be reproduced in a coordinate fashion and energy metabolism has to be raised to meet the demand, a need which is ideally met by Mg2þ. In contrast to early and fleeting changes in Ca2þ of mitogen-stimulated cells, the increase of Mg2þ persists for hours (Sanui and Rubin, 1978, 1982a), as would be expected of a second messenger for proliferation since the mitogen itself must remain on the cells for most, if not all, of the G1 period in order to produce its full effect on the initiation of DNA synthesis (Bolen and Smith, 1977; Brooks, 1977; Rubin and Steiner, 1975). Given the acute sensitivity of protein synthesis to free Mg2þ, its increase need not be great but must be tightly controlled by its extensive buffering. Another critic states that growth-stimulated cells do not exhibit a steep surge of Mg2þ, as they do of Ca2þ, so they are ‘‘not set up to use Mg2þ as a natural trigger’’ (Whitfield, 1982). In a similar vein, this critic

Mg2þ in Cell Growth Regulation and Transformation

49

endorses the growth regulatory role of Ca2þ because it tends to arrest cells near the G1/S border while Mg2þ deprivation affects cells in all parts of G1 (Whitfield, 1982). In both those respects, however, Mg2þ mimics the effects of growth factor application and removal while the ‘‘trigger’’ action of Ca2þ does not. In any case, the evidence for a primary role of Mg2þ in cell growth regulation is strong enough to invalidate the objections and to raise questions about the role of the other cations. The role of Mg2þ in regulation of protein synthesis and thereby of DNA synthesis and mitosis receives strong support from current molecular studies of the PI 3-K pathway of response to growth factors (Richardson et al., 2004). This pathway leads through a protein kinase cascade to mTOR kinase, which is considered a central regulator of protein synthesis (Schmelzle and Hall, 2000). mTOR accomplishes this role by phosphorylating serines and threonines on two protein substrates which then drive further steps that initiate translation, mainly of ribosomal proteins and elongation factors (Terada et al., 1994). The mTOR phosphorylations have a high Km for what was thought to be ATP (Dennis et al., 2001; Jaeschke et al., 2004) but must, in fact, be MgATP2, which is the true substrate for all phosphoryl transfer reactions in the cell. Further considerations indicate that the level of MgATP2 is dependent on the availability of Mg2þ which is increased by stimulation with growth factors (Grubbs, 1991; Horowitz and Tluczek, 1989; Ishijima et al., 1991; Sanui and Rubin, 1978, 1982a). Hence, the molecular studies confirm a central role for Mg2þ in growth regulation that had already been inferred from the more physiological earlier studies. The molecular studies, however, indicate that the role of Mg2þ in protein synthesis is exerted mainly through its effect on the concentration of MgATP2 as substrate for mTOR kinase with its high Km for this substrate. It may be significant for the coordination of protein synthesis with energy production that the kinases of carbohydrate metabolism, which phosphorylate low molecular weight substrates, have Km for MgATP2 similar to that for mTOR (Edelman et al., 1987; Garner and Rosett, 1973). Mg2þ and its chelation with ATP may therefore play as central a role in metabolic and growth regulation as ATP does as the source of energy.

ACKNOWLEDGMENTS I am grateful for the manuscript preparation and editing by Dorothy M. Rubin and the creation of Figs. 7 and 8 by Joel Ou. This paper is dedicated to the memory of Dr. Hisashi Sanui whose ingenuity and skill in measuring intracellular and surface-bound cations by atomic absorption spectrophotometry added immeasurably to understanding the role of cations in growth regulation. The effort for the paper was supported by the National Institutes of Health grant G13LM07483-03.

REFERENCES Abercrombie, M., and Ambrose, E. J. (1962). The surface properties of cancer cells: A review. Cancer Res. 22, 525–548. Achs, M. J., Garfinkel, D., and Opie, L. H. (1982). Computer simulation of glucose-perfused rat heart in a work jump. Am. J. Physiol. 243, R389–R399. Achs, M. J., Kohn, M. C., and Garfinkel, D. (1979). Computer simulation of metabolism in pyruvate-perfused rat heart. IV. Model behavior. Am. J. Physiol. 237, R174–R180. Alberts, B., Johnson, A., Lewis, J., Raff, M., Roberts, K., and Walter, P. (2002). ‘‘Molecular Biology of the Cell’’ (Preface). Garland Science, New York, NY. Aoki, M., Blazek, E., and Vogt, P. K. (2001). A role of the kinase mTOR in cellular transformation induced by the oncoproteins P3k and Akt. Proc. Natl. Acad. Sci. USA 98, 136–141. Aoki, M., Schetter, C., Himly, M., Batista, O., Chang, H. W., and Vogt, P. K. (2000). The catalytic subunit of phosphoinositide 3-kinase: Requirements for oncogenicity. J. Biol. Chem. 275, 6267–6275. Aoki, M., Vogt, P. K. (Eds.) (2004). ‘‘Retroviral Oncogenes and TOR.’’ Springer Verlag, Berlin, Germany. Balk, S. D., Whitefield, J. F., Youdale, T., and Braun, A. C. (1973). Roles of calcium, serum, plasma and folic acid in the control of proliferation of normal and Rous sarcoma virusinfected chicken fibroblasts. Proc. Natl. Acad. Sci. USA 70, 675–679. Barrandon, Y., and Green, H. (1987). Three clonal types of keratinocyte with different capacities for multiplication. Proc. Natl. Acad. Sci. USA 84, 2302–2306. Beretta, L., Gingras, A.-C., Svitin, Y. V., Hall, M. N., and Sonenberg, N. (1996). Rapamycin blocks the phosphorylation of 4E-BP1 and inhibits cap-dependent initiation of translation. EMBO J. 15, 658–664. Blair, D. G., Oskarsson, M. K., Wood, T. G., McClements, W. L., Fischinger, P. J., and Vande Woude, G. F. (1981). Activation of the transforming potential of a normal cell sequence: A molecular model for oncogenesis. Science 212, 941–943. Bolen, J. B., and Smith, G. L. (1977). Effects of withdrawal of a mitogenic stimulus on progression of fibroblasts in S phase: Differences between serum and purified multiplicationstimulating activity. J. Cell. Physiol. 91, 441–448. Bowen-Pope, D. F., and Rubin, H. (1977). Magnesium and calcium effects on uptake of hexoses and uridine by chick embryo fibroblasts. Proc. Natl. Acad. Sci. USA 74, 1585–1589. Bowen-Pope, D. F., and Rubin, H. (1983). Growth stimulatory precipitates of Ca2þ and pyrophosphate. J. Cell. Physiol. 117, 51–61. Bowen-Pope, D. F., Vidair, C., Sanui, H., and Rubin, H. (1979). Separate roles for calcium and magnesium in their synergistic effect on uridine uptake by cultured cells. Proc. Natl. Acad. Sci. USA 76, 1308–1312. Boynton, A. L., and Whitfield, J. F. (1976). The different actions of normal and supranormal calcium concentrations on the proliferation of Balb/c 3T3 mouse cells. In Vitro 12, 479–484. Boynton, A. L., Whitfield, J. F., Isaacs, R. J., and Morton, H. J. (1974). Control of 3T3 cell proliferation by calcium. In Vitro 10, 12–17. Brendler, T., Godefroy-Colburn, T., Yu, S., and Thach, R. E. (1981). The role of mRNA competition in regulating translation. J. Biol. Chem. 256, 11755–11761. Brooks, R. F. (1977). Continuous protein synthesis is required to maintain the probability of entry into S phase. Cell 12, 311–317. Brugge, J. S., and Erikson, R. L. (1977). Identification of a transformation-specific antigen induced by an avian sarcoma virus.

Mg2þ in Cell Growth Regulation and Transformation

51

Cameron, I. L., and Greulich, R. C. (1963). Evidence for an essentially constant duration of DNA synthesis in renewing epithelia of the adult mouse. J. Cell Biol. 18, 31–40. Carney, D. H., Glenn, K. C., and Cunningham, D. D. (1978). Conditions which affect initiation of animal cell division by trypsin and thrombin. J. Cell. Physiol. 95, 13–22. Carruthers, C., and Suntzeff, V. (1943). Chemical studies on the mode of action of methylcholanthrene on mouse epidermis. Cancer Res. 3, 744–748. Carvalho, A. P., Sanui, H., and Pace, N. (1963). Calcium and magnesium binding properties of cell membrane materials. J. Cell. Comparative Physiol. 62, 311–317. Castor, L. N. (1977). Responses of protein synthesis and degradation in growth control of WI-38 cells. J. Cell. Physiol. 92, 457–467. Ceccarini, C., and Eagle, H. (1971). pH as a determinant of cellular growth and contact inhibition. Proc. Natl. Acad. Sci. USA 68, 229–233. Chakraborty, A. K., Cichutek, K., and Duesberg, P. (1991). Transforming function of proto-ras genes depends on heterologous promoters and is enhanced by specific point mutations. Proc. Natl. Acad. Sci. USA 88, 2217–2221. Chang, E. H., Furth, M. E., Scolnick, E. M., and Lowy, D. R. (1982). Tumorigenic transformation of mammalian cells induced by a normal human gene homologous to the oncogene of Harvey murine sarcoma virus. Nature 297, 479–483. Cohen, S., and Stastny, M. (1968). Epidermal growth factor. III. The stimulation of polysome formation in chick embryo epidermis. Biochim. Biophys. Acta 166, 427–437. Collett, M. S., and Erikson, R. L. (1978). Protein kinase activity associated with the avian sarcoma virus src gene product. Proc. Natl. Acad. Sci. USA 75, 2021–2024. Coman, D. R. (1944). Decreased mutual adhesiveness, a property of cells from squamous cell carcinomas. Cancer Res. 4, 625–629. Corkey, B. E., Duszynski, J., Rich, T. L., Matschinsky, B., and Williamson, J. R. (1986). Regulation of free and bound magnesium in rat hepatocytes and isolated mitochondria. J. Biol. Chem. 261, 2567–2574. Courtneidge, S. A., Levinson, A. D., and Bishop, J. M. (1980). The protein encoded by the transforming gene of avian sarcoma virus (pp60src) and a homologous protein in normal cells (pp60proto-src) are associated with the plasma membrane. Proc. Natl. Acad. Sci. USA 77, 3783–3787. Cross, F. R., Garber, E. A., Pellman, D., and Hanafusa, H. (1984). A short sequence in the p60src N terminus is required for p60src myristylation and membrane association and for cell transformation. Mol. Cell. Biol. 4, 1834–1842. Croy, R. G., and Pardee, A. B. (1983). Enhanced synthesis and stabilization of Mr 68,000 protein in transformed BALB/c-3T3 cells: Candidates for restriction point control of cell growth. Proc. Natl. Acad. Sci. USA 80, 4699–4703. Dawson, R. M. C., and Hauser, H. (1970). Binding of calcium to phospholipids. In ‘‘Calcium and Cellular Function’’ (A. W. Cuthbert, Ed.), pp. 17–41. St. Martins Press, New York. deLong, R. P., Coman, D. R., and Zeidman, I. (1950). The significance of low calcium and high potassium content in neoplastic tissue. Cancer 3, 718–721. Dennis, P. B., Jaeschke, A., Saitoh, M., Fowler, B., Kozma, S. C., and Thomas, G. (2001). Mammalian TOR: A homeostatic ATP sensor. Science 294, 1102–1105. DePhilip, R. M., Rudert, W. A., and Lieberman, I. (1980). Preferential stimulation of ribosomal protein synthesis by insulin and in the absence of ribosomal and messenger ribonucleic acid formation. Biochemistry 19, 1662–1669. Dulbecco, R., and Elkington, J. (1975). Induction of growth in resting fibroblastic cell cultures by Caþþ. Proc. Natl. Acad. Sci. USA 72, 1584–1588. Dulic, V., Lees, E., and Reed, S. I. (1992). Association of human cyclin E with a periodic G1-S phase protein kinase. Science 257, 1958–1961.

52

Harry Rubin

Ebel, H., and Gu¨ nther, T. (1980). Magnesium metabolism: A review. J. Clin. Chem. Clin. Biochem. 18, 257–270. Edelman, A. M., Blumenthal, D. K., and Krebs, E. G. (1987). Protein serine/threonine kinases. Ann. Rev. Biochem. 56, 567–593. Evans, R. B., Morhenn, V., Jones, A. L., and Tomkins, G. M. (1974). Concomitant effects of insulin on surface membrane conformation and polysome profiles of serum-starved Balb/c 3T3 fibroblasts. J. Cell Biol. 61, 95–106. Fischer, A. (1946). ‘‘Biology of Tissue Cells.’’ Cambridge Univ. Press, Cambridge, UK. Fodge, D. W., and Rubin, H. (1973). Activation of phosphofructokinase by stimulants of cell multiplication. Nature New Biol. 246, 181–183. Fodge, D. W., and Rubin, H. (1975a). Differential effects of glucocorticoids on DNA synthesis in normal and virus transformed chick embryo cells. Nature 257, 804–806. Fodge, D. W., and Rubin, H. (1975b). Glucose utilization, pH reduction and density dependent inhibition in cultures of chick embryo fibroblasts. J. Cell. Physiol. 85, 635–642. Frantz, C. N., Nathan, D. G., and Scher, C. D. (1981). Intracellular univalent cations and the regulation of the Balb/c-3T3 cell cycle. J. Cell Biol. 88, 51–56. Garfinkel, D., and Garfinkel, L. (1988). Magnesium and regulation of carbohydrate metabolism at the molecular level. Magnesium 1, 249–261. Garfinkel, D., Kohn, M. C., and Achs, M. J. (1979). Computer simulation of metabolism in pyruvate perfused rat heart. V. Physiological implications. Am. J. Physiol. 237, R181–R186. Garner, P., and Rosett, T. (1973). The influence of Mg2þ/adenine nucleotide ratios and absolute concentration of Mg2þ/adenine nucleotide on the observed velocity of some kinase reactions. FEBS Lett. 34, 243–246. Gingras, A.-C., Raught, B., and Sonenberg, N. (2004). mTOR signaling to translation. In ‘‘TOR, Target of Rapamycin’’ (G. Thomas, D. M. Sabatini, and M. N. Hall, Eds.), Vol. 279, pp. 169–197. Springer Verlag, Berlin, Germany. Glaser, L., Whiteley, B., Rothenberg, P., and Cassel, D. (1984). Mitogenic polypeptides control ion flux in responsive cells. Bioessays 1, 16–20. Gribble, F. M., Loussouarn, G., Tucker, S. J., Zhao, C., Nichols, C. G., and Ashcroft, F. M. (2000). A novel method for measurement of submembrane ATP concentration. J. Biol. Chem. 275, 30046–30049. Grubbs, R. D. (1991). Effect of epidermal growth factor on Mg2þ homeostasis in BC3H-1 myocytes. Am. J. Physiol. 260, C1158–C1164. Grubbs, R. D. (2002). Intracellular magnesium and magnesium buffering. Biometals 15, 251–259. Gurney, T. (1969). Local stimulation of growth in primary cultures of chick embryo fibroblasts. Proc. Natl. Acad. Sci. USA 62, 906–911. Hamburger, R. N., Pious, D. A., and Mills, S. F. (1963). Antigenic specificities acquired from the growth medium by cells in tissue culture. Immunology 6, 439–449. Hartwell, L. H. (1991). Twenty-five years of cell cycle genetics. Genetics 129, 975–980. Hershko, A., Mamont, P., Shields, R., and Tomkins, G. T. (1971). Pleiotypic response. Nat. New Biol. 232, 206–211. Hesketh, T. R., and Moore, J. P. (1985). A common sequence of calcium and pH signals in the mitogenic stimulation of eukaryotic cells. Nature 313, 481–484. Horowitz, S. B., and Lau, Y.-T. (1988). A function that relates protein synthetic rates to potassium activity in vivo. J. Cell. Physiol. 135, 425–434. Horowitz, S. B., and Tluczek, L. J. M. (1989). Gonadotropin stimulates oocyte translation by increasing magnesium activity through intracellular potassium-magnesium exchange. Proc. Natl. Acad. Sci. USA 86, 9652–9656. Hosseini, J. M., and Elin, R. J. (1985). Magnesium variability of lymphocytes from cell culture. J. Amer. College Nutrition 4, 613–617.

Mg2þ in Cell Growth Regulation and Transformation

53

Hua, V., Wang, W. K., and Duesberg, P. (1997). Dominant transformation by mutated ras genes in vitro requires more than 100 times higher expression than is observed in cancers. Proc. Natl. Acad. Sci. USA 94, 9614–9619. Hunter, T., and Sefton, B. M. (1980). Transforming gene product of Rous sarcoma virus phosphorylates tyrosine. Proc. Natl. Acad. Sci. USA 77, 1311–1315. Ilan, J., and Ilan, J. (1978). Translation of maternal messenger ribonucleoprotein particles from sea urchin in a cell-free system from unfertilized eggs and product analysis. Dev. Biol. 66, 375–385. Ishijima, S., Sonoda, T., and Tatibana, M. (1991). Mitogen-induced early increase in cytosolic free Mg2þ concentration in single Swiss 3T3 fibroblasts. Am. J. Physiol. 261, C1074–C1080. Ishijima, S., and Tatibana, M. (1994). Rapid mobilization of intracellular Mg2þ by bombesin in Swiss 3T3 cells: mobilization through external Ca2þ—and tyrosine kinase-dependent mechanisms. J. Biochem. 115, 730–737. Jaeschke, A., Dennis, P. B., and Thomas, G. (2004). mTOR: A mediator of intracellular homeostasis. In ‘‘TOR, Target of Rapamycin’’ (G. Thomas, D. M. Sabatini, and M. N. Hall, Eds.), pp. 283–298. Springer Verlag, Berlin, Germany, Vol. 279, . Kamine, J., and Rubin, H. (1976). Magnesium required for serum stimulation of growth in cultures of chick embryo fibroblasts. Nature 263, 143–145. Kamps, M. P., Buss, J. E., and Sefton, B. M. (1985). Mutation of NH2-terminal glycine of p60src prevents both myristoylation and morphological transformation. Proc. Natl. Acad. Sci. USA 82, 4625–4628. Killander, D., and Zetterberg, A. (1965). A quantitative cytochemical investigation of the relationship between cell mass and initiation of DNA synthesis in mouse fibroblasts in vitro. Exp. Cell Res. 40, 12–20. Kim, J. H., Gelbard, A. S., and Perez, A. G. (1968). Inhibition of DNA synthesis by actinomycin D and cycloheximide in synchronized HeLa cells. Exp. Cell Res. 53, 478–487. Kinzler, K. W., and Vogelstein, B. (1996). Lessons from hereditary colorectal cancer. Cell 87, 159–170. Koff, A., Giordano, A., Desai, D., Yamashita, K., Harper, J. W., Elledge, S., Nishimoto, T., Morgan, D. O., Franza, B. R., and Roberts, J. M. (1992). Formation and activation of a cyclin E-cdk2 complex during the G1 phase of the human cell cycle. Science 257, 1689–1694. Krueger, J. G., Wang, E., and Goldberg, A. R. (1980). Evidence that the src gene product of Rous sarcoma virus is membrane associated. Virology 101, 25–40. Lansing, A. I., Rosenthal, T. B., and Kamen, M. D. (1948). Calcium ion exchanges in some normal tissues and in epidermal carcinogenesis. Arch. Bioche. 19, 177–183. Lardy, H. A., and Parks, R. E., Jr. (1956). Influences of ATP concentration on rates of some phosphorylation reactions. In ‘‘Enzymes: Units of Biological Function’’ (O. H. Gaebler, Ed.), pp. 584–587. Academic Press. Lau, Y.-T., Yassin, R. R., and Horowitz, S. B. (1988). Potassium salt microinjection into Xenopus oocytes mimics gonadotropin treatment. Science 240, 1231–1323. Ledbetter, M. L. S., and Lubin, M. (1975). Regulation of protein synthesis in human diploid fibroblasts by intracellular potassium levels. J. Cell. Biol. 67, 237a. Ledbetter, M. L. S., and Lubin, M. (1977). Control of protein synthesis in human fibroblasts by intracellular potassium. Exp. Cell Res. 105, 223–236. Levine, E. M., Becker, Y., Boone, C. W., and Eagle, H. (1965). Contact inhibition, macromolecular synthesis, and polyribosomes in cultured human diploid fibroblasts. Proc. Natl. Acad. Sci. USA 53, 350–356. Linder, M. E., and Burr, J. G. (1988). Nonmyristoylated p60v-src fails to phosphorylate proteins of 115–120 kDa in chicken embryo fibroblasts. Proc. Natl. Acad. Sci. USA 85, 2608–2612.

54

Harry Rubin

Liskay, R. M., Kornfeld, B., Fullerton, P., and Evans, R. (1980). Protein synthesis and the presence or absence of a measurable G1 in cultured Chinese hamster cells. J. Cell. Physiol. 104, 461–467. Lopez-Rivas, A., Adelberg, E. A., and Rozengurt, E. (1982). Intracellular Kþ and the mitogenic response of 3T3 cells to peptide factors in serum-free medium. Proc. Natl. Acad. Sci. USA 79, 6275–6279. Lostroh, A. J., and Krahl, M. E. (1973). Insulin action: Accumulation in vitro of Mg2þ and Kþ in rat uterus: Ion pump activity. Biochimica et Biophysica Acta. 291, 260–268. Lostroh, A. J., and Krahl, M. E. (1974). Magnesium, a second messenger for insulin: Ion translocation coupled to transport activity. Adv. Enzyme Regul. 12, 73–81. Lubin, M., Cahn, F., and Coutermarsh, B. A. (1982). Amiloride, protein synthesis, and activation of quiescent cells. J. Cell. Physiol. 113, 247–251. Matsushime, H., Roussel, M. F., Ashmun, R. A., and Sherr, C. J. (1991). Colony-stimulating factor 1 regulates novel cyclins during the G1 phase of the cell cycle. Cell 65, 701–713. McCutcheon, M., Coman, D. R., and Moore, F. B. (1948). Studies on invasiveness of cancer. Adhesiveness of malignant cells in various human adenocarcinomas. Cancer 1, 460–467. McKeehan, W., and Ham, R. G. (1978). Calcium and magnesium ions and the regulation of multiplication in normal and transformed cells. Nature 275, 756–758. Meselson, M., Nomura, M., Brenner, S., Davern, C., and Schlessinger, D. (1964). Conservation of ribosomes during bacterial growth. J. Mol. Biol. 9, 696–711. Miller, A. D., Curran, T., and Verma, I. M. (1984). c-fos protein can induce cellular transformation: a novel mechanism of activation of a cellular oncogene. Cell 36, 51–60. Moolenaar, W. H., Tertoolen, L. G. J., and de Laat, S. W. (1984). Growth factors immediately raise cytoplasmic free Ca2þ in human fibroblasts. J. Biol. Chem. 259, 8066–8069. Moolenaar, W. H., Tsien, R. Y., van der Saag, P. T., and de Laat, S. W. (1983). Naþ/Hþ exchange and cytoplasmic pH in the action of growth factors in human fibroblasts. Nature 304, 645–648. Moscatelli, D., Sanui, H., and Rubin, H. (1979). Effects of depletion of Kþ, Naþ, or Ca2þ, on DNA synthesis and cell cation content in chick embryo fibroblasts. J. Cell. Physiol. 101, 117–128. Murphy, E., Freudenrich, C. C., Levy, L. A., and London, R. E. (1989). Monitoring cytosolic free magnesium in cultured chicken heart cells by use of the fluorescent indicator Furaptra. Proc. Natl. Acad. Sci. USA 86, 2981–2984. O’Neill, C., Jordan, P., and Ireland, G. (1986). Evidence for two distinct mechanisms of anchorage stimulation in freshly explanted and 3T3 Swiss mouse fibroblasts. Cell 44, 489–496. Peck, E. J., and Ray, W. J. (1971). Metal complexes of phosphoglucomutase in vivo. J. Bio. Chem. 246, 1160–1167. Plagemann, P. G. W., Marz, R., and Wohlheuter, R. M. (1978). Uridine transport in Novikoff rat hepatoma cells and other cell lines and its relationship to uridine phosphorylation and phosphorolysis. J. Cell. Physiol. 97, 49–72. Prescott, D. M. (1968). Regulation of cell production. Cancer Res. 28, 1815–1820. Raizada, M. K., and Perdue, J. F. (1975). Insulin receptors in cultured normal chick embryo fibroblasts. J. Cell Biol. 67, 349a. Raju, B., Murphy, E., Levy, L. A., Hall, R. D., and London, R. E. (1989). A fluorescent indicator for measuring cytosolic free magnesium. Am. J. Physiol. 256, C540–C548. Revel, M., and Hiatt, H. H. (1965). Magnesium requirement for the formation of an active messenger RNA-ribosome-S-RNA complex. J. Mol. Biol. 11, 467–475. Reynolds, A. B., Roesel, D. J., Kanner, S. B., and Parsons, J. T. (1989). Transformation-specific tyrosine phosphorylation of a novel cellular protein in chicken cells expressing oncogenic variants of the avian cellular src gene. Mol. Cell. Biol. 9, 629–638.

Mg2þ in Cell Growth Regulation and Transformation

55

Richardson, C. J., Schalm, S. S., and Blenis, J. (2004). PI3-kinase and TOR: PIKTORing cell growth. Seminars in Dev. Biol. 15, 147–159. Rink, T. J., Tsien, R. Y., and Pozzan, T. (1982). Cytoplasmic pH and free Mg2þ in lymphocytes. J. Cell Biol. 95, 189–196. Rose, I. A. (1968). The state of magnesium in cells as estimated from the adenylate kinase equilibrium. Proc Natl. Acad. Sci. USA 61, 1079–1086. Rossow, P. W., Riddle, V. G. H., and Pardee, A. B. (1979). Synthesis of labile, serum-dependent protein in early G1 controls of animal cell growth. Proc. Natl. Acad. Sci. USA 76, 4446–4450. Rozengurt, E. (1986). Early signals in the mitogenic response. Science 234, 161–166. Rozengurt, E., and Heppel, L. A. (1975). Serum rapidly stimulates ouabain-sensitive 86Rbþ influx in quiescent 3T3 cells. Proc. Natl. Acad. Sci. USA 72, 4492–4495. Rozengurt, E., and Stein, W. D. (1977). Regulation of uridine uptake by serum and insulin in density-inhibited 3T3 cells. Biochim. Biophys. Acta. 464, 417–432. Rozengurt, E., Stein, W. D., and Wigglesworth, N. (1977). Uptake of nucleosides in densityinhibited cultures of 3T3 cells. Nature 267, 442–444. Rubin, H. (1971a). Growth regulation in cultures of chick embryo fibroblasts. In ‘‘Ciba Foundation Symposium on Growth Control in Cell Cultures’’ (G. E. W. Wolstenholme and J. Knight, Eds.), pp. 127–149. Churchill Livingstone, Edinburgh, UK. Rubin, H. (1971b). pH and population density in the regulation of animal cell multiplication. J. Cell Biol. 51, 686–702. Rubin, H. (1975a). Central role for magnesium in coordinate control of metabolism and growth in animal cells. Proc. Natl. Acad. Sci. USA 72, 3551–3555. Rubin, H. (1975b). Nonspecific nature of the stimulus to DNA synthesis in cultures of chick embryo cells. Proc. Natl. Acad. Sci. USA 72, 1676–1680. Rubin, H. (1976). Magnesium deprivation reproduces the coordinate effects of serum removal or cortisol addition on transport and metabolism in chick embryo fibroblasts. J. Cell. Physiol. 89, 613–626. Rubin, H. (1977). Antagonistic effects of insulin and cortisol on coordinate control of metabolism and growth in cultured fibroblasts. J. Cell. Physiol. 91, 249–260. Rubin, H. (1981). Growth regulation, reverse transformation, and adaptability of 3T3 cells in decreased Mg2þ concentration. Proc. Natl. Acad. Sci. USA 78, 328–332. Rubin, H. (1982). Effect of magnesium content on density-dependent regulation of the onset of DNA synthesis in transformed 3T3 cells. Cancer Res. 42, 1761–1768. Rubin, H. (2001). Synergistic mechanisms in carcinogenesis by polycyclic aromatic hydrocarbons and tobacco smoke: A biohistorical perspective with updates. Carcinogenesis 22, 1903–1930. Rubin, H. (in press). Magnesium: The missing element in molecular views of cell proliferation control. Bio. Essays. Rubin, H., and Chu, B. (1978). Reversible regulation by magnesium of chick embryo fibroblast proliferation. J. Cell. Physiol. 94, 13–19. Rubin, H., and Chu, B. (1982). Self-normalization of highly transformed 3T3 cells through maximized contact interaction. Proc. Natl. Acad. Sci. USA 79, 1903–1907. Rubin, H., and Fodge, D. (1974). Interrelationships of glycolysis, sugar transport and the initiation of DNA synthesis in chick embryo cells. In ‘‘Control of Proliferation in Animal Cells,’’ (B. Clarkson and R. Baserga, Eds.), Vol. 1, pp. 801–816. Cold Spring Harbor Laboratory. Rubin, H., and Hatie´ , C. (1968). Increase in the size of chick embryo cells upon cultivation in serum-containing medium. Dev. Biol. 17, 603–616. Rubin, H., and Koide, T. (1973). Stimulation of DNA synthesis and 2-deoxy-D-glucose transport in chick embryo cultures by excessive metal concentrations and by a carcinogenic hydrocarbon. J. Cell. Physiol. 81, 387–396.

56

Harry Rubin

Rubin, H., and Koide, T. (1975). Early cellular responses to diverse growth stimuli independent of protein and RNA synthesis. J. Cell. Physiol. 86, 47–58. Rubin, H., and Sanui, H. (1977). Complexes of inorganic pyrophosphate, orthophosphate and calcium as stimulants of 3T3 cells multiplication. Proc. Natl. Acad. Sci. USA 74, 5025–5030. Rubin, H., and Sanui, H. (1979). The coordinate response of cells to hormones and its mediation by the intracellular availability of magnesium. In ‘‘Hormones and Cell Culture,’’ (R. Ross and G. Sato, Eds.), Vol. 6, pp. 741–750. Cold Spring Harbor Conf. on Cell Proliferation. Rubin, H., and Steiner, R. (1975). Reversible alteration in the mitotic cycle of chick embryo cells in various states of growth regulation. J. Cell. Physiol. 85, 261–270. Rubin, H., Terasaki, M., and Sanui, H. (1978). Magnesium reverses inhibitory effects of calcium deprivation on coordinate response of 3T3 cells to serum. Proc. Natl. Acad. Sci. USA 75, 4379–4383. Rubin, H., Terasaki, M., and Sanui, H. (1979). Major intracellular cations and growth control: Correspondence among magnesium content, protein synthesis, and the onset of DNA synthesis in Balb/c 3T3 cells. Proc. Natl. Acad. Sci. USA 76, 3917–3921. Rubin, H., Vidair, C., and Sanui, H. (1981). Restoration of normal appearance, growth behavior and calcium content to transformed 3T3 cells by magnesium deprivation. Proc. Natl. Acad. Sci. USA 78, 2350–2354. Sanui, H. (1970). pH dependence of the effect of adenosine triphosphate and ethylenediaminetetraacetate on sodium and magnesium binding by cellular membrane fragments. J. Cell. Physiol. 75, 361–368. Sanui, H., and Pace, N. (1967). Effect of ATP, EDTA and EGTA on the simultaneous binding of Na, K, Mg and Ca by rat liver microsomes. J. Cell. Physiol. 69, 11–20. Sanui, H., and Rubin, H. (1977). Correlated effects of external magnesium on cation content and DNA synthesis in cultured chicken embryo fibroblasts. J. Cell. Physiol. 92, 23–32. Sanui, H., and Rubin, H. (1978). Membrane bound and cellular cationic changes associated with insulin stimulation of cultured cells. J. Cell. Physiol. 96, 265–278. Sanui, H., and Rubin, H. (1979a). Intracellular divalent and monovalent cation changes associated with ouabain inhibition of protein and DNA synthesis in Balb/c 3T3 cells. J. Cell Biol. 83, 8a. Sanui, H., and Rubin, H. (1979b). Measurement of total, intracellular and surface bound cations in animal cells grown in culture. J. Cell. Physiol. 100, 215–226. Sanui, H., and Rubin, H. (1982a). Changes of intracellular and externally bound cations accompanying serum stimulation of mouse Balb/c3T3 cells. Exp. Cell. Res. 139, 15–25. Sanui, H., and Rubin, H. (1982b). The role of magnesium in cell proliferation and transformation. In ‘‘Ions, Cell Proliferation, and Cancer,’’ (A. L. Boynton, W. L. Mckeehan, and J. F. Whitfield, Eds.), pp. 517–537. Academic Press. Sanui, H., and Rubin, H. (1984). Ionic changes associated with lead stimulation of DNA synthesis in Balb/c3T3 cells. Biological Trace Element Research 6, 289–307. Schlessinger, J. (1988). Signal transduction by allosteric receptor oligomerization. Trends Biochem. Sci. 13, 443–447. Schmelzle, T., and Hall, M. N. (2000). TOR, a central controller of cell growth. Cell 103, 253–262. Schneiderman, M. H., Dewey, W. C., and Highfield, D. P. (1971). Inhibition of DNA synthesis in synchronized Chinese hamster cells treated in G1 with cycloheximide. Exp. Cell Res. 67, 147–155. Schreier, M. H., and Noll, H. (1971). Conformational changes in ribosomes during protein synthesis. Proc. Natl. Acad. Sci. USA 68, 805–809. Schreier, M. H., and Staehelin, T. (1973). Initiation of mammalian protein synthesis: The importance of ribosome and initiation factor quality for the efficiency of in vitro systems. J. Mol. Biol. 73, 329–349.

Mg2þ in Cell Growth Regulation and Transformation

57

Schuldiner, S., and Rozengurt, E. (1982). Naþ/Hþ antiport in Swiss 3T3 cells: Mitogenic stimulation leads to cytoplasmic alkalinization. Proc. Natl. Acad. Sci. USA 79, 7778–7782. Sefton, B. M., and Rubin, H. (1970). Release from density dependent growth inhibition by proteolytic enzymes. Nature 227, 843–845. Sherr, C. J. (1993). Mammalian G1 cyclins. Cell 73, 1059–1065. Sherr, C. J. (1994). G1 phase progression: Cycling on cue. Cell 79, 551–555. Shimoyama, Y., Nagafuchi, A., Fujita, S., Gotoh, M., Takeichi, M., Tsukita, S., and Hirobashi, S. (1992). Cadherin dysfunction in a human cell line: Possible involvement of loss a -catenin expression in reduced cell-cell adhesiveness. Cancer Res. 52, 5770–5774. Shlatz, L., and Marinetti, G. V. (1972). Hormone-calcium interactions with the plasma membrane of rat liver cells. Science 176, 175–177. Smith, G. L. (1977). Increased ouabain-sensitive 86rubidiumþ uptake after mitogenic stimulation of quiescent chicken embryo fibroblasts with purified multiplication stimulating activity. J. Cell Biol. 73, 761–767. Stanners, C. P., and Becker, H. (1971). Control of macromolecular synthesis in proliferating and resting Syrian hamster cells in monolayer culture. J. Cell. Physiol. 77, 31–42. Stoker, M. (1964). Regulation of growth and orientation in hamster cells transformed by polyomavirus. Virology 24, 165–174. Stoker, M., O’Neill, C., Berryman, S., and Waxman, V. (1968). Anchorage and growth regulation in normal and virus-transformed cells. Int. J. Cancer 3, 683–693. Stoker, M., Shearer, M., and O’Neill, C. (1966). Growth inhibition of polyoma-transformed cells by contact with static normal fibroblasts. J. Cell. Sci. 1, 297–310. Swierenga, S. H. H., MacManus, J. P., and Whitefield, J. F. (1976). Regulation by calcium of the proliferation of heart cells from young adult rats. In Vitro 12, 31–36. Temin, H. M. (1971). Stimulation by serum of multiplication of stationary chicken cells. J. Cell. Physiol. 78, 161–170. Terada, N., Patel, H. R., Takase, K., Kohno, K., and Nairns, A. C. (1994). Rapamycin selectively inhibits translation of mRNAs encoding elongation factors and ribosomal proteins. Proc. Natl. Acad. Sci. USA 91, 11477–11481. Terasaki, M. (1983). Studies on magnesium in non-transformed and transformed Balb/c 3T3 cells. Ph.D. Dissertation, Univ. of California, Berkeley, CA. Terasaki, M., and Rubin, H. (1981). Relations between intracellular Mg2þ, protein synthesis and the phenotype of transformed Balb/c 3T3 cells. J. Cell. Biol. 91, 4a. Terasaki, M., and Rubin, H. (1985). Evidence that intracellular magnesium is present in cells at a regulatory concentration for protein synthesis. Proc. Natl. Acad. Sci. USA 82, 7324–7326. Tupper, J. T., Zorgniotti, F., and Mills, B. (1977). Potassium transport and content during G1 and S phase following serum stimulation of 3T3 cells. J. Cell. Physiol. 91, 429–440. Turner, M. K., Abrams, R., and Lieberman, I. (1968). Levels of ribonucleotide reductase activity during the division cycle of the L cell. J. Biol. Chem. 243, 3725–3728. Ullrich, A., and Schlessinger, J. (1990). Signal transduction by receptors with tyrosine kinase activity. Cell 61, 203–212. Umbas, R., Schalken, J. A., Aalders, T. W., Carter, B. S., Karthaus, H. F. M., Schaafsma, H. E., Debruyne, F. M. J., and Isaacs, W. B. (1992). Expression of the cellular adhesion molecule Ecadherin is reduced or absent in high-grade prostate cancer. Cancer Res. 52, 5104–5109. Vidair, C., and Rubin, H. (1981). Evaluation of Mg2þ as an intracellular regulator of uridine uptake. J. Cell. Physiol. 108, 317–325. Vidair, C., and Rubin, H. (in press). Mg2þ as activator of uridine phosphorylation and other cellular responses to growth factors. Proc. Natl. Acad. Sci. Vleminckx, K., and Kemler, R. (1999). Cadherins and tissue formation: Integrating adhesion and signaling. BioEssays 21, 211–220.

58

Harry Rubin

Weiss, R. A. (1970). The influence of normal cells on the proliferation of tumor cells in culture. Exp. Cell Res. 63, 1–18. Whitfield, J. F. (1982). The roles of calcium and magnesium in cell proliferation: An overview. In ‘‘Ions, Proliferation, and Cancer’’ (A. Boynton, W. L. McKeehan, and J. F. Whitfield, Eds.), pp. 283–294. Academic Press, New York. Wolf, F. I., Fasanella, S., Tedesco, B., Torsello, A., Sgambto, A., Faraglia, B., Palozza, P., Boninsegna, A., and Cittadini, A. (2004). Regulation of magnesium content during proliferation of mammary epithelial cells (HC-11). Front. Biosci. 9, 2056–2062. Won, K.-A., Xiong, Y., Beach, D., and Gilman, M. Z. (1992). Growth-regulated expression of D-type cyclin genes in human diploid fibroblasts. Proc. Natl. Acad. Sci. USA 89, 9910–9914. Zeidman, I. (1947). Chemical factors in the mutual adhesiveness of epithelial cells. Cancer Res. 7, 386–389. Zetterberg, A., and Engstro¨ m, W. (1983). Induction of DNA synthesis and mitosis in the absence of cellular enlargement. Exp. Cell Res. 144, 199–207. Zetterberg, A., and Killander, D. (1965). Quantitative cytophotometric and autoradiographic studies on the rate of protein synthesis during interphase in mouse fibroblasts in vitro. Exp. Cell Res. 40, 1–11. Zhou, H., and Duesberg, P. (1990). A retroviral promoter is sufficient to convert proto-src to a transforming gene that is distinct from the src gene of Rous sarcoma virus. Proc. Natl. Acad. Sci. USA 87, 9128–9132.

Presence and Influence of Human Papillomaviruses (HPV) in Tonsillar Cancer Hanna Mellin Dahlstrand and Tina Dalianis Department of Oncology-Pathology, Karolinska Institute, Karolinska University Hospital, 171 76, Stockholm, Sweden

I. II. III. IV.

Introduction Tonsillar Cancer Human Papillomavirus (HPV) Human Papillomavirus (HPV) in Tonsillar Cancer A. Frequency and Type of HPV in Tonsillar Cancer B. HPV and Tonsillar Cancer Patient Features C. HPV and Prognosis in Tonsillar Cancer D. HPV and Radiosensitivity in Tonsillar Cancer E. HPV and Correlation to Cell-Cycle Proteins in Tonsillar Cancer F. HPV and Genetic Instability in Tonsillar Cancer V. HPV and Other Tumors of the Head and Neck VI. HPV Vaccines VII. Conclusions References

Tonsillar cancer is the most common of the oropharyngeal carcinomas and human papillomavirus (HPV) has been found to be present in approximately half of all cases. Patients with HPV-positive tonsillar cancer have been observed to have a better clinical outcome than patients with HPV-negative tonsillar cancer. Moreover, patients with tonsillar cancer and a high viral load have been shown to have a better clinical outcome, including increased survival, compared to patients with a lower HPV load in their tumors. Recent findings show that HPV-positive tumors are not more radiosensitive and do not have fewer chromosomal aberrations than HPV-negative tumors, although some chromosomal differences may exist between HPV-positive and -negative tonsillar tumors. Current experimental and clinical data indicate that an active antiviral cellular immune response may contribute to this better clinical outcome. These data are also in line with the findings that the frequency of tonsillar cancer is increased in patients with an impaired cellular immune system. Thus, therapeutic and preventive HPV-16 antiviral immune vaccination trials may be worthwhile, not only in cervical cancer, but also in tonsillar cancer. ß 2005 Elsevier Inc.

Advances in CANCER RESEARCH 0065-230X/05 $35.00

59

Copyright 2005, Elsevier Inc. All rights reserved

60

Hanna Mellin Dahlstrand and Tina Dalianis

I. INTRODUCTION When zur Hausen (zur Hausen, 1976) proposed that cervical cancer might be caused by human papillomavirus (HPV), the scientific community accepted that HPV could potentially be involved in the development of some but definitely not all cervical cancers. Today, it is fully accepted that different types of HPVs are present and instrumental in the induction of almost all cervical cancers as well as some other types of human cancer (zur Hausen, 1996, 1999). In addition, several mechanisms by which HPVs exhibit their oncogenic potential have also been revealed (zur Hausen, 1996, 1999). The possibility that preventive vaccines against HPV may soon reach clinical practice (Koutsky et al., 2002; Lehtinen and Dillner, 2002) has drawn even more attention to the association of HPV with other types of cancer. However, HPV is only present in a proportion of other types of cancer, e.g., head and neck cancer, anogenital cancer, and nonmelanoma skin cancer, and much less is known about the role of HPV in these tumors (Alani and Munger, 1998; de Villiers, 1991, 1997; Gillison et al., 1999; Licitra et al., 2002; Mork et al., 2001; Snijders et al., 1992). Nonetheless, tonsillar carcinoma is of particular interest, since it is the head and neck cancer where HPV is most commonly found (Gillison et al., 2000; Mork et al., 2001; Paz et al., 1997; Snijders et al., 1996). Approximately half of all tonsillar cancers are HPV positive (Andl et al., 1998; Klussmann et al., 2001; Mellin et al., 2000; Paz et al., 1997). In addition, recent reports suggest that patients with HPV-positive tonsillar tumors have a lower risk of relapse and longer survival compared to patients with HPV-negative tonsillar tumors (Gillison et al., 2000; Mellin et al., 2000). These data motivate further comparisons between HPV-positive and HPV-negative tonsillar tumors with regard to clinical outcome, sensitivity to radiotherapy, biology of the tumor, and genetic stability in order to better understand possible options for treatment and vaccination studies. The purpose of this article is to review current knowledge on the status and significance of HPV in tonsillar cancer.

II. TONSILLAR CANCER Cancer of the palatine tonsil, in the lymphoid region called the Waldeyer’s ring, is usually referred to as tonsillar cancer. Tonsillar cancer is the most common of the oropharyngeal malignancies, and 75% of all tonsillar carcinomas are squamous cell carcinomas (Genden et al., 2003). As with all head and neck squamous cell carcinomas (HNSCC), smoking and alcohol abuse are regarded as the main etiological factors for tonsillar

HPV in Tonsillar Cancer

61

cancer and are known to account for 80 to 90% of all HNSCC (Decker and Golstein, 1982; Licitra et al., 2002). However, HNSCC and tonsillar cancer also occur in some 15 to 20% of patients without these risk factors (Gillison et al., 2001; Licitra et al., 2002). In many of these instances, viruses are most likely to be involved in the development of HNSCC, and data now indicate that high-risk types of HPV, similar to those observed in cervical cancer, are associated with a subset of HNSCC (Alani and Munger, 1998; de Villiers, 1991; Gillison et al., 2000, 2001; Gissmann et al., 1982; Mork et al., 2001; Naghashfar et al., 1985; Snijders et al., 1992; Syrjanen et al., 1983). Patients with tonsillar cancer do not normally seek health care until the tumor is fairly large and presents symptoms like swallowing-related pain or difficulties in swallowing (Mashberg and Samit, 1995). This is because small tumors generally do not cause any discomfort. Other common first symptoms are pain in the ear or a lump in the neck due to the tumor spreading to the lymph nodes (Mashberg and Samit, 1995). In advanced cases, vital functions such as breathing, eating, and speaking may be significantly affected. Further suffering may be caused by cancer growth in the face and the neck. Later on, the curative treatments of surgery and radiotherapy (Genden et al., 2003; Mellin et al., 2000) can be disabling and disfiguring. Tonsillar cancer is generally treated with (pre-operative or post-operative) full-dose radiotherapy (64 Gy) against the primary tumor and the neck. The extent of the surgical intervention depends on the size of the primary tumor, the presence of metastases in the neck lymph nodes, and the response to the radiotherapy given. Overall survival for patients with oropharyngeal cancer is 67% with stage I, 46% with stage II, 31% with stage III, and 32% with stage IV (Pugliano et al., 1997). However, the overall survival for patients with oropharyngeal cancer is only about 38% (Pugliano et al., 1997). Despite similar histology and stage, as well as standardized treatment, it is not easy to predict the outcome of each individual case. Hence, both predictive and prognostic markers would be of significant clinical value in order to tailor treatment for individual tonsillar cancer patients. This would allow for optimization of therapy to give the most efficient treatment with minimal impact on function and form. Moreover, given the high proportion of HPV-16 associated with tonsillar cancer (see following text), patients could obtain substantial benefit from the use of the same prophylactic and adjuvant therapeutic strategies that are being developed to prevent and/or treat HPV-associated anogenital cancers (for reviews, see Devaraj et al., 2003; Ling et al., 2000). However, before using such treatment, it is important to investigate in which cases this could be an option.

62

Hanna Mellin Dahlstrand and Tina Dalianis

III. HUMAN PAPILLOMAVIRUS (HPV) There are more than 100 HPV types, and for general reviews on HPV and cancer, as well as, more specifically, head, neck, and oral cancer, see, for example, Zur Hausen (1996), de Villiers (1997), Syrjanen (2003), and Scully (2002). Some HPV types are associated with common warts, while others are associated with chondylomas and papillomas. Finally, there are HPV types such as HPV 16, 18, 31, 33, and others that are associated with malignant tumors. Nevertheless, the genomes of all HPVs are similar and consist of double-stranded circular DNA with a size of 7 to 8 Kb. The genome is enclosed in a 52- to 55-nm viral capsid, and is arbitrarily divided into a noncoding region and two coding regions, the early and late regions. The early region encodes for the early proteins E1-E2 and E4-E7, which are important for pathogenesis and transformation, while the late region encodes for L1 and L2, the two capsid proteins. Of particular interest in this context is that among the HPV types associated with malignant tumors, E6 and E7 are classified as oncogenes. E6 binds to the cellular protein p53 and degrades it, while E7 binds to pRB and abrogates its function (Dyson et al., 1989, 1992; Scheffner et al., 1990). Under these conditions, the intracellular levels of normal p53 and pRB are reduced and this combination results in the inhibition of cell cycle control and facilitation of tumor development (for reviews, see Hanahan and Weinberg, 2000; zur Hausen, 1996). Also of interest is that the L1, the major capsid protein, can self assemble and form viruslike particles (Kirnbauer et al., 1993), which are useful for vaccination against HPV infections.

IV. HUMAN PAPILLOMAVIRUS (HPV) IN TONSILLAR CANCER A. Frequency and Type of HPV in Tonsillar Cancer HPV DNA has been shown to be present in 45 to 100% of all tonsillar tumors (Andl et al., 1998; Dahlgren et al., 2003; Gillison et al., 2000; Koskinen et al., 2003; Mellin et al., 2000; Ringstro¨ m et al., 2002; Snijders et al., 1992). The variation depends mainly on the type of material that is available for analysis and the methodology used for detection (as has been discussed) (Mellin, 2002; Mellin et al., 2002). In general, it is easier to  detect HPV in fresh-frozen (
70 C) tumors compared to formalin-fixed and paraffin-embedded tumors, where the DNA is degraded and where

HPV in Tonsillar Cancer

63

degradation progresses even after storage for decades (Mellin, 2002; Mellin et al., 2000, 2002). The most common and sensitive technique for the detection of HPV is based on polymerase chain reaction (PCR) technology (Mellin, 2002; Mellin et al., 2002). In the past, less sensitive methods such as Southern blots or in situ hybridization techniques have been used (Mellin, 2002; Syrjanen, 1990). Screening for HPV by PCR analysis is usually initially performed using HPV consensus/general primers (e.g., GP5þ/6þ, My9/10, CPI/IIG, FAP59/61) (de Roda Husman et al., 1995; Forslund et al., 1999; Manos et al., 1989; Tieben et al., 1993). These primer sets allow for the amplification of a wide range of HPV types and are useful for screening. Alternatively, they can be type-specific and identify only one specific HPV type (Hagmar et al., 1992). General primers are complementary to sequences (often in L1) in HPV that are highly conserved among many HPV types, while HPV type-specific primers bind to a sequence found in a single HPV type (often in E6 or E7) and do not cross-bind to other HPV types. Instead of an HPV type-specific PCR, HPV typing can also be performed by sequencing the PCR product obtained by a PCR run with general primers (Mellin et al., 2002). For HPV typing, HPV type-specific oligonucleotide probes using either enzyme immunoassays or Southern blot hybridizations are also commonly used (Herrero et al., 2003; van Houten et al., 2001). Without doubt, HPV type 16 is the type predominant in tonsillar cancer (Andl et al., 1998; Gillison et al., 2000; Klussmann et al., 2001; Koskinen et al., 2003; Mork et al., 2001; Paz et al., 1997; Snijders et al., 1992; Strome et al., 2002; Wilczynski et al., 1998). In most reports of HPV-positive tonsillar cancer biopsies, 85 to 100% contain HPV-16 followed by 0 to 7% containing HPV-33. HPV-31, HPV-59, or non-typeable HPVs are found even more rarely. In addition, when using DNA as well as RNA in situ hybridization, the viral genome and its transcription products (performed on HPV-16) have been located in cancer cells and nodal metastases but not to the surrounding stroma of the primary tonsillar tumor or the nodal metastases (Demetrick et al., 1990; Niedobitek et al., 1990; Snijders et al., 1992; Strome et al., 2002; Wilczynski et al., 1998).

B. HPV and Tonsillar Cancer Patient Features Patients with HPV-positive tonsillar tumors are less likely to be heavy smokers and drinkers, although it has been reported that HPV may have a synergistic effect with regard to tumor development in smokers (Gillison et al., 2000; Haraf et al., 1996; Herrero et al., 2003; Koch et al., 1999; Ringstro¨ m et al., 2002; Schwartz et al., 1998). A current concern regards the possible sexual transmission of HPV in oral and oropharyngeal

64

Hanna Mellin Dahlstrand and Tina Dalianis

squamous cell carcinoma (OSCC) (Herrero et al., 2003; Scully, 2002). This is suggested by the significant increase in tonsillar cancer reported among men in the United States from 1973 to 1995 (Frisch et al., 2000b). This may be explained by changes in sexual habits resulting in the increased transmission of HPV (Devaraj et al., 2003). It is also known that individuals with HPV-associated anogenital malignancies have an increased risk for a second primary cancer in the tonsils and oral cavity (Boice et al., 1985; Frisch and Biggar, 1999; Rabkin et al., 1992). In one of these studies, the increased risk was estimated to be 4.3-fold (Frisch and Biggar, 1999). Moreover, a similar study showed an increase in tonsillar cancer in women aged >50 years with a history of in situ cervical cancer, as well as an increased incidence of both tonsillar and tongue cancer in the husbands of cervical cancer patients (Hemminki et al., 2000). In contrast, patients with an HPV-unrelated cancer, e.g., colon cancer or breast cancer, had no increased risk of developing tonsillar cancer (Frisch and Biggar, 1999). Since the histology of the oral mucosa resembles that of the uterine cervix and other lower genital localizations, one can anticipate similar HPV infection patterns in the oral cavity as described for the genital tract (Syrjanen, 2003). HPV infection of the cervix is transmitted by sexual contact and there is a correlation between the prevalence of HPV, the number of sexual partners, and a low age at sexual debut (Oriel, 1971; Schiffman and Brinton, 1995; Syrjanen and Syrjanen, 1990). Orogenital contact has also been suggested to lead to HPV infections (Devaraj et al., 2003; Maden et al., 1992; Schwartz et al., 1998; Scully, 2002; Smith et al., 1998). In cervical cancer, which has been studied in more detail, HPV-involved cancer progression has been shown to be a multistep process. This process includes E6 and E7 transcription (of ‘‘oncogenic’’ HPV types), modification of cellular genes, and possibly also genetic susceptibility, an impaired cellmediated immunity, and co-factors such as smoking (Beskow and Gyllensten, 2002; Hanahan and Weinberg, 2000; Schiffman and Brinton, 1995; zur Hausen, 1996, 1999). Less is known about HPV-induced carcinogenesis at other tumor sites. However, it is reasonable to assume that the pathways are similar but not necessarily identical for tonsillar cancer. In the cervix, the immune system usually clears HPV infections within months or years. Only rarely after 10 to 30 years of latency do HPV infections progress to cervical carcinoma (Beskow and Gyllensten, 2002; Rome et al., 1987; Schiffman and Brinton, 1995; Syrjanen and Syrjanen, 1990). It is possible that this is also true for tonsillar cancer; however, the latency period is still unknown. An impaired cellular-mediated immune system (such as in HIV-infected or transplanted patients) results in an increase in HPV-induced lesions as well as an increase in HPV-associated cancers, including tonsillar cancer

HPV in Tonsillar Cancer

65

(Berkhout et al., 1995; Demetrick et al., 1990; de Villiers, 1997; Frisch et al., 2000a; Swoboda and Fabrizii, 1993). Evasion of the cell-mediated immune system is critical for HPV-transformed tumor cells. In cervical cancer, the expression of HLA class I antigens and accessory molecules is often downregulated (Cromme et al., 1994; Stanley, 2001; Stern, 1996). It is possible that similar mechanisms will also be found in tonsillar cancer. The role of humoral immunity in HPV infection is not well understood. In HPV-infected women, an IgG response to HPV-16 and HPV-18 is often observed 4 to 12 months after HPV DNA detection (Carter et al., 1996; Lehtinen and Paavonen, 2001). In the sera of patients with HNSCC, the presence of antibodies against HPV-16 is significantly more frequently observed compared to individuals not having HNSCC (Mork et al., 2001). However, while antibodies against the viral capsid proteins are markers of past or present infection, antibodies against E6 and E7 are markers more clearly associated with malignant disease (Herrero et al., 2003; Lehtinen and Paavonen, 2001).

C. HPV and Prognosis in Tonsillar Cancer In 1998, a survival analysis of 31 patients with tonsillar cancer using tumor pRB expression demonstrated a significantly better survival for patients with pRB-negative tumors (Andl et al., 1998). HPV presence and survival were not analyzed separately. However, there was an indication of a significant correlation between lack of pRB expression and presence of HPV (Andl et al., 1998). It was first reported in 2000 that HPV is a favorable prognostic factor in tonsillar cancer (Gillison et al., 2000; Mellin et al., 2000). In one of these studies (Mellin et al., 2000) on 60 patients with tonsillar cancer, it was found that 52% of the patients with HPV-positive tumors were tumor-free 3 years after diagnosis, as compared to 21% of patients with HPV-negative tumors (Fig. 1). Patients with HPV-positive tumors also exhibited a significantly longer 5-year survival compared to patients with HPV-negative tumors (53.5 compared to 31.5%, p ¼ 0.047, log-rank test). HPV was a favorable prognostic factor independent of tumor stage, age, gender, and grade of differentiation (Mellin et al., 2000). In the study by Gillison et al. (2000) on 253 head and neck cancer patients, 60 had oropharyngeal cancers (mostly tonsillar cancers) and disease-specific survival was significantly improved for the HPV-positive oropharyngeal cancer group compared to the HPV-negative group. However, among patients with non-oropharyngeal cancers, the disease-specific survival was similarly independent of HPV status (Gillison et al., 2000). Accordingly, the prognostic value of HPV did not seem to hold for head and neck cancer in general, but only for tonsillar

66

Hanna Mellin Dahlstrand and Tina Dalianis

Fig. 1 Number and percent of disease-free patients at 3 years after diagnosis for each stage and HPV status. Reproduced from Mellin et al., 2000, with permission from Wiley.

cancer specifically (Gillison et al., 2000; Paz et al., 1997; Riethdorf et al., 1997; Snijders et al., 1996). Additional reports on HPV as a favorable factor for tonsillar cancer have subsequently been reported (Dahlgren et al., 2003; Friesland et al., 2001; Mellin et al., 2002; Ringstro¨ m et al., 2002; Strome et al., 2002). Moreover, in a recent study on the prognostic value of HPV in HNSCC, HPV was not found to have any prognostic value for HNSCC as a whole (Koskinen et al., 2003). However, in this study, all tonsillar tumors were HPV positive and all patients with tonsillar cancer remained alive during the observation period (which for all HNSCC ranged between 1.4 and 89.6 months, mean 24.5 months) (Koskinen et al., 2003). In subsequent studies, the possible importance of the viral load and physical status of HPV on the clinical outcome was evaluated for tonsillar cancer by Mellin et al. (2002) and for HNSCC in general by Koskinen et al. (2003). Mellin et al. (2002) analyzed the presence of HPV in 22 fresh-frozen pretreatment tonsillar samples by general and type-specific PCR using a quantitative PCR. Eleven of the 22 analyzed patients had HPV-16 positive tonsillar cancer; the viral load ranged from 10 to 15,500 HPV-16 copies/cell with a mean of 190 copies/cell. The estimation of viral load in tonsillar cancer was thus in line with the previous study of Klussman et al. (2001). In the six tonsillar cancers and their metastases that were analyzed, the viral copy number per -actin varied between 5.8 and 152.6. Interestingly, Mellin et al. (2002) found that patients with >190 HPV-16 copies in their tumor cells had a significantly longer survival rate than did patients with <60 HPV-16 copies/cell (p ¼ 0.039, log-rank test), as shown in Fig. 2. It is

HPV in Tonsillar Cancer

67

possible that a stronger immune response is generated against tumor cells that contain a high viral content, which may explain why patients with a large viral load have a better survival. Koskinen et al. (2003) studied 61 fresh-frozen biopsies obtained at diagnosis from patients with squamous cell carcinoma of the hypopharynx (10), larynx (18), tongue (15), oral cavity (13), and tonsil (5). The frequency of HPV was determined by SPF10 PCR-screening with a general probe hybridization and INNO-LiPA HPV genotyping assay, while HPV quantification was determined by a real-time quantitative PCR (Koskinen et al., 2003). Using these procedures, 61% of the samples were HPV positive; 5/5 (100%) of the tonsillar samples, 11/15 (73%) of the tongue, and around 50% of the samples from the remaining locations were HPV positive, with HPV-16 as the predominant type (85%). There were large individual variations in HPV viral load in general. However, the median copy numbers of E6 DNA in tonsillar specimens were approximately 80,000 times higher than those in nontonsillar HNSCC types (Koskinen et al., 2003). One reason why HPV may be a favorable prognostic factor in tonsillar cancer is because it is present in large quantities and, hence, may induce a better immune response compared to that induced by HPV in nontonsillar locations. However, the number of tonsillar cancer patients in the study of Koskinen et al. (2003) was small (n ¼ 5). Since all five tonsillar cancer patients remained alive and

Fig. 2 Kaplan-Meier graph showing significantly better disease-specific survival in patients with tumors with HPV-16 copies/-actin compared to patients with tumors with HPV-16 copies/-actin in tonsillar cancer (p ¼ 0.039, log-rank test, n ¼ 11). Reproduced from Mellin et al., 2002, with permission from Wiley.

68

Hanna Mellin Dahlstrand and Tina Dalianis

had HPV-positive tumors, it was not possible to correlate clinical outcome to the presence or absence of HPV or high viral load (Koskinen et al., 2003). Clinical outcome could not be correlated to the physical status of HPV (Koskinen et al., 2003; Mellin et al., 2002). Mellin et al. (2002) studied the physical status of HPV by rliPCR (Kalantari et al., 2001). They found that almost all HPV-16 positive samples were either full length, full length and deleted, or deleted episomal, thus making a comparison between different physical forms impossible. Furthermore, the finding that HPV-16 is mainly episomal in tonsillar cancer was supported by an earlier study where the analysis of physical status was performed by Southern blot and two-dimensional gel electrophoresis. In this study, HPV-16 was found to be episomal, while HPV-33 was both episomal and integrated in tonsillar cancer (Snijders et al., 1992). In the study by Koskinen et al. (2003), the physical status of HPV was based on the assumption that the E2 gene is lost on integration, which is not always the case (Kalantari et al., 2001). A real-time PCR was performed by the amplification of the E2 and E6 genes simultaneously in separate reaction tubes. The presence of equal amounts of E2 and E6 copies was regarded as indicative of the presence of episomal viral genomes, while the presence of E6 amplification without E2 amplification was assumed to be indicative of the virus in an integrated form (Koskinen et al., 2003). Two tonsillar samples were suggested to have episomal forms and three samples were suggested to contain integrated forms of HPV. Since all patients remained alive, it was impossible to correlate the physical status of HPV to clinical outcome. In another study, fluorescence in situ hybridization (FISH) rather than PCR was used to determine the physical state of HPV in HNSCC (including 12 tonsillar carcinomas) (Hafkamp et al., 2003). It was found that 8/12 of the tonsillar tumors exhibited FISH staining, corresponding to integrated HPV-16; 7/8 tumors also harbored episomal HPV (Hafkamp et al., 2003). One tumor contained HPV-16 as well as HPV-18 in integrated and episomal forms (Hafkamp et al., 2003). Although it is not yet clear whether HPV-16 exists mainly in episomal or integrated forms or both, the studies above indicate that episomal HPV-16 can still induce malignant transformation in tonsillar cancer (Hafkamp et al., 2003; Koskinen et al., 2003; Mellin et al., 2002; Snijders et al., 1992). This has been similarly demonstrated in approximately one-third of cervical cancer cases (Cullen et al., 1991; Das et al., 1992; Kalantari et al., 2001; Watts et al., 2002). One mechanism for episomal transformation could be that extra-chromosomal HPV-16 exhibits genetic modifications in the long control region (LCR). This leads to an enhanced activity of the LCR, which may, in turn, influence the promotor activity of E6/E7 transcription (Chen et al., 1997; Dong et al., 1994; May et al., 1994; Watts et al.,

HPV in Tonsillar Cancer

69

2001). Whether the LCR is mutated in a similar way in tonsillar cancer is unknown, but it is possible.

D. HPV and Radiosensitivity in Tonsillar Cancer An alternative explanation for HPV as a positive prognostic factor in tonsillar cancer, other than that HPV-positive tumor cells may be more immune sensitive, is that HPV-positive tonsillar tumors are more radiosensitive than HPV-negative tonsillar tumors. Friesland et al. (2001) examined whether the favorable clinical outcome of patients with HPV-positive tonsillar cancer was due to a greater radiosensitivity of HPV-positive tonsillar cancer compared to HPV-negative tonsillar cancer. Forty patients with tonsillar cancer were selected—21 with tumors in complete remission (CR) after radiotherapy and 19 without complete remission (non-CR) (Friesland et al., 2001). The tumors were analyzed for presence of HPV by PCR and for overexpression of p53 by immunohistochemistry (IHC). The evaluation of response to radiotherapy was performed one month after completion of radiotherapy by clinical examination and, when required, using a biopsy from the primary tumor site (Friesland et al., 2001). The response was classified as CR when evidence of the tumor could not be observed and as non-CR when there was viable tumor remaining. Among the 40 patients, 34 had tumors with amplifiable DNA and could be evaluated for HPV status. There were no statistically significant differences in sensitivity to radiotherapy between patients with tumors with a different HPV or p53 status (assayed by IHC) (Friesland et al., 2001). This was unexpected since both CR and HPV are favorable prognostic factors for survival (Friesland et al., 1999, 2001; Mellin et al., 2000). Furthermore, when following the patients for two years, HPV and p53 status could still not be correlated with the radiotherapy response. In absolute numbers, there were more p53-negative, HPV-negative patients in the non-CR group compared to the CR group; however, the number of patients was too small for further statistical analysis (Friesland et al., 2001). A more recent analysis concerned 65 tonsillar cancer patients who had received preoperative radiotherapy (Mellin, 2002). Although a tendency toward a better response was observed for HPV-positive patients (71% in CR) compared to HPV-negative patients (53% in CR), this difference was not statistically significant (p ¼ 0.134, 2 test) (Mellin, 2002). Moreover, when these patients were included in a Kaplan-Meier survival analysis, it was found that a CR induced by radiotherapy may be more crucial than HPV status (p ¼ 0.00002, log-rank test). In the HPV-positive group, the patients in CR had a significantly better survival compared to the

70

Hanna Mellin Dahlstrand and Tina Dalianis

HPV-positive patients without CR (p ¼ 0.0011, log rank). Similarly, in the HPV-negative group, the patients with CR had a significantly better survival compared to the HPV-negative patients without CR (p ¼ 0.00006, log-rank test). Nevertheless, although not statistically significant, the HPVpositive patients in the CR group had an 82% five-year survival rate compared to 61% in the HPV-negative patients (p ¼ 0.26, log-rank test). In line with these data, a study on HPV and local control after radiotherapy in oropharyngeal cancer reported that HPV-positive tumors seemed to be more sensitive to radiotherapy (Lindel et al., 2001). However, only 14% of the patients were found to be HPV positive in this study and the results were not statistically significant. As has been mentioned, there is still no simple straightforward explanation as to why patients with HPV-positive tonsillar cancer survive better than those with HPV-negative tonsillar cancer. Nonetheless, it is obvious that the explanation cannot be based on differences in radiosensitivity alone. So far there is only a nonsignificant indication in the limited number of cases that have been tested that HPV-positive tonsillar cancer responds slightly better than HPV-negative tonsillar cancer to radiotherapy. It is likely that other tumor biological differences, such as p53 status, between tonsillar cancers harboring oncogenic HPV and tumors lacking HPV may explain differences in radiotherapy sensitivity. This was suggested in a study by Obata et al. (2000) where oropharyngeal cancer with wild-type p53 seemed to respond better to radiotherapy compared to tumors with mutated p53. The role of p53 will be discussed further. In summary, the presence of HPV in tonsillar cancer does not have a major impact on tumor radiosensitivity. The combined data still suggest that the prognostic value of HPV (particularly if correlated to a high viral load) could mainly be due to an immune response to HPV.

E. HPV and Correlation to Cell-Cycle Proteins in Tonsillar Cancer Abrogation of normal p53 function appears to be a vital step for cancer development. This allows the tumor to overcome the normal ‘‘checkpoints’’ of cell regulation and proliferate despite the accumulation of mutations. It is known that E6 (of ‘‘high risk’’ or ‘‘oncogenic’’ HPVs) can bind and degrade p53 (Scheffner et al., 1990; zur Hausen, 1996). Hence, in HNSCC, p53 function may be disabled by the E6 oncogene or by gene aberration in the p53 pathway. Furthermore, E7 of oncogenic HPV types can bind to another vital tumor suppressor pRb, leading to its dysfunction and contributing to dysregulation of the cell cycle (Dyson et al., 1989, 1992; zur Hausen, 1996).

HPV in Tonsillar Cancer

71

Generally, in cervical cancer, most cases are HPV positive and p53 is not mutated. However, in the few cases where cervical cancer has been shown to be HPV negative, p53 mutations have been found (Crook et al., 1992). In tonsillar cancer, the p53 status has yet to be proven and the situation is different and more complex since only about 50% of tonsillar cancers are reported to be HPV positive (Gillison et al., 2000; Mellin et al., 2000). Moreover, several reports have been published on the presence of mutated p53 in normal and pre-neoplastic mucosa of the aerodigestive tract of smokers and in the normal mucosa of aged individuals (Dolcetti et al., 1992; Gusterson et al., 1991; Pavelic et al., 1994). Furthermore, it has been reported that p53 mutation is more common in smokers than in former and nonsmokers, and that HPV is more common in the latter group (Koch et al., 1999). It is thus theoretically possible to anticipate the presence of p53 mutations independent of or in combination with HPV in HNSCC in smokers. Lewensohn-Fuchs et al. (1994) reported the occurrence of aberrant p53 expression measured by immunohistochemistry (IHC) in HPV-positive tonsillar cancer (2 cases). In a later sequence analysis of p53 in one of the HPV-positive tonsillar cancers with aberrant p53 expression, an inframe deletion of intron 7 in the p53 gene was demonstrated (Magnusson et al., 1995). Moreover, it was subsequently shown that normal tissue close to HPV-negative, p53 IHC-negative tumors could express aberrant p53 (Munck-Wikland et al., 1997). A number of reports have been published on the combined presence of HPV and mutated p53 in tonsillar cancers (or oropharyngeal cancers with a dominance of tonsillar cancers) (Balz et al., 2003; Friesland et al., 2001; Gillison et al., 2000; Snijders et al., 1994). In general, when p53 is estimated by IHC, p53 overexpression in tonsillar cancer is demonstrated in approximately 50% of all cases, irrespective of HPV status (Friesland et al., 2001; Snijders et al., 1994). However, when p53 is analyzed by sequencing, mutations are not always observed (Balz et al., 2003; Snijders et al., 1994). In one study, 4/8 (50%) of the HPV-positive tonsillar tumors had elevated p53 expression when tested by IHC, but only 1/10 (10%) showed a p53 mutation (in addition to one silent mutation, yielding no change in amino acids and p53 function) (Snijders et al., 1994). Until 1995, around 40% of HNSCC were found to be p53 mutated when using sequencing methods (Sidransky, 1995). However, in a later HNSCC study by Balz et al. (2003), p53 aberrations were found in 80% of the cases. In this instance, sequencing of the entire coding region of p53 was performed and not restricted to exon 5–8, as commonly done previously. In this same study, p53 mutations were shown to be equally frequent in tumors of the hypopharynx (27/30; 90%), the larynx (35/44; 80%), and the oral cavity (11/14; 79%), but less frequent in oropharyngeal tumors (14/33; 45%). Correspondingly, the prevalence of

72

Hanna Mellin Dahlstrand and Tina Dalianis

HPV-16/18 transcripts was considerably higher in carcinomas of the oropharynx (17/33; 52%) as compared to those of the hypopharynx (8/30; 27%), the oral cavity (4/14; 29%), or the larynx (8/44; 18%) (Balz et al., 2003). Interestingly, as many as 77% of the tumors with wild-type p53 expressed E6, while only 18% of the cases exhibited the combined presence of p53 mutations and E6 transcripts. In another study by Hafkamp et al. (2003), the presence of HPV was observed in 9 out of 16 (53%) oropharyngeal tumors. When these tumors were analyzed by PCR-SSCP for exon 5–8 of the p53 gene, p53 mutations could not be detected (Hafkamp et al., 2003). The fact that tonsillar tumors with HPV demonstrate p53 by IHC and possible p53 overexpression does not exclusively reflect the p53 mutation status (Mineta et al., 1998; Riethdorf et al., 1997). It is known that in normal cells the wild-type p53 tumor suppressor protein has a short half-life and normal p53 is therefore assumed to be undetectable by IHC (Soong et al., 1996). In cells with mutated p53, the level of the protein is stabilized, and the protein appears to be ‘‘overexpressed.’’ This is at least partly due to the inability of mutated p53 to induce transcription of MDM2, a protein that both degrades p53 and inhibits p53 transcription (Haupt et al., 1997; Thut et al., 1995). However, it has also been suggested that wild-type p53 may be stabilized in tumors by disturbances in the degradation of p53 or normally elevated due to the presence of DNA damage (Mineta et al., 1998, and references therein). Furthermore, there may be p53 mutations that do not stabilize p53 sufficiently enough to be detected by IHC. Alternatively, p53 mutations may alter the p53 structure in such a way that the antibody does not react with the epitope of the protein (Saunders et al., 1999). Nevertheless, it has been shown that in 80% of the reported p53 mutations in HNSCC, p53 was also IHC positive, while only 54% of the IHC-positive cases had mutated p53 when examined by sequencing (Riethdorf et al., 1997). With regard to pRB function, in cervical cancer and cervical dysplasia, where pRB is dysfunctional, it has been demonstrated that the cell cycle protein p16INKa is overexpressed (Klaes et al., 2001; Sano et al., 1998). Furthermore, it has been shown that high levels of p16INKa correlate with inactive pRB (Parry et al., 1995). Thus, overexpression of p16INKa can been seen as a surrogate biomarker for the presence of high-risk HPV in these lesions (Sano et al., 1998). Interestingly, in the study by Andl et al. (1998), there was a correlation between the presence of HPV and p16INK4A overexpression with simultaneous pRB downregulation in 6 out of 9 cases. As previously mentioned, Andl and associates also found a correlation between lack of pRB expression and presence of HPV and that the pRB-defective tumors showed an overexpression of p16INK4A (Andl et al., 1998). Similarly, a separate study also demonstrated a correlation between HPV (types 16 and

HPV in Tonsillar Cancer

73

33) and an overexpression of p16INK4A (Klussmann et al., 2003). Sixteen out of 18 HPV-positive tonsillar carcinomas showed p16 expression, whereas no HPV-negative tonsillar carcinomas displayed p16 (Klussmann et al., 2003). To conclude, the data show that p53 mutations can exist in HPV-positive HNSCC (including tonsillar cancer), but that they are less common in HPVpositive than HPV-negative HNSCC despite positive p53 by IHC (Andl et al., 1998; Balz et al., 2003; Brachman et al., 1992; Gillison et al., 2000; Hafkampf et al., 2003; Snijders et al., 1994; van Houten et al., 2001). Moreover, it must be mentioned that p53 DNA contact mutations generally have a strong negative impact on clinical outcome in HNSCC (Erber et al., 1998). It cannot therefore be completely excluded that the presence of wild-type p53, despite p53 IHC staining, contributes to the better clinical outcome in patients with HPV-positive tonsillar cancer. Finally, a few current studies in tonsillar cancer have shown a correlation between the presence of HPV, pRB negativity, and overexpression of p16INK4A (Andl et al., 1998; Klussmann et al., 2003), which is in line with recent findings in cervical cancer (Klaes et al., 2001; Sano et al., 1998).

F. HPV and Genetic Instability in Tonsillar Cancer Regardless of the p53 status or sensitivity to radiotherapy, HPV has still been indicated to be a favorable prognostic factor in tonsillar cancer (Dahlgren et al., 2003; Friesland et al., 2001; Gillison et al., 2000; Mellin et al., 2000, 2002, 2003). To find a possible explanation for this, a number of tonsillar cancer biopsies were examined with regard to their genetic instability (Dahlgren et al., 2003; Mellin et al., 2003). The degree of DNA aberration (diploid or aneuploid) was analyzed by Image Cytometry (ICM) (Mellin et al., 2003) and the chromosomal composition was analyzed by comparative genomic hybridization (CGH) (Dahlgren et al., 2003). All tonsillar cancers showed genetic aberrations. However, HPV-positive tonsillar cancers had a tendency to be somewhat less affected than HPV-negative tonsillar cancers when analyzed both by ICM and by CGH (Dahlgren et al., 2003; Mellin et al., 2003). Using ICM, the degree of DNA aberration was examined to study whether HPV positive and negative tumors differed in DNA content and whether the extent of DNA aberration also affected clinical outcome (Mellin et al., 2003). The DNA content was estimated in 58 primary tonsillar tumors. A normal diploid cell nuclear DNA content was referred to as a 2c value (c is the haploid genome equivalent). The fraction (percent) of cancer cells exceeding 2.5c was indicated as the 2.5c exceeding rate (2.5c ER), while the percent of cancer cells exceeding 5c was referred to as the 5c exceeding

74

Hanna Mellin Dahlstrand and Tina Dalianis

rate (5c ER). Cancer cells with a DNA content value above 2.5c were regarded as either proliferating diploid cells or aneuploid cells, whereas cells with a DNA content value above 5c ER were regarded to be aneuploid (hyperploid). A lesion was classified as diploid if none of the cells exceeded 5c ER and less than 35% of the cells were between 2.5c and 5c ER. Most of the examined tumors showed a high degree of aneuploidy, including a mean of 17.5% of the cells in 5c ER, and only 7 (12%) of the tumors were demonstrated to be diploid. Cancer patients with a 5c ER below the mean value were, to a higher degree, disease-free after 3 years and had a better survival compared to cancer patients with a 5c ER above the mean value. However, these disparities were not statistically significant. HPV-positive tumors tended to have a lower mean 5c ER, 13 as compared to 22% for the HPV-negative tumors (p ¼ 0.066, 2 test). Furthermore, significantly fewer HPV-positive tumors had a 5c ER above the mean value compared to the HPV-negative tumors (p ¼ 0.026, 2 test). Nevertheless, independent of DNA content, patients with HPV-positive cancer were, to a higher degree, disease-free 3 years after diagnosis compared to patients with HPV-negative cancer (Table I) (Mellin et al., 2003). In the study by Dahlgren and associates (Dahlgren et al., 2003), 25 of the original 40 original tonsillar cancer biopsies gave reliable profiles when analyzed by CGH. Of the 25, 15 (60%) were HPV positive. The complete data are summarized in Table II. Among the 15 HPV-positive tonsillar cancers, the gains ranged between 0 and 10 per case (mean 2.3) and the losses between 0 and 6 per case (mean 2.1), which results in an average number of chromosomal aberrations (ANCA) value of 4.5. The most frequently occurring gain in this tumor group was seen on chromosome 3q23-qter, where 11/15 cases (73%) had an increased copy number, whereas the most frequently occurring loss (7 cases, 47%) was seen on chromosome 11q14-q25 (Tables II and III). Amplifications were seen for 3 cases on chromosome 3q with the minimal region 3q24-q27 and chromosomes 8, 9,

Table I HPV Statusa and Meanb Value of 5c ER Correlated to Disease Free 3 Years after Diagnosisc

Disease-free patients

HPVþ and 17.5% in 5c ER

HPVþ and 17.5% in 5c ER

HPV
and 17.5% in 5c ER

HPV
and 17.5% in 5c ER

Total

2/3 (67%)

9/15 (60%)

2/12 (17%)

2/12 (17%)

15/42 (36%)

aAll HPV-positive tumors were also HPV-16 positive, except for one tumor that contained both HPV-16 and HPV-33. bMean value of 5c ER was 17.5%. cReproduced from Mellin et al. (2003) with permission from Anticancer Research.

75

HPV in Tonsillar Cancer

Table II

HPV and CGH Data of Squamous Cell Carcinomas of the Tonsila

Case no.b

HPV

2-2548 3-2549 4-2551

þ þ þ

5-2552 7-2553

þ


9-2759 10-2557

þ þ

11-2758




13-2559 14-2560 15-2561 17-2562 18-2564 22-2568 23-2569 25-2578 28-2571 30-2572


þ þ
þ þ

þ þ

32-2579 36-2573 38-2760






40-2581 41-2574 44-2576 46-2577


þ þ þ

DNA losses – – 3pter-p12, 9pter-p12, 11q13-q25, 18pter-p11.2, 18q21-q23 – 3p, 4p, 5, 9p, 10p, 11q13-q25, 13, 15, 18q 13, 16q 2q35-q37, 3p22-p12, 11q, 13, 14q24-q32, 16 11q14-qter, 13, 21 – 11q13-q25 4p, 11q13-q25, 16q – 4q28-q35, 13 7p, 7q11.1-q21, 14q, 16q – 7q22-q36, 10q23-q26 3p, 14 11p, 11q14-q25, 18q21-q23 – 10q, 11q13-q25 4, 11q13-q25, 13, 18q12-q23

2q14.3-q23 10q, 11q – 11q13-q25, 20pter-p11.1

DNA gains 3q13.1-q29 – 3q23-q29, 5pter-p12, 11q12-q13, 12pter-p11.2, 18q11.1-q21

11q12-q13 3q, 7q, 8q24.3, 11q12-q13, 12pter-p11.2, 14, 19q, 20, 22q13-q13 3q21-qter, 8q23-qter 3q, þþ3q24-q27c, 5, 8, þþ9p, þþ9q13-q34, 10, 11p12, 12, 17, 18 1q21-q41, 3q26.1-qter, 8q21.1-qter, 11q13, 18q, 20q – 3q 16p, 17 7q11.2-q31 – 3q, 8, 10, þþ20p, 20q – 3q, 7p21-p11.2, 7q11.1-q22 3q, 20p þþ3q, þþ8, 10p, 20 – 3q, 17, 20q, 22 1pter-p34.3, 1q, 2q11.1-q31, þþ7q, 8q22-q24.3, 9p24-p12, 11q11-q13, 12p, 14q22-q32, 15, 16p, 17q, þþ17q11.2-q12, 18q11.1-12, 19p 8, 12, 17q25 þþ3q 3 3q21-q29, 20q11.1-q13.3

aReproduced from Dahlgren et al. (2003) with permission from Wiley. bCase number and NCBI SKY/CGH database accession number. cAmplified.

and 20p in one tumor each (Table II). Among the 10 HPV-negative samples, the gains ranged between 0 and 16 (mean 4.0) and the losses between 0 and 9 (mean 2.1), resulting in an ANCA value of 6.1. The most frequent gains were seen on chromosomes 3q26.1-qter 7q11.1-22 and 8q24.3, where

76

Hanna Mellin Dahlstrand and Tina Dalianis

Table III Differences of Gains and Losses of HPVa Positive and Negative Tonsillar Cancerb

All cases HPVþ cases HPV
cases

þ3q

þ7q


11q

15 (60%) 11 (73%) 4 (40%)

4 (16%) 0 (0%) 4 (40%)

11 (44%) 7 (47%) 4 (40%)

þ3q/
11q 9 (36%) 6 (40%) 3 (30%)

aHPV DNA-positive according to PCR data, with GP5þ/6þ or CPI/IIG primers. bReproduced from Dahlgren et al. 2003 with permission from Wiley.

4 cases (40%) showed copy number increases (Table II and III). Two amplifications were seen in this subgroup, one on chromosome 7q and one on chromosome 17q11.2-q12. The most frequent loss mapped to chromosome 11q14q-q25, where 4 cases (40%) had copy number losses (Tables II and III). In summary, when studying all 25 samples, gains on 3q, 8q, 20q, 11q14-qter, and 13q were common. Furthermore, of the 15 cases with a gain on chromosome 3q, 9 (of which 3 had amplification of 3q24-q27) had loss of chromosome 11q14-q25 as well (Table III). In this study, there was also a better survival for patients with HPV-positive tumors (p ¼ 0.002, log-rank test). In addition, some distinct features of the karyograms of the HPV-positive and -negative groups could be identified (Table III). Presence of 3q24-increase was significantly higher (p ¼ 0.049) in HPV-positive tumors (73%) compared to HPV-negative tumors (40%), and an increase of 7q11.2-q22 was only present in HPV-negative tumors (p ¼ 0.017). The minimal affected region on chromosome 3 was 3q24-qter, which also appeared to be amplified in three HPV-positive samples, while no amplification was seen in the HPV-negative group (Dahlgren et al., 2003). The frequent gain on 3q, particularly in the HPV-positive samples, argues in favor of HPV as a possible etiological agent in tonsillar cancer since gain of 3q is a frequent and early event in cervical cancer where the role of HPV is undisputed (Heselmeyer et al., 1996). Moreover, a very similar pattern with a significant chromosomal gain of 3q (with the smallest region 3q22-25) was also obtained in HPV-positive vulvar cancers (Allen et al., 2002). In vulvar cancer, however, a gain of 8q21 (and not chromosome 7) was observed in HPV-negative cancers, while the chromosome arms of 3p and 11q were lost in both categories of vulvar cancer (Allen et al., 2002). Of functional interest, Redon and associates (2002) concluded, after highresolution amplicon mapping and transcriptional analysis, that cyclin L is mapped to 3q25 and is likely to be involved in head and neck cancers. Second, the RNA component of the human telomerase gene hTERC maps to chromosome band 3q26. The fact that this particular genomic imbalance

HPV in Tonsillar Cancer

77

occurs in HPV-positive tumors could point to this gene as a reasonable candidate for functional interactions of HPV proteins with telomerase, as has been demonstrated previously (Veldman et al., 2001). Hence, in line with other studies of HNSCC, the karyotypes were complex and the genomic imbalances were mapped to almost all chromosomes (Struski et al., 2002). The pattern of DNA gains and losses, however, was not random. Common chromosomal imbalances in the entire material were gains of chromosomes 3q24-qter, 8q23-qter, 20q, 11q12-q13, 12p, and 17q, and losses 11q14-q25 and 13. The gains of 3q and 8q in this explicit study of tonsillar cancer were similar to previous observations in HNSCC (Bergamo et al., 2000). Nevertheless, the loss of chromosome 3p, which is common in other reports on HNSCC, was not frequent in these tonsillar cancers (Bockmuhl et al., 1998). This may be due to the fact that the study of Dahlgren et al. (2003) only comprised tonsillar cancer, whereas other reports have included tumors from different locations in the head and neck. Furthermore, gains of 11q12-q13 are common in HNSCC, and several putative oncogenes map to this region. The result that both HPV-positive and HPV-negative tumors contained a high degree of DNA content and that few tumors displayed diploidy was in accordance with what has been observed in cervical cancer (Lorenzato et al., 2001; Rihet et al., 1996; Skyldberg et al., 2001). Nonetheless, independent of genetic instability and chromosomal setup in both the studies of Mellin et al. (2003) and Dahlgren et al. (2003), a better clinical outcome was observed for patients with HPV-positive tumors compared to HPV-negative tumors.

V. HPV AND OTHER TUMORS OF THE HEAD AND NECK The prevalence of HPV in HNSCC varies significantly when comparing different reports (Gillison et al., 2000; Haraf et al., 1996, Klussmann et al., 2001; McKaig et al., 1998; Ringstro¨ m et al., 2002; van Houten et al., 2001). However, in general, HPV is more commonly found in HNSCC patients younger than 60 years of age (Ringstro¨ m et al., 2002; Snijders et al., 1996). The variation in presence of HPV in HNSCC is most likely explained by differences in the composition of the material, including tumor site, the preparation and storage of the material, as well as the methods applied for analysis (Mellin, 2002; Mellin et al., 2002). However, it is also possible that HPV is detected less frequently in locations other than the tonsil, since it is present in lower copies/cell in the other locations (Koskinen et al., 2003). Nevertheless, the overall prevalence of HPV in HNSCC is 14 to

78

Hanna Mellin Dahlstrand and Tina Dalianis

35% when detected by PCR technique, 25% by Southern blot hybridization, 18% by in situ hybridization, and 6% by dot blot hybridization (Gillison et al., 2000; Haraf et al., 1996; Klussmann et al., 2001; McKaig et al., 1998; Ringstro¨ m et al., 2002; van Houten et al., 2001). Moreover, the two most common locations for HPV after tonsillar cancer are in tongue cancer (19–100%) and laryngeal cancer (10–50%) (Koskinen et al., 2003; Matzow et al., 1998; Mellin, 2002; Syrjanen, 2003). HPV-16 is found in the majority of HNSCC, followed by HPV-33. HPV-31, HPV-18, and others are found less frequently (Gillison et al., 2000; Koskinen et al., 2003). In summary, HPV, more specifically HPV-16, is rather commonly found in HNSCC, in particular, in patients below the age of 60. This information could be of potential interest in future HPV preventive and therapeutic vaccination studies.

VI. HPV VACCINES As has been mentioned, HPV is present in approximately half of all tonsillar tumors and is a favorable prognostic factor for clinical outcome (Dahlgren et al., 2003; Friesland et al., 2001; Gillison et al., 2000; Mellin, 2003; Mellin et al., 2000, 2002). This appears not to be due to differences in radiosensitivity between HPV-positive and HPV-negative tonsillar tumors (Friesland et al., 2001), or to the fact that HPV-positive tumors are somewhat less genetically unstable compared to HPV-negative tonsillar tumors (Dahlgren et al., 2003; Mellin et al., 2003). In fact, patients with a high HPV load in their tonsillar tumors seem to have a longer survival compared to those with a low HPV load (Mellin et al., 2002). In addition, HPV has not been found to be a favorable prognostic factor in other nontonsillar HPV-positive HNSCC tumors (Gillison et al., 2000), where the viral load is much lower, as demonstrated by Koskinen et al. (2003). These findings suggest that an immune response against the virus may contribute toward a better clinical outcome (Mellin et al., 2002). If so, it could be important to enhance this antiviral immune response in HPV-positive tonsillar cancer patients in order to increase disease-free survival and also to possibly limit the extent of the potentially disabling treatment that patients receive. Moreover, it is also possible that the induction of an antiviral immune response could be useful for patients with tonsillar cancer or other HNSCC tumors with a lower HPV load. As also has been discussed, several lines of evidence suggest that cellmediated immune responses are important in controlling both HPV infections and HPV-associated neoplasms (for a review, see Devaraj et al., 2003). First, the prevalence of HPV-related diseases, including both infections and

HPV in Tonsillar Cancer

79

neoplasms, is increased in transplant patients and immunodeficiency virus (HIV) infected patients (Schafer et al., 1991). Both these patient types have impaired cellular immunity. Second, animal studies have demonstrated that in immunized animals, the induction of HPV-specific cytotoxic T-cells by HPV peptide vaccination can protect mice from a subsequent challenge with an HPV-16 transformed tumor cell line (Feltkamp et al., 1993). Therefore, HPV vaccines with potential for therapeutic efficacy should generate enhanced HPV-specific cell-mediated immune responses. For prevention, however, it may be sufficient to elicit an antibody response (Breitburd et al., 1995; Heidari et al., 2002; Koutsky et al., 2002; Lehtinen et al., 2003; Suzich et al., 1995; Vlastos et al., 2003; for a review, see Devaraj et al., 2003). Hypothetically, there are at least three different approaches to HPV vaccine development, as reviewed in detail previously (Devaraj et al., 2003). The first approach is to use a prophylactic vaccine and prevent virus from establishing an infection in the epithelium, mainly through the induction of neutralizing antibodies against the viral capsid. In this way, HPV-induced neoplasia will be inhibited. A prophylactic vaccine may, however, be of less benefit to individuals who already have an established infection. The second approach would therefore be to induce a cellular immune response (both CD4þ and CD8þ) to prevent and induce the regression of neoplastic lesions. The strategy used in this kind of therapeutic approach is known as antigen-specific immunotherapy, in which effector cells, particularly T-cells, are primed against HPV epitopes known to be expressed by the neoplastic cells (tumor-specific antigens) such as E6 and E7. Finally, the third approach would be to combine prophylaxis and therapy in one vaccine to cover the needs of people who are newly exposed to high-risk virus and for people with current infection and neoplasia. In the past, HPV vaccine development has been hampered by difficulties in producing HPV in culture and the lack of animal models for HPV infections. However, the discovery that the L1 capsid protein spontaneously assembles to form empty capsids known as virus-like particles (VLPs) (Hagensee et al., 1993; Kirnbauer et al., 1993; Zhou et al., 1991) and the use and development of different animal systems has resulted in the development of potential candidates for future prophylactic and therapeutic vaccines. Knowledge regarding prophylactive vaccines was initially obtained by conducting VLP immunization experiments on cutaneous and mucosal animal papillomaviruses, such as cottontail rabbit papillomavirus (CRPV), canine oral papillomavirus (COPV), and bovine papillomavirus (BPV) (Breithburd et al., 1995; Suzich et al., 1995). VLP vaccination strategies in a murine polyomavirus system were also reported to successfully prevent

80

Hanna Mellin Dahlstrand and Tina Dalianis

viral infection, not only in mice with a normal immune system but also in T-cell deficient immune mice (Heidari et al., 2002; Vlastos et al., 2003). HPV vaccination development has also benefited from the use of tumorigenic mouse lines, such as C3 and TC-1, that express HPV-16 E6 and E7 proteins. These tumor lines have been used in different in vivo studies, demonstrating the immunogenic potential of E6- and E7-derived products against tumor outgrowth (Feltkamp et al., 1993; Lin et al., 1996). The field of HPV vaccination is now in a much stronger position than one or two decades ago. Successful, preventive strategies using HPV VLPs have been reported (Koutsky et al., 2002; Lehtinen et al., 2003; J. T. Schiller, personal communication). The use of preventive vaccines based on HPV-16 and HPV-18 VLPs, such as those used in clinical trials today, may also be beneficial for the prevention of HPV-positive HNSCC, including tonsillar cancers. However, therapeutic vaccines are much less established, even though experimental systems have been described (Feltkamp et al., 1993; Gerard et al., 2001; Gunn et al., 2001; Lin et al., 1996; Meneguzzi et al., 1991; and for review, see Devaraj et al., 2003). Nevertheless, clinical trials using HPV-16, E6/E7 chimeric VLPs, or HPV-16 specific peptides to boost the immune system against established HPV infection are presently ongoing in patients with cervical cancer (J. T. Schiller and C. Melief, personal communications). It is definitely important to use these strategies in HNSCC as well, particularly in tonsillar cancers, where a high HPV load is a favorable prognostic factor, indicating the potential impact of a strong immune response. If these strategies work, it is possible that these patients may need less traumatic treatment in the future than that required today. In summary, successful preventive vaccination strategies to avoid HPV infection in the cervix and ongoing therapeutic vaccination attempts in cervical cancer should also be applied for the prevention of HPV infection in the head and neck region. Adjuvant treatment in patients with HPV-positive tonsillar tumors should also be attempted.

VII. CONCLUSIONS HPV, with a predominance of HPV-16, is present in approximately half of all tonsillar tumors and is a favorable prognostic factor for clinical outcome (Gillison et al., 2000; Mellin et al., 2000, 2002, 2003). This appears not to be due to differences in radiosensitivity between HPV-positive and HPV-negative tonsillar tumors (Friesland et al., 2001) or to the fact that HPV-positive tumors are somewhat less genetically unstable than HPVnegative tonsillar tumors (Dahlgren et al., 2003; Mellin et al., 2003). The fact that patients with high HPV viral loads in their tumors seem to have a

HPV in Tonsillar Cancer

81

longer survival than do patients with lower viral loads suggests that there could be an immune response against the virus that contributes to the better clinical outcome (Mellin et al., 2002). If so, it might be important to therapeutically enhance this antiviral-immune response. This would also emphasize the reason for conducting expanded preventive vaccination trials using preventive anti-HPV-16 vaccines.

ACKNOWLEDGMENTS This work was supported in part by the Swedish Cancer Foundation, the Stockholm Cancer Society, the Stockholm City Council, and Karolinska Institute.

REFERENCES Alani, R. M., and Munger, K. (1998). Human papillomaviruses and associated malignancies. J. Clin. Oncol. 16, 330–337. Allen, D. G., Hutchins, A.-M., Hammet, F., White, D. J., Scurry, J. P., Tabrizi, S. N., Garland, S. M., and Armes, J. E. (2002). Genetic aberrations detected by comparative genomic hybridisation in vulvar cancers. Br. J. Cancer 86, 924–928. Andl, T., Kahn, T., Pfuhl, A., Nicola, T., Erber, R., Conradt, C., Klein, W., Helbig, M., Dietz, A., Weidauer, H., and Bosch, F. X. (1998). Etiological involvement of oncogenic human papillomavirus in tonsillar squamous cell carcinomas lacking retinoblastoma cell cycle control. Cancer Res. 58, 5–13. Balz, V., Scheckenbach, K., Go¨ tte, K., Bockmu¨ hl, U., Petersen, I., and Bier, H. (2003). Is the p53 inactivation frequency in squamous cell carcinomas of the head and neck underestimated? Analysis of p53 exons 2–11 and human papillomavirus 16/18 E6 transcripts in 123 unselected tumor specimens. Cancer Research 63, 1188–1191. Bergamo, N. A., Rogatto, S. R., Poli-Frederico, R. C., Reis, P. P., Kowalski, L. P., Zielenska, M., and Squire, J. A. (2000). Comparative genomic hybridization analysis detects frequent overrepresentation of DNA sequences at 3q, 7p, and 8q in head and neck carcinomas. Cancer Genet. Cytogenet. 119, 48–55. Berkhout, R. J., Tieben, L. M., Smits, H. L., Bavnick, J. N., Vemeer, B. J., and ter Schegget, J. (1995). Nested PCR approach for detection and typing of epidermodysplasia verruciformisassociated human papillomavirus types in cutaneous cancers from renal transplant recipients. J Clin. Microbiol. 33, 690–695. Beskow, A. H., and Gyllensten, U. B. (2002). Host genetic control of HPV 16 titer in carcinoma in situ of the cervix uteri. Int. J. Cancer 101, 526–531. Bockmuhl, U., Wolf, G., Schmidt, S., Schwendel, A., Jahnke, V., Dietel, M., and Petersen, I. (1998). Genomic alterations associated with malignancy in head and neck cancer. Head Neck 20, 145–151. Boice, J. D., Jr., Day, N. E., Andersen, A., Brinton, L. A., Brown, R., Choi, N. W., Clarke, E. A., Coleman, M. P., Curtis, R. E., Flannery, J. T., Hakama, M., Hakulinen, T., Howe, G. R., Jensen, O. M., Kleinerman, R. A., Magnin, D., Magnus, K., Makela, K., Malker, B., Miller, A. B., Nelson, N., Patterson, C. C., Pettersson, F., Pompe-Kirn, V., Primic-Zakelj, M., Prior, P.,

82

Hanna Mellin Dahlstrand and Tina Dalianis

Ravnihar, B., Skeet, R. G., Skjerven, J. E., Smith, P. G., Sok, M., Spengler, R. F., Storm, H. H., Stovall, M., Tomkins, G. W. O., and Wall, C. (1985). Second cancers following radiation treatment for cervical cancer. An international collaboration among cancer registries. J. Natl. Cancer Inst. 74, 955–975. Brachman, D. G., Graves, D., Vokes, E., Beckett, M., Haraf, D., Montag, A., Dunphy, E., Mick, R., Yandell, D., and Weichselbaum, R. R. (1992). Occurrence of p53 gene deletions and human papillomavirus infection in human head and neck cancer. Cancer Res. 52, 4832–4836. Breitburd, F., Kirnbauer, R., Hubbert, N. L., Nonnenmacher, B., Trin-Dinh-Desarquet, C., Orth, G., Schiller, J. T., and Lowy, D. R. (1995). Immunization with viruslike particles from cottontail rabbit papillomavirus (CRPV) can protect against experimental CRPV infection. J. Virol. 69, 3959–3963. Carter, J. J., Koutsky, L. A., Wipf, G. G., Christensen, N. D., Lee, S. K., Kuypers, J., Kiviat, N., and Galloway, D. A. (1996). The natural history of human papillomavirus type 16 capsid antibodies among a cohort of university women. J. Infect. Dis. 174, 927–936. Chen, Z., Storthz, K. A., and Shillitoe, E. J. (1997). Mutations in the long control region of human papillomavirus DNA in oral cancer cells, and their functional consequences. Cancer Res. 57, 1614–1619. Cromme, F. V., Airey, J., Heemels, M. T., Ploegh, H. L., Keating, P. J., Stern, P. L., Meijer, C. J., and Walboomers, J. M. (1994). Loss of transporter protein, encoded by the TAP-1 gene, is highly correlated with loss of HLA expression in cervical carcinomas. J. Exp. Med. 179, 335–340. Crook, T., Wrede, D., Tidy, J., Mason, W., Evans, D., and Vousden, K. (1992). Clonal p53 mutation in primary cervical cancer: Association with human-papillomavirus-negative tumours. Lancet 339, 1070–1073. Cullen, A. P., Reid, R., Campion, M., and Lorincz, A. T. (1991). Analysis of the physical state of different human papillomavirus DNAs in intraepithelial and invasive cervical neoplasm. J. Virol. 65, 606–612. Dahlgren, L., Mellin, H., Wangsa, D., Heselmeyer-Hadad, K., Bjo¨ rnesta˚ hl, L., Lindholm, J., Munck-Wikland, E., Auer, G., Ried, T., and Dalianis, T. (2003). Comparative genomic hybridization analysis of tonsillar cancer reveals a different pattern of genomic imbalances in human papillomavirus-positive and -negative tumors. Int. J. Cancer 107, 244–249. Das, B. C., Shama, J. K., Gopalakrishna, V., and Luthra, U. K. (1992). Analysis by polymerase chain reaction of the physical state of human papillomavirus type 16 DNA in cervical preneoplastic and neoplastic lesions. J. Gen. Virol. 73, 2327–2336. Decker, J., and Goldstein, J. (1982). Risk factors in head and neck cancer. New Engl. J. Med. 306, 1151–1155. Demetrick, D. J., Inoue, M., Lester, W. M., Kingma, I., Duggan, M. A., and Paul, L. C. (1990). Human papillomavirus type 16 associated with oral squamous carcinoma in a cardiac transplant recipient. Cancer 66, 1726–1731. de Roda Husman, A. M., Walboomers, J. M., van den Brule, A. J., Meijer, C. J., and Snijders, P. J. (1995). The use of general primers GP5 and GP6 elongated at their 30 ends with adjacent highly conserved sequences improves human papillomavirus detection by PCR. J. Gen. Virol. 76, 1057–1062. Devaraj, K., Gillison, M. L., and Wu, T.-C. (2003). Development of HPV vaccines for HPVassociated head and neck squamous cell carcinoma. Crit. Rev. Oral. Biol. Med. 14, 345–362. de Villiers, E. M. (1991). Viruses in cancers of the head and neck. Adv. Otolaryngol. 46, 116–123. de Villiers, E. M., Lavergne, D., McLaren, K., and Benton, E. C. (1997). Prevailing papillomavirus types in non-melanoma carcinomas of the skin in renal allograft recipients. Int. J. Cancer 73, 356–361.

HPV in Tonsillar Cancer

83

Dolcetti, R., Doglioni, C., Maestro, R., Gasparotto, D., Barszan, L., Pastore, A., Romanelli, M., and Boiocchi, M. (1992). p53 over-expression is an early event in the development of human squamous-cell carcinoma of the larynx: genetic and prognostic implications. Int. J. Cancer 52, 178–182. Dong, X. P., Stubenrauch, F., Beyer-Finkler, E., and Pfister, H. (1994). Prevalence of deletions of YY1-binding sites in episomal HPV 16 DNA from cervical cancers. Int. J. Cancer 58, 803–808. Dyson, N., Guida, P., Munger, K., and Harlow, E. (1992). Homologous sequences in adenovirus E1A and human papillomavirus E7 proteins mediate interaction with the same set of cellular proteins. J. Virol. 66, 6893–6902. Dyson, N., Howley, P. M., Munger, K., and Harlow, E. (1989). The humanpapillomavirus-16 E7 protein is able to bind to the retinoblastoma gene product. Science 243, 934–937. Erber, R., Conradt, C., Homann, N., Enders, C., Finckh, M., Dietz, A., Weidauer, H., and Bosch, F. X. (1998). TP53 DNA contact mutations are selectively associated with allelic loss and have a strong clinical impact in head and neck cancer. Oncogene 16, 1671–1679. Feltkamp, M. C., Smits, H. L., Vierboom, M. P., Minnaar, R. P., de Jongh, B. M., Drijfhout, J. W., ter Schegget, J., Melief, C. J., and Kast, W. M. (1993). Vaccination with cytotoxic T lymphocyte epitope-containing peptide protects against a tumor induced by human papillomavirus type 16-transformed cells. Eur. J. Immunol. 23, 2242–2249. Forslund, O., Antonsson, A., Nordin, P., Stenquist, B., and Hansson, B. G. (1999). A broad range of human papillomavirus types detected with a general PCR method suitable for analysis of cutaneous tumours and normal skin. J. Gen. Virol. 80, 2437–2443. Friesland, S., Fernberg, J. O., Lundell, G., Munck-Wikland, E., Strander, H., and Lewensohn, R. (1999). Prognostic impact of complete remission after preoperative irradiation of tonsillar carcinoma: A retrospective analysis of the Radiumhemmet data, 1980–1995. Int. J. Radiat. Onc. Biol. and Physics 45, 1259–1266. Friesland, S., Mellin, H., Dalianis, T., Munck-Wikland, E., and Lewensohn, R. (2001). Human papillomavirus (HPV) and p53 immunostaining in advanced tonsillar carcinoma–relation to radiotherapy response and survival. Anticancer Res. 21, 529–534. Frisch, M., and Biggar, R. J. (1999). Aetiological parallel between tonsillar and anogenital squamous-cell carcinomas. Lancet 354, 1442–1443. Frisch, M., Biggar, R. J., and Goedert, J. J. (2000a). Human papillomavirus-associated cancers in patients with human immunodeficiency virus infection and acquired immunodeficiency syndrome. J. Natl. Cancer Inst. 92, 1500–1510. Frisch, M., Hjalgrim, H., Jaeger, A. B., and Biggar, R. J. (2000b). Changing patterns of tonsillar squamous cell carcinoma in the United States. Cancer Causes Control 11, 489–495. Genden, E. M., Ferlito, A., Bradley, P. J., Rinaldo, A., and Scully, C. (2003). Neck disease and distant metastases. Oral Oncol. 39(3), 207–212. Gerard, C. M., Baudson, N., Kraemer, K., Bruck, C., Garcon, N., Paterson, Y., Pan, Z. K., and Pardoll, D. (2001). Therapeutic potential of protein and adjuvant vaccinations on tumour growth. Vaccine 19, 2583–2589. Gillison, M. L., Koch, W. M., Capone, R. B., Spafford, M., Westra, W. H., Wu, L., Zahurak, M. L., Daniel, R. W., Viglione, M., Symer, D. E., Shah, K. V., and Sidransky, D. (2000). Evidence for a causal association between human papillomavirus and a subset of head and neck cancers. J. Natl. Cancer Inst. 92, 709–720. Gillison, M. L., Koch, W. M., and Shah, K. V. (1999). Human papillomavirus in head and neck squamous cell carcinoma: Are some head and neck cancers a sexually transmitted disease? Curr. Opin. Oncol. 11, 191–199. Gillison, M. L., and Shah, K. V. (2001). Human papillomavirus-associated head and neck squamous cell carcinoma: Mounting evidence for an etiologic role for human papillomavirus in a subset of head and neck cancers. Curr. Opin. Oncol. 13, 183–188.

84

Hanna Mellin Dahlstrand and Tina Dalianis

Gissmann, L., Diehl, V., Schultz-Coulon, H. J., and zur Hausen, H. (1982). Molecular cloning and characterization of human papillomavirus DNA derived from a laryngeal papilloma. J. Virol. 44, 393–400. Gunn, G. R., Zubair, A., Peters, C., Pan, Z. K., Wu, T. C., and Paterson, Y. (2001). Two Listeria monocytogenes vaccine vectors that express different molecular forms of human papilloma virus-16 (HPV-16) E7 induce qualitatively different T cell immunity that correlates with their ability to induce regression of established tumors immortalized by HPV-16. J. Immunol. 167, 6471–6479. Gusterson, B. A., Anbazhagan, R., Warren, W., Midgely, C., Lane, D. P., O’Hare, M., Stamps, A., Carter, R., and Jayatilake, H. (1991). Expression of p53 in premalignant and malignant squamous epithelium. Oncogene 6, 1785–1789. Hafkamp, H., Speel, E. J. M., Haesevoets, A., Bot, F. J., Dinjens, W. N. M., Ramaekers, F. C. S., Hopman, A. H. N., and Manni, J. J. (2003). A subset of head and neck squamous cell carcinomas exhibits integration of HPV 16/18 DNA and overexpression of p16INK4A and p53 in the absence of mutations in p53 exons 5–8. Int. J. Cancer 107, 394–400. Hagensee, M. E., Yaegashi, N., and Galloway, D. (1993). Self-assembly of human papillomavirus type 1 capsids by expression of the L1 protein alone or by coexpression of the L1 and L2 capsid proteins. J. Virol. 67, 315–322. Hagmar, B., Johansson, B., Kalantari, M., Petersson, Z., Skyldberg, B., and Walaas, L. (1992). The incidence of HPV in a Swedish series of invasive cervical carcinoma. Med. Oncol. Tumor Pharmacother. 9, 113–117. Hanahan, D., and Weinberg, R. A. (2000). The hallmarks of cancer. Cell 100, 57–70. Haraf, D. J., Nodzenski, E., Brachman, D., Mick, R., Montag, A., Graves, D., Vokes, E. E., and Weichselbaum, R. R. (1996). Human papilloma virus and p53 in head and neck cancer: Clinical correlates and survival. Clin. Cancer Res. 2, 755–762. Haupt, Y., Maya, R., Kazaz, A., and Oren, M. (1997). Mdm2 promotes the rapid degradation of p53. Nature 387, 296–299. Heidari, S., Vlastos, A., Ramqvist, T., Clark, B., Griffin, M. I., Garcia, M., Perez, M., Amati, P., and Dalianis, T. (2002). Immunization of T-cell deficient mice against polyomavirus infection using viral pseudocapsids or temperature sensitive mutants. Vaccine 20, 1571–1578. Hemminki, K., Dong, C., and Frisch, M. (2000). Tonsillar and other upper aerodigestive tract cancers among cervical cancer patients and their husbands. Eur. J. Cancer 9, 433–437. Herrero, R., Castellsague, X., Pawlita, M., Lissowska, J., Kee, F., Balaram, P., Rajkumar, T., Sridhar, H., Rose, B., Pintos, J., Fernandez, L., Idris, A., Sanchez, M. J., Nieto, A., Talamini, R., Tavani, A., Bosch, F. X., Reidel, U., Snijders, P. J., Meijer, C. J., Viscidi, R., Munoz, N., and Franceschi, S. (2003). IARC Multicenter Oral Cancer Study Group. Human papillomavirus and oral cancer: The International Agency for Research on Cancer multicenter study. J. Natl. Cancer Inst. 95, 1772–1783. Heselmeyer, K., Schro¨ ck, E., Du Manior, S., Blegen, H., Shah, K., Steinbeck, R., Auer, G., and Ried, T. (1996). Gain of chromosome 3q defines the transition from severe dysplasia to invasive carcinoma of the uterine cervix. Proc. Natl. Acad. Sci. USA 93, 479–484. Kalantari, M., Blennow, E., Hagmar, B., and Johansson, B. (2001). Physical state of HPV16 and chromosomal mapping of the integrated form in cervical carcinomas. Diagn. Mol. Pathol. 10, 46–54. Kirnbauer, R., Taub, J., Greenstone, H., Roden, R., Durst, M., Gissman, L., Lowy, D. R., and Schiller, J. T. (1993). Efficient self-assembly of human papillomavirus type 16 L1 and L1-L2 into virus-like particles. J. Virol. 67, 6929–6936. Klaes, R., Friedrich, T., Spitkovsky, D., Ridder, R., Rudy, W., Petry, U., Dallenbach-Hellweg, G., Schmidt, D., and von Knebel Doeberitz, M. (2001). Overexpression of p16(INK4A) as a

HPV in Tonsillar Cancer

85

specific marker for dysplastic and neoplastic epithelial cells of the cervix uteri. Int. J. Cancer 92, 276–284. Klussmann, J. P., Gultekin, E., Weissenborn, S. J., Wieland, U., Dries, V., Dienes, H. P., Eckel, H. E., Pfister, H. J., and Fuchs, P. G. (2003). Expression of p16 protein identifies a distinct entity of tonsillar carcinomas associated with human papillomavirus. Am. J. Pathol. 162, 747–753. Klussmann, J. P., Weissenborn, S. J., Weiland, U., Dries, V., Kolligs, J., Jungehuelsing, M., Eckel, H. E., Dienes, H. P., Pfister, H. J., and Fuchs, P. G. (2001). Prevalence, distribution, and viral load of human papillomavirus 16 DNA in tonsillar carcinomas. Cancer 92, 2875–2884. Koch, W. M., Lango, M., Sewell, D., Zahurak, M., and Sidransky, D. (1999). Head and neck cancer in nonsmokers: A distinct clinical and molecular entity. Laryngoscope 109, 1544–1551. Koskinen, W. J., Chen, R. W., Leivo, I., Ma¨ kitie, A., Ba¨ ck, L., Kontio, R., Suuronen, R., Lindqvist, C., Auvinen, E., Molijn, A., Quint, W. G., Vaheri, A., and Altonen, L.-M. (2003). Prevalence and physical status of human papillomavirus in squamous cell carcinomas of the head and neck. Int. J. Cancer 107, 401–406. Koutsky, L. A., Ault, K. A., Wheeler, C. M., Brown, D. R., Barr, E., Alvarez, F. B., Chiacchierini, L. M., and Jansen, K. U. (2002). A controlled trial of a human papillomavirus type 16 vaccine. N. Engl. J. Med. 347, 1645–1651. Lehtinen, M., and Dillner, J. (2002). Preventive human papillomavirus vaccination. Sex Transm. Infec. 78, 4–6. Lehtinen, M., Malm, C., Apter, D., Heikkila¨ , R., Heino, P., Rimpila¨ , K., Ziliacus, R., and Paavonen, J. (2003). Preventive vaccines against papillomavirus and cervix cancer will soon enter clinical practice (Swedish). La¨kartidningen 100, 3408–3412. Lehtinen, M., and Paavonen, J. (2001). Efficacy of preventive human papillomavirus vaccination. Int. J. STD AIDS 12, 771–776. Lewensohn-Fuchs, I., Munck-Wikland, E., Berke, Z., Magnusson, K., Pallesen, G., Auer, G., ˚ berg, B., Rubio, C., Kuylenstierna, R., Wiman, K. G., and Lindholm, J., Linde, A., A Dalianis, T. (1994). Involvement of aberrant p53 expression and human papillomavirus in carcinoma of the head, neck, and esophagus. Anticancer Res. 14, 1281–1286. Licitra, L., Bernier, J., Grandi, C., Merlano, M., Bruzzi, P., and Lefebvre, J. L. (2002). Cancer of the oropharynx. Crit. Rev. Oncol. Hematol. 41, 107–122. Lin, K.-Y., Guarnieri, F. G., Staveley-O’Carroll, K. F., Levitsky, H. I., August, J. T., Pardoll, D. M., and Wu, T. C. (1996). Treatment of established tumors with a novel vaccine that enhances major histocompatibility class II presentation of tumor antigen. Cancer Res. 56, 21–26. Lindel, K., Beer, K. T., Laissue, J., Greiner, R. H., and Aebersold, D. M. (2001). Human papillomavirus positive squamous cell carcinoma of the oropharynx: A radiosensitive subgroup of head and neck carcinoma. Cancer 92, 805–813. Ling, M., Kanayama, M., Roden, R., and Wu, T. C. (2000). Preventive and therapeutic vaccines for human papillomavirus-associated cervical cancers. J. Biomed. Sci. 7, 341–356. Lorenzato, M., Clavel, C., Masure, M., Nou, J. M., Bouttens, D., Evrard, G., Bory, J. P., Maugard, B., Quereux, C., and Birembaut, P. (2001). DNA image cytometry and human papillomavirus (HPV) detection help to select smears at high risk of high-grade cervical lesions. J. Pathol. 194, 171–176. Maden, C., Beckman, A. M., Thomas, D. B., McKnight, B., Sherman, K. J., Ashley, R. L., Corey, L., and Daling, J. R. (1992). Human papillomaviruses, herpes simplex viruses, and the risk of oral cancer in men. Am. J. Epidemiology 135, 1093–1102. Magnusson, K. P., Berke, Z., Munck-Wikland, E., Lewensohn-Fuchs, I., Klein, G., Dalianis, T., and Wiman, K. G. (1995). In frame deletion of intron 7 in the p53 gene in a human tonsil tumor. Int. J. Oncolog. 6, 27–29.

86

Hanna Mellin Dahlstrand and Tina Dalianis

Manos, M. M., Ting, Y., Wright, D. K., Lewis, A. J., Broker, T. R., and Wolinsky, S. M. (1989). The use of polymerase chain reaction amplification for the detection of genial human papillomaviruses. Molecular Diagnostics of Human Cancer, Cancer Cells 7, 209–214. Mashberg, A., and Samit, A. (1995). Early diagnosis of asymptomatic oral and oropharyngeal squamous cancers. CA Cancer J. Clin. 45, 328–351. Matzow, T., Boysen, M., Kalantari, M., Johansson, B., and Hagmar, B. (1998). Low detection rate of HPV in oral and laryngeal carcinomas. Acta Oncol. 37, 73–76. May, M., Dong, X. P., Beyer-Finkler, E., Stubenrauch, F., Fuchs, P. G., and Pfister, H. (1994). The E6/E7 promoter of extrachromosomal HPV16 DNA in cervical cancers escapes from cellular repression by mutation of target sequences for YY1. EMBO J. 13, 1460–1466. Meneguzzi, G., Cerni, C., Kieny, M. P., and Lathe, R. (1991). Immunization against human papillomavirus type 16 tumor cells with recombinant vaccinia viruses expressing E6 and E7. Virology 181, 62–69. McKaig, R. G., Baric, R. S., and Olshan, A. F. (1998). Human papillomavirus and head and neck cancer: Epidemiology and molecular biology. Head and Neck 20, 250–265. Mellin, H. (2002). Human Papillomavirus in Tonsillar Cancer. Karolinska University Press, Stockholm. Mellin, H., Dahlgren, L., Munck-Wikland, E., Lindholm, J., Rabbani, H., Kalantari, M., and Dalianis, T. (2002). Human papillomavirus type 16 is episomal and a high viral load may be correlated to better prognosis in tonsillar cancer. Int. J. Cancer 102, 152–158. Mellin, H., Friesland, S., Auer, G., Dalianis, T., and Munck-Wikland, E. (2003). Human papillomavirus and DNA ploidy in tonsillar cancer–correlation to prognosis. Anticancer Res. 23, 2821–2828. Mellin, H., Friesland, S., Lewensohn, R., Dalianis, T., and Munck-Wikland, E. (2000). Human papillomavirus (HPV) DNA in tonsillar cancer: Clinical correlates, risk of relapse, and survival. Int. J. Cancer 89, 300–304. ˚ kerwall, J., and Wenerberg, J. (1998). p53 Mineta, H., Borg, A˚ ., Dictor, M., Wahlberg, P., A mutation, but not p53 overexpression, correlates with survival in head and neck squamous cell carcinoma. Br. J. Cancer 78, 1084–1090. Mork, J., Lie, A. K., Glattre, E., Hallmans, G., Jellim, E., Koskela, P., Moller, B., Pukkala, E., Schiller, J. T., Youngman, L., Lehtinen, M., and Dillner, J. (2001). Human papillomavirus infection as a risk factor for squamous-cell carcinoma of the head and neck. N. Engl. J. Med. 344, 1125–1131. Munck-Wikland, E., Blegen, H., Lewensohn-Fuchs, I., Malec, M. J., Auer, G., Kuylenstierna, R., Dalianis, T., Wiman, G. K., and Lindholm, J. (1997). P53 overexpression in normal and dysplastic cells adjacent to p53 and HPV negative invasive tumours. Int. J. Oncology 11, 97–104. Naghashfar, Z., Sawada, E., Kutcher, M. J., Swancar, J., Gupta, J., Daniel, R., Kashima, H., Woodruff, J. D., and Shah, K. (1985). Identification of genital tract papillomaviruses HPV-6 and HPV-16 in warts of the oral cavity. J. Med. Virol. 17, 313–324. Niedobitek, G., Pitteroff, S., Herbst, H., Shepherd, P., Finn, T., Anagnostopoulos, I., and Stein, H. (1990). Detection of human papillomavirus type 16 DNA in carcinomas of the palatine tonsil. J. Clin. Pathol. 43, 918–921. Obata, A., Eura, M., Sasaki, J., Saya, H., Chikamatsu, K., Tada, M., Iggo, R. D., and Yumoto, E. (2000). Clinical significance of p53 functional loss in squamous cell carcinoma of the oropharynx. Int. J. Cancer 89, 187–193. Oriel, J. D. (1971). Natural history of genital warts. Br. J. Vener. Dis. 47, 1–13. Parry, D., Bates, S., Mann, D. J., and Peters, G. (1995). Lack of cyclin D-Cdk complexes in Rb-negative cells correlates with high levels of p16INK4/MTS1 tumour suppressor gene product. EMBO J. 14, 503–511.

HPV in Tonsillar Cancer

87

Pavelic, Z. P., Li, Y. Q., Stambrook, P. J., McDonald, J. S., Munck-Wickland, E., Pavelic, K., Dacic, S., Danilovic, Z., Pavelic, L., Mugge, R. E., Wilson, K., Nguyen, C., and Gluckman, J. (1994). Overexpression of p53 protein is common in premalignant head and neck lesions. Anticancer Res. 14, 2259–2266. Paz, I. B., Cook, N., Odom-Maryon, T., Xie, Y., and Wilczynski, S. P. (1997). Human papillomavirus (HPV) in head and neck cancer. An association of HPV 16 with squamous cell carcinoma of Waldeyer’s tonsillar ring. Cancer 79, 595–604. Pugliano, F. A., Piccirillo, J. F., Zequeira, M. R., Emami, B., Perez, C. A., Simpson, J. R., and Fredrickson, J. M. (1997). Clinical-severity staging system for oropharyngeal cancer: Fiveyear survival rates. Arch. Otolaryngol. Head Neck Surg. 123, 1118–1124. Rabkin, C. S., Biggar, R. J., Melbye, M., and Curtis, R. E. (1992). Second primary cancers following anal and cervical carcinoma: Evidence of shared etiologic factors. Am. J. Epidemiol. 136, 54–58. Redon, R., Hussenet, T., Bour, G., Caulee, K., Jost, B., Muller, D., Abecassis, J., and du Manoir, S. (2002). Amplicon mapping and transcriptional analysis pinpoint cyclin L as a candidate oncogene in head and neck cancer. Cancer Res. 62, 6211–6217. Riethdorf, S., Friedrich, R. E., Ostwald, C., Barten, M., Gogacz, P., Gundlach, K. K., Schlechte, H., Bregenzer, T., Riethdorf, L., and Loning, T. (1997). p53 gene mutations and HPV infection in primary head and neck squamous cell carcinomas do not correlate with overall survival: A long-term follow-up study. J. Oral Pathol. Med. 26, 315–321. Ringstro¨ m, E., Peters, E., Hasegawa, E. M., Posner, M., Liu, M., and Kelsey, K. T. (2002). Human papillomavirus type 16 and squamous cell carcinoma of the head and neck. Clin. Cancer Res. 8, 3187–3192. Rihet, S., Lorenzato, M., and Clavel, C. (1996). Oncogenic human papillomaviruses and ploidy in cervical lesions. J. Clin. Pathol. 49, 892–896. Rome, R. M., Chanen, W., and Pagano, R. (1987). The natural history of human papillomavirus (HPV) atypia of the cervix. Aust. N. Z. J. Obstet. Gynaecol. 27, 287–290. Sano, T., Oyama, T., Kashiwabara, K., Fukuda, T., and Nakajima, T. (1998). Expression status of p16 protein is associated with human papillomavirus oncogenic potential in cervical and genital lesions. Am. J. Pathol. 153, 1741–1748. Saunders, M. E., MacKenzie, R., Shipman, R., Fransen, E., Gilbert, R., and Jordan, R. C. (1999). Patterns of p53 gene mutations in head and neck cancer: Full-length gene sequencing and results of primary radiotherapy. Clin. Cancer Res. 5, 2455–2463. Schafer, K., Friedmann, W., Mielke, M., Schwartlander, B., and Koch, M. A. (1991). The increased frequency of cervical dysplasia-neoplasia in women infected with the human immunodeficiency virus is related to the degree of immunosuppression. Am. J. Obstet. Gynecol. 164, 593–599. Scheffner, M., Werness, B. A., Huibregtse, J. M., Levine, A. J., and Howley, P. M. (1990). The E6 oncoprotein encoded by human papillomavirus types 16 and 18 promotes the degradation of p53. Cell 63, 1129–1136. Schiffman, M. H., and Brinton, L. A. (1995). The epidemiology of cervical carcinogenesis. Cancer 76, 1888–1901. Scully, C. (2002). Oral squamous cell carcinoma; from an hypothesis about a virus, to concern about possible sexual transmission. Oral Oncol. 38, 227–234. Schwartz, S. M., Daling, J. R., Doody, D. R., Wipf, G. C., Carter, J. J., Madeleine, M. M., Mao, E. J., Fitzgibbons, E. D., Huang, S., Beckmann, A. M., McDougall, J. K., and Galloway, D. A. (1998). Oral cancer risk in relation to sexual history and evidence of human papillomavirus infection. J. Natl. Cancer Inst. 90, 1626–1636. Sidransky, D. (1995). Molecular genetics of head and neck cancer. Curr. Opin. Oncol. 7, 229–233.

88

Hanna Mellin Dahlstrand and Tina Dalianis

Skyldberg, B., Fujioka, K., Hellstro¨ m, A. C., Sylven, L., Moberger, B., and Auer, G. (2001). Human papillomavirus infection, centrosome aberration, and genetic stability in cervical lesions. Mod. Pathol. 14, 279–284. Smith, E. M., Hoffman, H. T., Summersgill, K. S., Kirchner, H. L., Turek, L. P., and Haugen, T. H. (1998). Human papillomavirus and risk of oral cancer. Laryngoscope 108, 1098–1103. Snijders, P. J., Cromme, F. V., van den Brule, A. J. C., Meijer, C. J., and Walboomers, J. M. (1994). Papillomaviruses and cancer of the upper digestive and respiratory tracts. Curr. Top. Microbiol. Immunol. 186, 177–198. Snijders, P. J., Cromme, F. V., van den Brule, A. J., Schrijnemakers, H. F., Snow, G. B., Meijer, C. J., and Walboomers, J. M. (1992). Prevalence and expression of human papillomavirus in tonsillar carcinomas, indicating a possible viral etiology. Int. J. Cancer 51, 845–850. Snijders, P. J. F., Scholes, A. G. M., Hart, A., Jones, A. S., Vaughan, E. D., Woolgar, J. A., Meijer, C. J. L. M., Walboomers, J. M., and Field, J. K. (1996). Prevalence of mucosotropic human papillomaviruses in squamous-cell carcinoma of the head and neck. Int. J. Cancer 66, 464–469. Soong, R., Robbins, P. D., Dix, B. R., Grieu, F., Lim, B., Knowles, S., Williams, K. E., Turbett, G. R., House, A. K., and Iacopetta, B. J. (1996). Concordance between p53 protein overexpression and gene mutation in a large series of common human carcinomas. Hum. Pathol. 27, 1050–1055. Stanley, M. (2001). Immunobiology of papillomavirus infections. J. Reprod. Immunol. 52, 45–59. Stern, P. L. (1996). Immunity to human papillomavirus-associated cervical neoplasia. Adv. Cancer Res. 69, 175–211. Strome, S. E., Savva, A., Brisse, A. E., Gostout, B. S., Lewis, J., Clayton, A. C., McGovern, R., Weaver, A. L., Persing, D., and Kasperbauer, J. L. (2002). Squamous cell carcinoma of the tonsils: A molecular analysis of HPV associations. Clin. Cancer Res. 8, 1093–1100. Struski, S., Doco-Fenzy, M., and Cornillet-Lefebvre, P. (2002). Compilation of published comparative genomic hybridization studies. Cancer Genet. Cytogenet. 135, 63–90. Suzich, J. A., Ghim, S. J., Palmer-Hill, F. J., White, W. I., Tamura, J. K., Bell, J. A., Newsome, J. A., Jenson, A. B., and Schlegel, R. (1995). Systemic immunization with papillomavirus L1 protein completely prevents the development of viral mucosal papillomas. Proc. Natl. Acad. Sci. USA 92, 11553–11557. Swoboda, A., and Fabrizii, V. (1993). Tonsillar carcinoma in a renal graft recipient treated with cyclosporine A. Clin. Nephrol. 39, 272–274. Syrjanen, S. M. (1990). Basic concepts and practical applications of recombinant DNA techniques in detection of human papillomavirus (HPV) infection. Review article. Apmis 98, 95–110. Syrjanen, S. M. (2003). Human papillomavirus infections and oral tumors. Med. Microbiol. Immunol. 192, 123–128. Syrjanen, K., and Syrjanen, S. (1990). Epidemiology of human papilloma virus infections and genital neoplasia. Scand. J. Infect. Dis 69(Suppl.), 7–17. Syrjanen, K., Syrjanen, S., Lamberg, M., Pyrho¨ nen, S., and Nuutinen, J. (1983). Morphological and immunohistochemical evidence suggesting human papillomavirus (HPV) involvement in oral squamous cell carcinogenesis. Int. J. Oral Surg. 12, 418–424. Thut, C. J., Chen, J. L., Klemm, R., and Tjian, R. (1995). p53 transcriptional activation mediated by coactivators TAFII40 and TAFII60. Science 267, 100–104. Tieben, L. M., ter Schegget, J., Minnaar, R. P., Bouwes Bavnick, J. N., Berkhout, R. J., Vermeer, B. J., Jebbink, M. F., and Smits, H. L. (1993). Detection of cutaneous and genital HPV types in clinical samples by PCR using consensus primers. J. Virol. Methods 42, 265–279. van Houten, V. M., Snijders, P. J., van den Brekel, M. W., Kummer, J. A., Meijer, C. J., van Leeuwen, B., Denkers, F., Smeele, L. E., Snow, G. B., and Brakenhoff, R. H. (2001).

HPV in Tonsillar Cancer

89

Biological evidence that human papillomaviruses are etiologically involved in a subgroup of head and neck squamous cell carcinomas. Int. J. Cancer 93, 232–235. Veldman, T., Horikawa, I., Barrett, J. C., and Schlegel, R. (2001). Transcriptional activation of the telomerase hTERT gene by human papillomavirus type 16 E6 oncoprotein. J. Virol. 75, 4467–4472. Vlastos, A., Andreasson, K., Tegerstedt, K., Hollanderova, D., Heidari, S., Forstova, J., Ramqvist, T., and Dalianis, T. (2003). VP1 pseudocapsids, but not a glutathione-S-transferase VP1 fusion protein, prevent polyomavirus infection in a T-cell immune deficient experimental mouse model. J. Med. Virol. 70, 293–300. Watts, K. J., Thompson, C. H., Cossart, Y. E., and Rose, B. R. (2001). Variable oncogene promoter activity of human papillomavirus type 16 cervical cancer isolates from Australia. J. Clin. Microbiol. 39, 2009–2014. Watts, K. J., Thompson, C. H., Cossart, Y. E., and Rose, B. R. (2002). Sequence variation and physical state of human papillomavirus type 16 cervical cancer isolates from Australia and New Caledonia. Int. J. Cancer 97, 868–874. Wilczynski, S. P., Lin, B. T., Xie, Y., and Paz, I. B. (1998). Detection of human papillomavirus DNA and oncoprotein overexpression are associated with distinct morphological patterns of tonsillar squamous cell carcinoma. Am. J. Pathol. 152, 145–156. Zhou, J., Sun, X. Y., Stenzel, D. J., and Frazer, I. H. (1991). Expression of vaccinia recombinant HPV 16 L1 and L2 ORF proteins in epithelial cells is sufficient for assembly of HPV virionlike particles. Virology 185, 251–257. zur Hausen, H. (1976). Condylomata acuminata and human genital cancer. Cancer Res. 36, 794. zur Hausen, H. (1996). Papillomavirus infections—a major cause of human cancers. Biochim. Biophys. Acta 1288, F55–F78. zur Hausen, H. (1999). Papillomaviruses in human cancers. Proc. Assoc. Am. Physicians 111, 581–587.

T-Cell Transformation and Oncogenesis by g2-Herpesviruses Armin Ensser and Bernhard Fleckenstein Institut fu¨r Klinische und Molekulare Virologie, Friedrich-Alexander-Universita¨t Erlangen-Nu¨rnberg, 91054 Erlangen, Germany

I. Natural Occurrence of the Viruses and Pathology II. Genome Structure III. Replication and Persistence of T-Lymphotropic Rhadinoviruses A. The Immediate-Early and Early Genes B. The Latency-Associated Nuclear Antigen IV. Viral Homologs to Cellular Genes A. Homologs to Enzymes of Nucleotide Metabolism and Growth Regulation B. Complement Regulatory Proteins C. Antiapoptotic Viral Proteins D. Viral Cytokines and Receptor Proteins E. The Viral Superantigen Homolog and a New Signaling Adaptor Protein V. Oncogenesis A. Transformation-Related Proteins of Rhadinoviruses B. The Transformation-Associated Protein STP C. The Tyrosine Kinase–Interacting Protein (Tip) of Herpesvirus saimiri Subgroup C D. The Tio Oncogene of Herpesvirus ateles VI. Growth Transformation of Human T Cells by Rhadinoviruses A. Establishing Transformed T-Cell Lines B. The Phenotype of Rhadinovirus Transformants C. Application of T-Cell Growth Transformation VII. Rhadinovirus Vectors for Gene Therapy VIII. Conclusions References

g2-Herpesviruses, also termed rhadinoviruses, have long been known as animal pathogens causing lymphoproliferative diseases such as malignant catarrhal fever in cattle or T-cell lymphoma in certain Neotropical primates. The rhadinovirus prototype is Herpesvirus saimiri (HVS), a T-lymphotropic agent of squirrel monkeys (Saimiri sciureus); Herpesvirus ateles (HVA) is closely related to HVS. The first human rhadinovirus, human herpesvirus type 8 (HHV-8), was discovered a decade ago in Kaposi’s sarcoma (KS) biopsies. It was found to be strongly associated with all forms of KS, as well as with multicentric Castleman’s disease and primary effusion lymphoma (PEL). Since DNA of this virus is regularly found in all KS forms, and specifically in the spindle cells of KS, it was also termed KS-associated herpesvirus (KSHV). Several simian rhadinoviruses related to KSHV have been discovered in various Old World primates, Advances in CANCER RESEARCH 0065-230X/05 $35.00

91

Copyright 2005, Elsevier Inc. All rights reserved

92

Armin Ensser and Bernhard Fleckenstein though they seem only loosely associated with pathogenicity or tumor induction. In contrast, HVS and HVA cause T-cell lymphoma in numerous non-natural primate hosts; HVS strains of the subgroup C are capable of transforming human and simian T-lymphocytes to continuous growth in cell culture and can provide useful tools for T-cell immunology or gene transfer. Here, we describe their natural history, genome structure, biology, and pathogenesis in T-cell transformation and oncogenesis. ß 2005 Elsevier Inc.

I. NATURAL OCCURRENCE OF THE VIRUSES AND PATHOLOGY The rhadinovirus (g2-herpesvirus) prototype, Herpesvirus saimiri (HVS), is regularly found in squirrel monkeys (Saimiri sciureus), while the related Herpesvirus ateles (HVA) can be isolated at a high rate from spider monkeys (Ateles spp., A.) (reviewed in Fleckenstein and Desrosiers, 1982). The natural habitat of these primate host species is in South and Middle American rainforests. Squirrel monkeys are usually infected via saliva within the first 2 years of life, and HVS can be isolated from peripheral blood cells of the persistently infected squirrel monkeys by co-cultivation on permissive cell cultures. Several animals have been found to be coinfected with isolates of different HVS subgroups (see following text). The virus does not cause disease or tumors, and it establishes lifelong persistence in the species (Melendez et al., 1968). The HVA isolate #810 from A. geoffroyi (Melendez et al., 1972) is officially classified as ateline herpesvirus type 2, whereas isolate #73 and related strains (#87, 93, 94) from A. paniscus were designated as ateline herpesvirus type 3 (Falk et al., 1974). Like HVS, HVA is not pathogenic in its natural host. In other New World primate species such as tamarins (Saguinus spp., S.), common marmosets (Callithrix jacchus, C.), or owl monkeys (Aotus trivirgatus), the experimental infection with HVS causes acute peripheral T-cell lymphoma within less than 2 months (Melendez et al., 1969; Wright et al., 1976; reviewed in Fleckenstein and Desrosiers, 1982). Similar to HVS, HVA causes acute T-cell lymphoma in various New World primate species including tamarins (S. oedipus) and owl monkeys (Hunt et al., 1972). The pathological changes are similar to those observed after HVS infection. The experimental infection is usually performed intramuscularly or intravenously. Intramuscular injection of purified HVS virion DNA also causes disease in susceptible primates (Fleckenstein et al., 1978b). HVS strains were classified into the three subgroups—A, B, and C—depending on the pathogenic properties and on the sequence divergence in the left-terminal nonrepetitive genomic region (Desrosiers and Falk, 1982; Medveczky et al., 1984, 1989). The major representative strains are

T-Cell Transformation and Oncogenesis

93

A11 (Falk et al., 1972) for subgroup A, B-S295C (Melendez et al., 1968) and B-SMHI (Daniel et al., 1975) for subgroup B, and C488 (Biesinger et al., 1990) and C484 (Medveczky et al., 1984) for subgroup C. Viruses of subgroup B are considered to be less oncogenic; subgroup C strains have the strongest oncogenic properties. Tamarins are susceptible to viruses of all subgroups, while subgroup B viruses were reported not being able to cause disease in adult common marmosets (reviewed in Fickenscher and Fleckenstein, 2002). The parental strain C488, as well as various viral deletion mutants, cause acute peripheral T-cell lymphoma within only a few weeks in common marmosets or in cottontop tamarins (S. oedipus) (Duboise et al., 1998a,b; Ensser et al., 2001; Glykofrydes et al., 2000; Knappe et al., 1998a,b). Remarkably, high intravenous doses of HVS strain C488 can induce a similar fulminant disease in Old World monkeys such as rhesus and cynomolgus monkeys (Macaca mulatta; M. fascicularis). The pathological findings in macaques were similar to those in New World primates, and the disease in cynomolgus monkeys was described as a pleomorphic peripheral T-cell lymphoma or, alternatively, as a pleomorphic T-lymphoproliferative disorder (Alexander et al., 1997; Knappe et al., 2000a). Whereas HVS infection or pathogenicity has not been reported in rodents, a nonpermissive infection and tumor induction was described in New Zealand white rabbits, although with variable efficiency (Ablashi et al., 1985; Medveczky et al., 1989). HVS and HVA can be isolated from peripheral blood cells of persistently infected monkeys or from leukemic animals, presumably from infected T cells, by co-cultivation with permissive owl monkey kidney (OMK) cells (Daniel et al., 1976; Falk et al., 1972). Both HVS and HVA replicate in OMK cells (Daniel et al., 1976), but while HVS induces cell lysis and grows to high titers, HVA remains mostly cell-associated with syncytia formation. As a result, supernatants of such cultures have lower and unstable HVA titers. A series of transformed T-cell lines were derived from leukemias or tumors of virus-infected tamarins and could be cultivated continuously for several years (reviewed in Fleckenstein and Desrosiers, 1982). While virus particles were found initially in most cases, virus production was lost after prolonged culture. The episomal DNA is heavily methylated in such cell lines (Desrosiers et al., 1979), and some of these cell lines carried rearrangements or large deletions in the episomal HVS genomes (Kaschka-Dierich et al., 1982). Marmoset and tamarin T-cells can be transformed by HVS to stable T-cell lines in vitro and are designated as semi-permissive, since virus particles are released in low titers (Desrosiers et al., 1986; Kiyotaki et al., 1986; Schirm et al., 1984; Szomolanyi et al., 1987). Likewise, HVA also transforms T cells of certain New World monkey species such as cottontop tamarins in culture, yielding cytotoxic T-cell lines (Falk et al., 1978; reviewed in Fleckenstein and Desrosiers, 1982).

94 Fig. 1 Genome structure of selected rhadinoviruses. The genome structures of the rhadinoviruses are shown with special respect to the variable areas harboring nonconserved viral genes or genes with homology to cellular counterparts (white boxes). KSHV/HHV-8 (Neipel et al., 1997; Russo et al., 1996), rhesus rhadinovirus (RRV; Alexander et al., 2000; Searles et al., 1999), herpesvirus saimiri (HVS; Albrecht et al., 1992a; Ensser et al., 2003), herpesvirus ateles (HVA; Albrecht, 2000). Conserved genomic regions encoding virus genes with typical herpesvirus functions are shown in black. Abbreviations: bZIP, basic-leucine zipper protein; CCPH, complement control protein homologue; DHFR, dihydrofolate reductase; FGARAT,

T-Cell Transformation and Oncogenesis

95

II. GENOME STRUCTURE HVS is the type species of the rhadinovirus subfamily (g2-herpesviruses) (Roizman et al., 1992). The term ‘‘rhadino’’ viruses uses the ancient Greek word ‘‘i
’’ for ‘‘fragile’’ (Roizman et al., 1992), because the genomic viral DNA splits upon isopyknic centrifugation into two classes of highly different density: the terminal repetitive H-DNA (high density, high G þ C content) without coding capacity and the L-DNA containing viral proteincoding genes (low density, low G þ C content). The intact rhadinoviral so-called M-genome has intermediate density in CsCl gradients (M-DNA). The size of the total M-DNA genome is variable due to different numbers of H-DNA segments attached to both ends of the linear virion genome. Two strains of HVS were sequenced, #A11 (Albrecht et al., 1992a) and the highly oncogenic subgroup C strain #C488 (Ensser et al., 2003). The long unique L-DNA of HVS A11 has 112930 bp with 34.5% G þ C, whereas the H-DNA is composed of multiple tandem repeats of 1444 bp with 70.8% G þ C. In strain C488, the L-DNA comprises 113027 bp, and it is flanked by two distinct repeat units of 1318 and 1458 bp. The shorter H-DNA unit is a subset of the longer repeat unit of which 140 bp are deleted. The size of the packaged C488 M-genome is approximately 155 kbp, ranging from 130 to 160 kbp due to variable numbers of terminal H-DNA segments (Ensser et al., 2003). The HVS L-DNA genomes contain at least 76 to 77 open reading frames and 5 to 7 U-RNAs (Albrecht et al., 1992a; Ensser et al., 2003; Ho¨ r et al., 2001). HVA strain #73 has a genome structure similar to HVS (Albrecht, 2000; Fleckenstein et al., 1978a) (Fig. 1). The long unique L-DNA containing all known virus genes has 108409 bp with 36.6% G þ C, and the terminal repetitive H-DNA without coding capacity contains multiple tandem repeats of 1582 bp with 77.1% G þ C. The HVA genome contains 73 ORFs and only two genes for U-RNA-like transcripts (Albrecht, 2000; Albrecht et al., 1999). The typical rhadinoviral genome is characterized by the gene blocks of typical herpesvirus genes, which are highly conserved between the herpesvirus families; these blocks are flanked or interspersed by genes unique to the rhadinovirus subfamily and strain (Albrecht and Fleckenstein, 1990; Gompels et al., 1988) (Fig. 1). Among these are transformation-related oncogenes and viral homologues of cellular genes, which will be described. While most genes are well conserved

formylglycineamide ribotide amidotransferase; FLIP, FLICE inhibitory protein; gp, glycoprotein; GPCR, G-protein coupled receptor; HSUR or HAUR, HVS or HVA-encoded URNA; IRF, interferon regulatory factor; MIP, macrophage-inflammatory protein; SAg, superantigen homolog; Stp, saimiri transformation-associated protein; Tio, two-in-one protein; Tip, tyrosine kinase–interacting protein; TS, thymidylate synthase.

96

Armin Ensser and Bernhard Fleckenstein

between different HVS strains, there is extensive sequence variation at the so-called left end of the HVS L-DNA and in the region of the R transactivator gene orf50 and the glycoprotein gene orf51 (Biesinger et al., 1990; Ensser et al., 2003; Ho¨ r et al., 2001; Thurau et al., 2000). Most HVS genes also have a conserved homolog in the HVA genome, including the viral SAG, Cyclin, GPCR homologs. HVA, however, does not encode homologs to the ORF12, vIL17, vCD59, or vFLIP (Fig. 1). The impression is rather convincing that HVA resembles a variant of HVS, which has either collected a smaller set of cell-homologous genes or has secondarily lost several genes. The mutation analysis of the HVS genome will be facilitated in the future by the availability of oncogene-deleted HVS A11S4 (White et al., 2003) and transforming HVS C488 bacmid clones (Grillho¨ sl et al., 2004); a defective subgenomic bacmid clone of the HVS strain #C484 L-DNA is also available (Collins et al., 2002).

III. REPLICATION AND PERSISTENCE OF T-LYMPHOTROPIC RHADINOVIRUSES While some amount of evidence has accumulated on the requirements of the related B-lymphotropic MHV-68 and KSHV for DNA replication and persistence, less is known about the replication mechanisms of T-lymphotropic rhadinoviruses in general or of HVS in particular. The HVS A11 origin of lytic replication was mapped to the untranslated region upstream of the thymidylate synthase gene (Lang and Fleckenstein, 1990; Schofield, 1994). A putative origin in the left-terminal region of the L-DNA was described to mediate plasmid maintenance in strain C484 (Kung and Medveczky, 1996). This region is neither conserved between different HVS strains, nor is it required for viral replication or episomal persistence in strain C488 or C484 (Ensser et al., 1999; Medveczky et al., 1989). In transformed human T cells, HVS persists as stable non-integrated episomes at high copy number (Biesinger et al., 1992). There are no indications yet for the genetic correlate of a plasmidlike origin of replication and of the viral factors involved.

A. The Immediate-Early and Early Genes The classification of HVS genes to the immediate-early (IE) phase of infection has been difficult and was mostly based on experiments using cycloheximide to inhibit viral protein synthesis. In contrast to herpes

T-Cell Transformation and Oncogenesis

97

simplex virus, this resulted in the rather asynchronous infection of tissue culture cells by HVS (Randall et al., 1985). The orf14 (ie14) transcripts encode a gene product with sequence homology to murine superantigens (Sag) (Thomson and Nicholas, 1991), but an essential regulatory function of the orf14/sag is unlikely, since it was shown to be dispensable for virus replication (Duboise et al., 1998a; Knappe et al., 1998b). The other recognized IE gene orf57 (ie57) codes for a nuclear phosphoprotein of 52 kDa (Nicholas et al., 1988; Randall et al., 1984) with structural and functional homology with ICP27/IE63 of herpes simplex virus and EBV BMLF1/EB2. Correspondingly, IE57 stimulates the expression of unspliced RNA and represses the expression of spliced transcripts (Whitehouse et al., 1998). IE57 is further involved in nuclear RNA export (Goodwin et al., 1999) and colocalizes to and redistributes nuclear components of the splicing machinery (Cooper et al., 1999). Thus, the ie57 post-transcriptional regulator appears to be the sole regulatory viral IE gene. A strong viral transactivator function was mapped to the delayed-early gene orf50 (Nicholas et al., 1991), the homolog to the R transactivator of EBV (BRLF1). The orf50 gene codes for a larger protein ORF50A and for a smaller C-terminal variant ORF50B due to differential promoter usage and splicing. The transactivation domain is located in the C-terminus of both ORF50 proteins and binds to the TATA-binding protein in the basal transcription complex (Hall et al., 1999). In strain A11, only ORF50A transactivated reporters carrying the promoters of responsive virus genes, such as orf6 and orf57 (Whitehouse 1997, 1998). While ie57 is highly conserved between subgroup A and C, the genomic orf50 region encoding this major viral transactivator was found to be strongly divergent (Ensser et al., 2003; Thurau et al., 2000), and in contrast to strain A11, the ORF50B protein of C488 had full transactivation properties (Thurau et al., 2000). Post-transcriptional inhibition of spliced orf50A transcripts by the IE57 can not have functional relevance in the context of strain C488, suggesting that orf50 exerts the dominant function for regulation of replication, at least in HVS strain C488. Neither HVS nor HVA encodes a homolog to the bZip/Zta of EBV or KSHV (Sinclair, 2003).

B. The Latency-Associated Nuclear Antigen The HVS ORF73 protein of strain A11 and C488 localizes to the host cell nucleus and, like the latent nuclear antigen LANA of KSHV, it can associate with host cell chromosomal DNA (Hall et al., 2000; Scha¨ fer et al., 2003). Similar to the KSHV LANA, the A11 ORF73 protein can form homodimers and associate with mitotic chromosomes via defined carboxyterminal domains (Calderwood et al., 2004). It has been claimed that, similar to

98

Armin Ensser and Bernhard Fleckenstein

KSHV LANA, the C488 ORF73 attaches the viral genomic DNA to metaphase chromosomes, and it seems to bind motifs derived from H-DNA terminal repeat sequences of HVS C488; however, the evidence was derived from transformed C. jacchus T-cells that are semipermissive for lytic replication, and the specificity of the ORF73 interaction with the homologous H-DNA terminal repeat motif has not been demonstrated (Verma and Robertson, 2003). The A11 ORF73 is highly expressed in HVS-infected human epithelial cancer cell lines; it can associate with the cellular p32, thereby coactivating heterologous as well as its own promoter (Hall et al., 2002). A11 ORF73 further binds to GSK-3 , which is involved in WNT-signaling and possibly contributes to oncogenic transformation (Fujimuro and Hayward, 2003). Although not detectable by Northern blotting from C488-transformed human T cells (Fickenscher et al., 1996), transcripts of this gene are found by RT-PCR. The HVS C488 orf73 gene product can downregulate the orf50A and B promoters and prevents the ORF50-mediated activation of viral replication gene promoters, and it can block the initiation of the lytic replication cascade. This suggests that HVS ORF73 can control the transition between rhadinoviral latency and lytic replication (Scha¨ fer et al., 2003) (Fig. 2). In analogy to ORF73, the LANA protein of KSHV has also been shown to downregulate the orf50/Rta promoter (Lan et al., 2004). The KSHV LANA transactivates its own promoter and is negatively regulated by ORF50 (Jeong et al., 2004). Whether the HVS orf73 promoter is also repressed by the KSHV ORF50/Rta is not known.

IV. VIRAL HOMOLOGS TO CELLULAR GENES Rhadinoviruses like HVS and KSHV contain numerous viral genes that are homologs to cellular genes and which seem to have been pirated from cellular DNA. Since the viral gene copies are typically intronless, a role for reverse transcription, possibly involving reverse transcriptase activity of endogenous retroviruses, may be theorized. This interesting hypothesis was discussed extensively; however, experimental results have not yet proven it. There are a few of these possibly captured cellular genes that are unique to specific viruses, while some others are common to several rhadinoviruses (Fig. 1) or to the g-herpesviruses including EBV. This may suggest that the uptake of cellular genes is a rare event during herpesvirus evolution. Most of these cellular homologs can be categorized into two major groups: (A) genes related to nucleotide metabolism or cellular growth control and (B–E) genes that modulate innate or adaptive immune functions (discussed in the following text).

Fig. 2 The LANA protein is an upstream regulator of the lytic gene expression cascade. (A) In the absence of HVS ORF73, both forms of ORF50 protein are expressed and activate expression of delayed-early genes involved in DNA-replication, leading to expression of late viral genes. (B) ORF73 expression downregulates the transcriptional activity of ORF50 promoters and prevents ORF50 expression and the initiation of the lytic viral replication cycle via delayed-early gene expression. (C) Higher doses of ORF73 begin to repress the ORF73 promoter, ORF73-mediated repression of the ORF50 promoter is released, and lytic replication can proceed.

100

Armin Ensser and Bernhard Fleckenstein

A. Homologs to Enzymes of Nucleotide Metabolism and Growth Regulation Homologs to enzymes of the nucleotide metabolism include a dihydrofolate reductase (orf2, DHFR), which is not present in HVA, and a functional thymidylate synthase (orf70, TS). In addition, orf3 and orf75 encode large tegument proteins, which share local homology to formylglycineamide ribotide amidotransferase (FGARAT) (summarized in Ensser et al., 2003). The functions of these enzymes may possibly augment the free nucleotide pool and could thus facilitate DNA synthesis and virus replication. Orf2 and the adjacently located viral U-RNA genes are dispensable for virus replication and T-cell transformation (Ensser et al., 1999). HVS orf72 codes for a functional viral cyclin D that is expressed in semipermissive HVS-transformed marmoset T cells and is resistant to cyclin-dependent kinase inhibitors p16Ink4a, p21Cip1, and p27Kip1. This deregulation pushes the cell cycle toward the S phase, thereby supporting virus replication in permissive cells (Jung et al., 1994; Nicholas et al., 1992; Swanton et al., 1997). Although enhanced replication might secondarily promote transformation or tumor induction, the viral cyclin seems to provide only auxiliary functions, since it is not required for replication, T-cell transformation, and lymphomagenesis (Ensser et al., 2001). We have found that in HVS C488, the viral cyclin D can enhance viral reactivation from transformed primate T-cells, and it enhances S phase entry of transformed T-cells under growth limiting conditions. This finding is similar to observations in MHV-68, where the viral cyclin is also dispensable for pathogenesis but enhances reactivation of virus from latently infected cells (van Dyk et al., 1999, 2000).

B. Complement Regulatory Proteins HVS has two genes that are functional regulators of the complement system. A complement control protein homologue (CCPH) is encoded by the orf4, which inhibits C3 convertase, an enzyme involved in the initiation of early steps in complement activation (Albrecht and Fleckenstein, 1992; Fodor et al., 1995). The orf15 is a viral variant of CD59, which prevents the insertion of the membrane attack complex formed by C8 and C9, and thus blocks the terminal complement cascade (Albrecht et al., 1992b; Rother et al., 1994). T-cell stimulatory functions have been described for cellular CD59 via CD2 (Deckert et al., 1995; Korty et al., 1991), but analogous functional data do not exist for the viral CD59.

T-Cell Transformation and Oncogenesis

101

C. Antiapoptotic Viral Proteins The HVS orf16 and orf71 encode potent inhibitors of apoptotic cell death. The ORF16 protein is a viral Bcl-2 homologue that can stabilize mitochondrial membranes and is able to inhibit apoptosis induced by Sendai virus infection or by treatment with Fas-ligand, dexamethasone, menadione, or by irradiation (Bellows et al., 2000; Derfuss et al., 1998; Nava et al., 1997). Both Bcl-XL and viral Bcl-2 inhibit cell death induced by either cell-autonomous (independent of death receptors) or receptor-mediated mechanisms, depending on the cell type studied. The ORF71, which is not preserved in HVA, is a viral FLICE (FADD-like interleukin 1-converting enzymelike protease) inhibitory protein (vFLIP). vFLIP interacts with cellular FADD (Fas-associated protein with death domains) and FLICE via homophilic interaction of their respective deatheffector domains; this can block the formation of the death-signal-induced signaling complex (DISC), and it consequently prevents caspase 8 (FLICE) activation. The vFLIP inhibited death-receptor-dependent apoptosis and partially protected permissive OMK cells from Fas-dependent apoptosis at a late stage of infection (Thome et al., 1997). However, the vFLIP was dispensable for virus replication to high virus titers, T-cell transformation, and lymphoma induction (Glykofrydes et al., 2000).

D. Viral Cytokines and Receptor Proteins The cellular homolog of HVS orf13, initially termed ctla8 (Rouvier et al., 1993), codes for IL-17, a cytokine produced specifically by CD4þ T cells. Indeed, the viral orf13/vIL-17 has led to the discovery of its cellular homolog. IL-17 induces the secretion of IL-6, IL-8, G-CSF, and PG-E2 by fibroblasts, endothelial, or epithelial cells, and it can promote the proliferation and maturation of CD34þ hematopoetic progenitor cells into neutrophils. Among various other functions, IL-17 was shown to support T-cell proliferation (Fossiez et al., 1998; Yao et al., 1995). The viral IL-17, which is unique to HVS, is functionally not distinguishable from its cellular counterpart. However, deletions of the HVS C488 orf13/vIL-17 did not affect virus replication and oncogenicity (Knappe et al., 1998a). G-protein coupled receptors (GPCR) are found in most rhadinoviruses, in many other herpesviruses, and in poxviruses. The HVS orf74 (ECRF3, vGPCR) encodes a viral IL-8 receptor (IL-8R) that was classified to the low-affinity B type of IL-8R (Ahuja and Murphy, 1993; Murphy, 1994; Nicholas et al., 1992). Viral IL-17, produced by virus-infected cells, could induce IL-8 on neighboring stroma cells. This IL-8 might then bind to the

102

Armin Ensser and Bernhard Fleckenstein

viral IL-8R on the virus-infected cell and lead to further activation. However, the existence of such a paracrine stimulation mechanism is speculative. In addition to IL-8, and with much higher capacity, it has been shown that ORF74 can bind the other CXC-Chemokines GCP-2 and GROalpha but, in contrast to the KSHV ORF74, not IP10. Downstream signaling (via Gi, G12/13, but not Gq) resulted in transcription factor activation that was constitutive via SRE-elements, while CREB, NFAT, and NFkB were dependent on ligand-binding (Rosenkilde et al., 2004).

E. The Viral Superantigen Homolog and a New Signaling Adaptor Protein Superantigens crosslink the superantigen-specific V chains of the T-cell receptor with major histocompatibility complex (MHC) class II molecules on accessory cells; this leads to efficient antigen-independent stimulation of a significant part of the T-cell repertoire. The viral immediate early gene orf14/ vSag revealed local homology to the superantigen (Sag) of mouse mammary tumor virus (MMTV) and to murine mls superantigens (Thomson and Nicholas, 1991). Although a recombinant viral IE14/vSag protein bound to MHC class II molecules and stimulated T-cell proliferation, no selective advantage was found for specific V families that would be typical for superantigens, neither after in vitro stimulation of human T cells with IE14/vSag (Duboise et al., 1998a; Yao et al., 1996), nor after infection and transformation with HVS (Knappe et al., 1997). Therefore, the IE14 may be more appropriately designated as a superantigen homolog or a mitogen. Expression of the IE14/vSag protein has not been demonstrated in HVS-transformed human T-cells. The roles of the HVS C488 ie14/vsag in transformation of human and simian T-cells in vitro and in the pathogenicity of HVS C488 has remained controversial. In one line of studies, deletion of ie14/vsag from HVS C488 neither impaired its ability to induce tumors in cottontop tamarins (S. oedipus), nor to transform human and simian T cells in vitro (Knappe et al., 1997, 1998b). In another study, ie14/vsag deletion mutants were apathogenic in common marmosets (C. jacchus) and were unable to transform C. jacchus T cells in vitro. Surprisingly, this virus did not even persist in infected animals (Duboise et al., 1998b). This disagreement could be explained either by species-specific differences between New World primate species, by differences between the recombinant viruses, or in the specific experimental procedures, such as the comparably lower virus titers used for the experimental infection in the latter study. The neighboring orf12 has local homology to the K3 and K5 genes of KSHV that suppress antigen presentation by MHC (Coscoy and Ganem,

T-Cell Transformation and Oncogenesis

103

2000; Ishido et al., 2000a,b; Means et al., 2002), but it is dispensable for replication and human T-cell transformation in vitro (Knappe et al., 1998a). Orf12 is not preserved in HVA. The orf5 is conserved in different HVS strains and also in HVA. It has been recognized to encode a viral adapter protein with structural similarity to LAT. The small 89 to 91 aa protein has an aminoterminal myristoylation site and several SH2 binding sites. In transfected 293T epithelial kidney cells and Jurkat cells that overexpress the ORF5 protein, it interact with SH2-domain containing signaling proteins and augmented TCRsignaling activity; thereby it might facilitate persistent infection. Expression in HVS-transformed or infected human or monkey cells is uncertain or it seems to be restricted to the lytic replication phase (Lee et al., 2004). The absence of ORF5 expression in latently infected cells would make a contribution to the transformation of human T-cells less likely. HVS, like other rhadinoviruses, has acquired a number of cellular genes, but most, if not all, of these genes seem dispensable for virus replication, Tcell transformation in culture, and pathogenesis in susceptible New World primates. Similar to KSHV infected B-cell lines, expression of the cellhomologous viral genes is mostly found during lytic virus replication, but not in transformed human lymphocytes (Fickenscher et al., 1996; Knappe et al., 1997).

V. ONCOGENESIS A. Transformation-Related Proteins of Rhadinoviruses The variable region at the left end of the HVS L-DNA harbors the major factors responsible for induction of T-cell leukemia and T-cell transformation in vitro (Chou et al., 1995; Desrosiers et al., 1985a, 1986; Duboise et al., 1998b; Koomey et al., 1984; Murthy et al., 1989). In subgroup A and B strains, there is only one gene at this position, termed stpA or stpB (saimiri transformation-associated protein of subgroup A or B strains) (Ho¨ r et al., 2001; Murthy et al., 1989). The virus strains of subgroup C carry two open reading frames at the homologous location (Biesinger et al., 1990). These were later named stpC (saimiri transformation-associated protein of subgroup C strains) (Jung and Desrosiers, 1991) and tip (tyrosine kinase interacting protein) (Biesinger et al., 1995) (Figs. 3, 4). In the closely related HVA, two small open reading frames were identified at the left end of the HVA genome, which had local homology to tip as well as stpC. These small orfs were found to represent the two exons of the tio (two-in-one) gene by RT-PCR from HVA-transformed monkey cells (Albrecht et al., 1999).

104

Armin Ensser and Bernhard Fleckenstein

Fig. 3 Comparison of Herpesvirus saimiri and Herpesvirus ateles L-DNA left terminal regions. The oncoproteins Stp, Tip, or Tio are encoded at the variable region. Stp, saimiri transformation-associated protein of the respective subgroup A, B, or C; Tio, two-in-one protein of HVA; Tip, tyrosine kinase–interacting protein of HVS subgroup C; HSUR or HAUR, HVS or HVA-encoded URNA; FGARAT, formylglycineamide ribotide amidotransferase.

There are only two genes for HVA U-RNA like transcripts (HAUR) in the immediate context of Tio, and there is no dhfr gene (Albrecht, 2000; Albrecht et al., 1999) (Fig. 3). StpA and B share limited sequence homology with StpC, but are structurally unrelated to Tip (Fig. 4). While not required for viral replication, deletion of either stpC or tip abolishes the transformation by HVS in vitro and pathogenicity in vivo (Duboise et al., 1996, 1998b; Knappe et al., 1997;

T-Cell Transformation and Oncogenesis

105

Medveczky et al., 1993). Both stpC and tip are transcribed into a single bicistronic mRNA from a common promoter directed toward the left genomic end of the L-DNA, where tip is situated downstream of stpC. The transcription of stpC/tip is regulated similarly to cellular IE genes in human T cells, and no obvious viral factors seem to be involved. In C488-transformed human T cells, stpC and tip are the only constitutively transcribed viral genes found so far. While the StpC protein can be readily detected by immunoblotting, Tip requires more sensitive immune complex kinase assays when expressed at physiological levels in HVS transformed T-cells (Biesinger et al., 1995; Fickenscher et al., 1996, 1997).

B. The Transformation-Associated Protein STP The stpA of strain A11 encodes a 164 aa phosphoprotein with a predicted molecular mass of 17 kDa, which migrates at an apparent molecular mass of 26 and 35 kDa (Lee et al., 1997). The StpC protein of strain C488 is a 102 aa perinuclear membrane-associated phosphoprotein with a predicted molecular mass of approximately 10 kDa, but it migrates with an apparent molecular mass of 20 kDa. The N-terminus of StpC consists of 17 mostly charged amino acids. The C-terminus is a hydrophobic region, which probably serves as an anchor to perinuclear membranes. In between are 18 collagen tripeptide repeats (GPX)n, which may mediate multimerization of the protein (Fig. 4) (Biesinger et al., 1990; Fickenscher et al., 1997; Jung and Desrosiers, 1991, 1992, 1994). StpA of strain A11 and stpC of strain C488 transfected rodent fibroblasts formed foci in vitro and induced tumors in nude mice (Jung et al., 1991). StpA transgenic mice developed polyclonal peripheral T-cell lymphoma, while an stpC transgene induced epithelial tumors (Kretschmer et al., 1996; Murphy et al., 1994). StpA was reported to bind to and to be phosphorylated by the nonreceptor tyrosine kinase Src (Lee et al., 1997). StpA can form a complex with Src and STAT3 in transfected 293 and 3T3 cells, leading to STAT3 phosphorylation, nuclear translocation, and activation STAT3-induced gene expression (Chung et al., 2004). The nontransforming StpB also associates with Src (Choi et al., 2000). StpC was shown to interact with the small G-protein Ras and stimulated mitogen-activated protein (MAP) kinase activity (Jung and Desrosiers, 1995). Both proteins StpA and StpC interact with TNFassociated factors (TRAFs), leading to nuclear factor kappa B (NFB) activation (Lee et al., 1999). StpC specifically activates NFkB via the TRAF/NIK/IKK pathway independently of TNF (Sorokina et al., 2004). Although the precise biochemical mechanisms still have to be resolved in detail, the transforming potential of StpA and StpC is well established.

106

T-Cell Transformation and Oncogenesis

107

C. The Tyrosine Kinase–Interacting Protein (Tip) of Herpesvirus saimiri Subgroup C A 40 kDa phosphoprotein that coprecipitated with the T-cell-specific nonreceptor tyrosine kinase p56lck was observed in HVS strain C488-transformed T-cells by a phosphotransferase assay. The protein was thus named tyrosine kinase–interacting protein (Tip) and could be identified as the gene product of the leftmost C488 reading frame (Biesinger et al., 1995). Tip of HVS strain C488 has 256 aa and a predicted molecular mass of 29 kDa. Considerable interstrain variation of the tip genes has been observed (Ensser et al., 2003; Greve et al., 2001), but the common denominators are an N-terminal glutamate-rich region, which is duplicated in some strains, followed by one or two serine-rich regions, a bipartite kinase interacting domain, and a C-terminal hydrophobic domain which anchors the molecule at the inside of the plasma membrane. The kinase-interacting domain consists of nine amino acids with homology to the C-terminal regulatory regions of various Src kinases (CSKH), and a proline-rich SH3domain-binding sequence (SH3B); both motifs are required for the interaction with the kinase (Biesinger et al., 1995; Jung et al., 1995a) (Fig. 4). Several tyrosine residues, three of which are conserved between all strains investigated, can be a substrate for the tyrosine kinase Lck. Tip binding to Lck modulates the kinase activity: Data from transient assays using native as well as recombinant molecules, in transfected cells or cell-free systems, showed that binding of Tip to Lck results in a strong activation of the enzyme (Fickenscher et al., 1997; Hartley et al., 1999; Lund et al., 1997a; Wiese et al., 1996). Recombinant Lck was activated by recombinant Tip in vitro, showing that their interaction is direct and does not depend on other T-cell molecules. Although the SH3-binding domain of Tip can interact with several Src-family kinases (Schweimer et al., 2002), the activation after interaction of Lck and Tip in T-cells seems rather selective; although p60fyn and p53/56lyn are active in transformed T-cells, they were neither activated by Tip nor did they phosphorylate Tip (Fickenscher et al., 1997; Wiese et al., 1996). When compared with their non-infected parental clones, HVS C488-transformed human T-cell clones had increased basal Fig. 4 Structural diversity of the major gamma herpesvirus signaling proteins. The HVS oncoproteins Stp, Tip, or HVA Tio, the KSHV K1 and K15, RRV R1 proteins are shown along with the EBV LMP1, LMP2a. CSKH, C-terminal Src kinase homology region; ITAM, immunoreceptor tyrosine-containing activation motif (green/blue); SH3b, domain interacting with Src-family kinase SH3 domain (yellow); SH2b, domain interacting with Src-family kinase SH2 domain; S-rich, serine rich motif; TRAF, potential binding site for TNF-R associated factors; Y and YY, tyrosine and double-tyrosine residues, putative signaling motifs; Black, Transmembrane domain; N or C, amino- or carboxyterminus of the protein, respectively.

108

Armin Ensser and Bernhard Fleckenstein

levels of tyrosine phosphorylation (Wiese et al., 1996). Tip could activate Lck even when the regulatory tyrosines Y394 and Y505 of Lck had been mutated, suggesting a novel mechanism of Lck activation (Hartley et al., 1999). Since constitutively active Lck mutants have an oncogenic potential, stimulation of Lck signaling by Tip would contribute to the activated phenotype of HVS-transformed T cells and to the transformation process, whereby Tip and StpC would act in synergy. In addition to Lck, phosphorylated signal transducer and activator of transcription (STAT) factors 1 and 3 were immunoprecipitated with Lck and Tip-C484 (Lund et al., 1997b). The 48 amino-acid Lck-binding domain of Tip, including the CSKH and SH3B domains, was sufficient for activation of Lck and STAT3 in Jurkat cells (Lund et al., 1999). Tyrosine phosphorylation at a YXPQ motif of Tip (Y72; identical to Y114 of C488), which conforms to a consensus binding site for STAT factors, was required for STAT binding and transcription activation (Hartley and Cooper, 2000). However, when we introduced a corresponding Tip Tyrosine 114 to Phenylalanine (Y114F) mutation to generate recombinant HVS C488-TipY114F, this mutation abolished the constitutive STAT3 phosphorylation but was still efficiently able to transform human T-cells (Heck et al., 2005). On the other hand, when Tip was highly expressed in stable tip-transfected Jurkat T-cells, low basal levels of tyrosine phosphorylation, impaired response to T-cell receptor activation, and downregulation of CD3 and CD4 were observed. Tip further partially reversed the transformed phenotype of fibroblasts, which had lost contact inhibition after transfection with a constitutively active mutant of Lck. These effects were even more pronounced when Tip Tyrosine 114 was mutated in position 114 to Serine (Y114S) to enhance its binding to Lck (Guo et al., 1997; Jung et al., 1995b). This led to a model similar to the role of latent membrane protein (LMP)-2A in EBV infection; LMP-2A inhibits B-cell receptor-mediated signaling in EBV-transformed B cells, thereby favoring the latent form of infection by blocking EBV replication (Fruehling and Longnecker, 1997; Miller et al., 1994). Accordingly, Tip would be a functional antagonist of StpC. The activation of Lck and the inhibition of T-cell signaling by Tip may be two different aspects of the same function, since the activation of Lck by Tip might trigger negative feedback mechanisms in stably transfected Jurkat cells expressing high levels of Tip. In a similar line of evidence, lymphocytes transduced with Tip-expressing retroviral vectors showed increased levels of Lck-dependent Fas-mediated apoptosis and decreased IL-2 expression (Hasham and Tsygankov, 2004). T-cell signaling is a complex regulatory process; therefore, the observation of conflicting effects of Tip under different experimental conditions is not unexpected. Another possible reason for this is the low expression level of Tip in such HVS-transformed primary human and simian T-cells, while p56lck is abundantly expressed;

T-Cell Transformation and Oncogenesis

109

any possible changes in the Lck enzymatic activity after binding of Tip may be masked by the excess free Lck. Tip function has evolved in the context of the natural infection of squirrel monkeys, where uncontrolled T-cell transformation seems not to occur (Greve et al., 2001). A Tip-binding lysosomal 80 kDa protein with N-terminal WD repeat and C-terminal coiled-coil domain was identified by affinity chromatography and mass spectrometry. After coexpression of high levels of Tip and p80 in 293 and Jurkat cells, interaction of the N-terminal serine-rich region of Tip with p80 induces enlarged endosomal vesicles and recruits Lck and TCR complex into these vesicles for trafficking or degradation. Since Tip is constitutively present in lipid rafts, it was found that Tip can recruit p80 into lipid rafts; the C-terminus of Tip seems to interact with Lck to recruit TCR complex to lipid rafts, and TCR and Lck then may interact with p80 to initiate the aggregation and internalization of the lipid raft domain and thereby downregulate the TCR complex (Park et al., 2002, 2003). Previously, the same group had identified a Tip-associated cellular protein of 65 kDa termed Tap that could contribute to T-cell activation by Tip (Yoon et al., 1997). Meanwhile, Tap appears to be an RNA export factor that has no known T-cell-specific functions as yet (Gru¨ ter et al., 1998).

D. The Tio Oncogene of Herpesvirus ateles A spliced gene with two exons was detected in the left-terminal L- to H-DNA transition region of HVA strain #73. The derived viral protein shares local sequence homology with StpC and Tip of HVS and was therefore termed Tio (‘‘two in one’’; Figs. 3, 4). Tio is expressed in HVA-transformed simian T cells. After cotransfection, Tio bound to and was phosphorylated by the Src kinases Lck or Src (Albrecht et al., 1999). Human T cells can be successfully transformed by recombinant HVS C488 in which the stpC/tip genes were replaced by either the HVA #73 region containing the promoter and both exons of tio, or a cDNA of tio transcribed from a heterologous promoter; thus, Tio can substitute for both StpC and Tip in human T-cell transformation (Albrecht et al., 2004). Interestingly, the phenotype of T-cells transformed by recombinant Tio-encoding HVS-C488 showed a preference for transformation in the absence of IL-2, while transformation was rather inefficient in its presence; this may hint at qualitative or quantitative differences in usage of T-cell signaling pathways of the Tio protein compared to the combined StpC plus Tip. Since human T cells have not been susceptible to transformation with various HVA strains, it is so far unknown whether other factors in the HVS background support transformation or if this failure to transform human T cells simply reflects the cell-associated growth of HVA strains. Factors that might contribute are the lack of an unknown viral

110

Armin Ensser and Bernhard Fleckenstein

receptor binding protein preventing efficient infection of human T-cells, or differences in the replication and maintenance of the viral episome.

VI. GROWTH TRANSFORMATION OF HUMAN T CELLS BY RHADINOVIRUSES A. Establishing Transformed T-Cell Lines Primary human T-cell culture is laborious and requires repeated stimulation with a mitogen or specific antigen in the presence of accessory cells expressing the appropriate MHC restriction elements. Since they are limited in their natural life span, it is rarely possible to grow primary T lymphocytes to large numbers. The efficient immortalization in vitro of human B lymphocytes by EBV is an established laboratory method (Henle et al., 1967). The method has been helpful for the analysis of the human B-cell repertoire and function, and since the EBV transformed lymphoblastoid B cells retain their antigen-presenting capability, they are widely used to study T-cellular antigen specificity. EBV can also immortalize human T-cells, but with low efficiency (Groux et al., 1997). Rapidly proliferating T-lymphoblastic tumor cell lines such as Jurkat (Schneider et al., 1977) are frequently used as a cellular and biochemical model for primary human T cells. However, such tumor-derived cell lines usually display a strongly altered phenotype with respect to signal transduction, gene regulation, and proliferation control. Another problem is the identity of the studied materials, since contaminations of cultures with other rapidly growing lymphoid cell lines are frequent (Drexler et al., 2003). HTLV-1 can transform human T cells, but these cell lines lose the T-cell-receptor complex, the cytotoxic activity, and the dependence on IL-2 after prolonged cultivation (Inatsuki et al., 1989; Miyoshi et al., 1981; Yamamoto et al., 1982; Yssel et al., 1989). The observation that HVS strain C488 stimulated human T lymphocytes to stable antigen-independent growth in culture offered a practical method of T-cell transformation (Biesinger et al., 1992). These growth-transformed human T cells retained many essential T-cell functions including the MHCrestricted antigen-specific reactivity of their parental T-cell clones (Bro¨ ker et al., 1993; De Carli et al., 1993; Weber et al., 1993). These observations have opened up a novel research direction which links T-cell biology, signal transduction pathways, and transforming viral functions. The virus can not be isolated from transformed human T-cell cultures (Biesinger et al., 1992); this is in contrast to HVS-transformed Old and New World monkey T lymphocytes that produce infectious viral particles in many cases. Although the formal proof that the virus can never be

T-Cell Transformation and Oncogenesis

111

reactivated from transformed human T lymphocytes is difficult, neither treatment with phorbol esters, nucleoside analogues, nor other drugs that can cause reactivation of other viruses such as EBV or KSHV, nor specific or nonspecific stimulation of the T cells could induce virion production (Fickenscher et al., 1996). StpC and Tip are the only viral proteins which have been regularly demonstrated in HVS-transformed human T cells. The noncoding viral U-RNA genes (HSUR, HVS URNA) were abundantly expressed, similarly to the EBER RNAs of EBV. However, the deletion of all HSUR did not influence virus replication or T-cell transformation (Ensser et al., 1999). While the viral gene stpC/tip was strongly and inducibly transcribed into a bicistronic message (Fickenscher et al., 1996; Medveczky et al., 1993), other viral transcription was rarely detected. The gene ie14 was abundantly transcribed for a few hours only after stimulation of transformed human T cells with phorbol ester, and the IE gene ie57, the early gene orf50, and the viral thymidylate synthase gene were found transcribed at extremely low abundance or only after additional T-cell stimulation (Knappe et al., 1997; Thurau et al., 2000). These findings argue for a strong block of virus replication in C488-transformed human T cells that is downstream of the expression of the regulatory genes orf50 and ie57. The specific properties of HVS-transformed T cells and the transformation procedure have been comprehensively reviewed (Ensser and Fleckenstein, 2004; Ensser et al., 2002; Fickenscher and Fleckenstein, 2001, 2002). Briefly, the infection of peripheral blood mononuclear cells, cord blood mononuclear cells, thymocytes, or established T-cell clones by HVS C488 results in T-cell lines that continuously grow without restimulation with antigen or mitogen and do not require the presence of feeder or antigen-presenting cells.

B. The Phenotype of Rhadinovirus Transformants The HVS strain C488 is commonly used for the targeted transformation of human T cells; other subgroup C virus strains were able to transform human T cells, but to a varying extent (Fickenscher et al., 1997). The morphology of such lines resembles the irregular shape of T blasts; they carry nonintegrated viral episomes in high copy number and have a normal karyotype (Troidl et al., 1994). The surface phenotype of the transformed T lymphocytes resembles mature, activated T cells that are CD4 þ CD8 or CD4CD8þ T cells and carry  - or g-type T-cell receptors; transformation of established T-cell clones demonstrated that the phenotype and HLA-restriction of the parental T cells is conserved. If transformed by the same virus strain,  and g-clones were similar with respect to viral persistence, virus gene expression, proliferation, and Th1-type cytokine production. The phenotype of HVS-transformed T cells is remarkably

112

Armin Ensser and Bernhard Fleckenstein

stable for many months in culture, a significant technical advance over immortalization of human T cells by the hybridoma technique or infection with HTLV-1 (Biesinger et al., 1992; Knappe et al., 2000a; Meinl et al., 1995; reviewed in Fickenscher and Fleckenstein, 2002). HVS-transformed cell lines were strongly stimulated by CD2-CD58 interaction, which led to IL-2 production and enhanced proliferation. Binding of CD2 to its ligand CD58 can be mimicked by antibodies directed against the CD2.1 epitope alone. For activation of primary T cells, the simultaneous ligation of CD2.1 and a different epitope, CD2.2 or CD2R, is necessary (Meuer et al., 1984; Mittru¨ cker et al., 1992). Stimulation of CD3, CD4, and the IL-2 receptors was measured by signal transduction parameters, by proliferation, and by Interferon-g ([IFN]-g) production. The basal proliferation activity, which is probably due to CD2 contacts with CD58-bearing cells, may interfere with the demonstration of antigen specificity; this can be reduced by using monoclonal antibodies against CD58 and CD2 together with HLA-transfected mouse L cells as antigen-presenting cells, or by starving the cells prior to antigen presentation. Responses to antigen contact can be measured by proliferation and different cytokine production; IFN-g production is the most reliable parameter (Bro¨ ker et al., 1993; Weber et al., 1993). One EBV specific cytotoxic T-lymphocyte line that retained its antigen-specific reactivity after HVS transformation was reported (Berend et al., 1993). The cytokines IL-2 and IL-3 are secreted by the cells after activation by mitogenic or antigenic stimuli, and antibodies against CD25 (IL-2R), IL-2, or IL-3 suppressed the growth rate; both cytokines seem to support autocrine growth (De Carli et al., 1993; Mittru¨ cker et al., 1992). Although many normal T-cell functions are preserved, a few specific cellular and biochemical features are typically changed in comparison with their parental cells. First, this relates to the hyperresponsiveness to CD2 ligation (Mittru¨ cker et al., 1992), the degree of which has been found to depend on the transforming strain of HVS subgroup C (Fickenscher et al., 1997). The transformed T cells express both the CD2 and the ligand CD58 at high density on their surface. This could lead to autostimulation and cell contact seems necessary for their growth, since deprivation by limiting dilution immediately halts the growth of HVS-infected human T cells. The CD2 hyperresponsiveness likely contributes to the spontaneous proliferation and to the transformed phenotype of HVS-transformed human T cells. However, transformed T cells from a patient lacking CD18/LFA-1 expression could not be stimulated via the CD2 pathway, suggesting that the CD2 hyperreactivity might not be essential for the HVS transformation (Allende et al., 2000). Second, the protein tyrosine kinase p53/56lyn is expressed and enzymatically active in HVS transformed T cells (Fickenscher et al., 1997; Wiese et al., 1996). Lyn is usually expressed in B cells but not in T cells, and is also found in T cells immortalized by HTLV-1 (Yamanashi

T-Cell Transformation and Oncogenesis

113

et al., 1989). Third, HVS-transformation shifts the range of cytokines secreted by stimulated T cells toward a Th1 profile: IL-4 and IL-5 production is diminished and secretion of IL-2 and IFN-g is increased in comparison with parental cells. The IFN-g secretion can be stimulated to particularly high levels (Bro¨ ker et al., 1993; De Carli et al., 1993; Weber et al., 1993). In addition, many of these clones secrete large amounts of the macrophage inflammatory proteins (MIP)-1a and MIP-1b. Using representational difference analysis, a specifically overexpressed novel cellular gene ak155 was identified in HVS-transformed T cells. The AK155 protein, also termed IL-26, is a member of the IL-10 cytokine family. IL-26 is transcribed at low levels in various T-cell types and in native peripheral blood cells, but not in B cells. The IL-26 protein forms homodimers, similarly to IL-10 (Knappe et al., 2000b). IL-26 can induce STAT activation in epithelial cells via a receptor that is a heterodimer of IL20R1 and IL10R2 chains; however, the functional IL-26 receptor is not found on T cells (Ho¨ r et al., 2004; H. Fickenscher, personal communication). Thus, it is unlikely that IL-26 contributes to HVS-mediated T-cell transformation, but it may be involved in the lymphocyte-epithelium interaction which is typical for various gamma herpesviruses.

C. Application of T-Cell Growth Transformation T-cell transformation by HVS C488 was successfully used to study T cells from primary human immunodeficiencies and, in many cases, has been the only way to cultivate and amplify the patients’ cells for further research. HVS-transformed T-cell lines have been established from patients with genetic T-cell defects involving the CD3g chain (Rodriguez-Gallego C. et al., 1996; Zapata et al., 1999), IL-2Rg chain (Stephan et al., 1996), CD95/Fas (Bro¨ ker et al., 1997), IL-12R (Altare et al., 1998), MHC class II (Alvarez-Zapata et al., 1998), Wiskott-Aldrich syndrome (Gallego et al., 1997), and CD18/LFA-1 (Allende et al., 2000). HVS-transformed T lymphocytes from macaque monkeys resemble their transformed human counterparts in many respects, although CD4þ CD8þ double-positive T cells, which are uncommon in humans, were frequently observed. However, the frequent reactivation of foamy virus, with which most rhesus monkeys in primate centers are infected, has frequently limited this experimental approach (Akari et al., 1996; summarized in Fickenscher and Fleckenstein, 2001). Antigen-specific simian T-cell lines retained their reactivity after transformation, and MHC class II-expressing transformed cells were able to present the antigen to each other in the absence of autologous presenter cells (Meinl et al., 1997). In contrast to their human counterparts, however, when tissue culture supernatants were centrifuged to

114

Armin Ensser and Bernhard Fleckenstein

concentrate virus particles, macaque T-cell lines were shown to shed low amounts of virus particles in many cases (Alexander et al., 1997; Knappe et al., 2000a). After reinfusion of autologous transformed T cells into the donor macaques, the infused T cells persisted for extended periods. The reinfused animals did not develop disease, and they were protected against tumor induction by challenge with the HVS C488 virus (Knappe et al., 2000a). Apart from being a tumor virus of New World monkeys, the behavior of HVS C488 in various Old World monkey systems is relevant, since macaques are a common animal model for the situation in humans. HVS-transformed human CD4þ T cells provide a productive system for T-lymphotropic viruses such as HHV-6 (F. Neipel and B. Fleckenstein, unpublished data) and human immunodeficiency virus (HIV) (Nick et al., 1993). The prototype viruses HIV-1IIIB and HIV-2ROD, a poorly replicating HIV-2 strain, as well as primary clinical isolates replicate in HVS-transformed T cells. Macrophage-tropic HIV isolates replicated without selecting for subtypes with changed phenotype or cell tropism (summarized in Fickenscher and Fleckenstein, 2001).

VII. RHADINOVIRUS VECTORS FOR GENE THERAPY Foreign genes have been inserted into the genome of HVS by various techniques including homologous recombination and subsequent limiting dilution (Desrosiers et al., 1985b), homologous recombination and subsequent enrichment via a cointroduced selective marker gene (Alt et al., 1991; Grassmann and Fleckenstein, 1989), direct ligation into HVS-DNA via engineered unique restriction enzyme cleavage sites (Duboise et al., 1996), homologous recombination with simultaneous replacement of a SEAP marker gene and subsequent limiting dilution (Duboise et al., 1998b), reconstitution from overlapping cosmids (Ensser et al., 1999), or by engineering of bacterial artificial chromosomes (Grillho¨ sl et al., 2004; White et al., 2003). Both oncogenic and nontransforming HVS variants have been used as eukaryotic expression vectors. HVS-deletion mutants without the leftterminal transformation-associated L-DNA region neither cause malignant disease in animals nor do they transform simian lymphocytes in culture (Desrosiers et al., 1984, 1986; Duboise et al., 1998b). In the past, such HVS vectors had been used to define the transforming functions of the HTLV-1 X region, revealing Tax as a transforming principle of HTLV-1 for human T cells (Grassmann et al., 1989, 1992). HVS-mediated expression of the proto-oncogene c-fos could not transform primary fibroblasts (Alt and Grassmann, 1993). In HVS-deletion mutants which still contained the tip gene, the stpC oncogene was successfully substituted by cellular ras (Guo

T-Cell Transformation and Oncogenesis

115

et al., 1998), by the K1 gene of KSHV/HHV-8 (Lee et al., 1998), or by R1 of rhesus monkey rhadinovirus (RRV) (Fig. 1) (Damania et al., 1999). Infection and sometimes limited replication of HVS was observed in various human cell types (Daniel et al., 1975; Simmer et al., 1991; Stevenson et al., 1999, 2000b). HVS-vectors can efficiently transduce human mesenchymal cells including bone marrow stroma cells (Frolova-Jones et al., 2000), while infection of human hematopoietic progenitor cells was less efficient and with a tendency toward partially differentiated cells (Stevenson et al., 1999). Although totipotent mouse embryonic stem (ES) cells were infected under drug selection with rather stable transgene expression (Stevenson et al., 2000a), the infection of differentiated murine or rat cells is mostly inefficient, since murine or rat cells are not permissive for HVS replication. It is notable that HVS C488-transformed human T cells are also not tumorigenic in nude or SCID mice, but could induce xenogenic graft-versus-host disease, similar to primary human T cells (Huppes et al., 1994). Oncogene-deleted HVS C488 expressing an EGFP/HSV-TK fusion gene were used to efficiently transduce tumor cell lines in vitro, which could then be ablated by application of the prodrugs Ganciclovir and BrdU (Hoggarth et al., 2004). However, a previous attempt to improve the biological safety of transforming HVS vectors by inserting the prodrug-activating thymidine kinase gene of herpes simplex virus looked promising in vitro (Hiller et al., 2000b), but failed in vivo: The recombinant HVS-TK viruses not only showed no response to the administration of ganciclovir but induced tumors even more rapidly than the wild-type HVS control (Hiller et al., 2000a). Episomally persisting, nonintegrating HVS vectors could be used for therapy of hereditary genetic disorders, like cytokine receptor deficiencies (Altare et al., 2001). Since gene transfer into primary human T cells by transfection or retroviral transduction methods has remained difficult, the maintained funcional phenotype of HVS-transformed T lymphocytes, and the option to simultaneously expand T cells by transformation, makes HVS-vectors an attractive alternative for gene delivery into human T cells (Fickenscher and Fleckenstein, 2002). They may even be considered for therapeutic redirection of human T-cell antigen specificity as a tool for experimental cancer therapy applications. However, replication-deficient vector variants are necessary and a number of biosafety aspects are to be clarified.

VIII. CONCLUSIONS Prior to the discovery of KSHV, the interest in rhadinoviruses has focused on HVS, the long-established prototype of the herpesvirus subgroup, and the virus-induced fulminant simian T-cell lymphomas. These can serve as an

116

Armin Ensser and Bernhard Fleckenstein

experimental model for general tumor development, although a comparable virus-associated acute peripheral pleomorphic T-cell lymphoma is not known yet in humans. HVS leukemogenesis has also been discussed as a model for tumor induction by KSHV. Although the specific biology of the respective virus-associated tumors (KS versus T-cell lymphoma) is not directly comparable, it is obvious that these viruses have many genomic features in common. Since the putative disease-relevant genes are variable between HVS and KSHV, HVS was proposed and used as a vector for KSHV genes in order to study their pathogenic potential. A main argument in favor of this application is that HVS replicates in permissive epithelial cells, whereas it is difficult to manipulate KSHV in absence of a classical permissive system. Most rhadinoviruses have sequestered a specific set of homologs to cellular genes which is variable in different rhadinoviruses. The integration sites are concentrated in the terminal L-DNA regions, which could facilitate integration by the process of viral replication that seems to originate in these areas of the genome. Whereas functional tests for such pirated genes are not possible in the absence of a replicative system, such as in the case of KSHV, many cell-homologous genes of HVS were shown by deletion mutagenesis to be dispensable for virus replication, T-cell transformation, and pathogenicity. This has led to a new hypothesis, which puts the function of the cell-homologous genes closer to the mechanisms of persistence in the natural host species than to transformation or pathogenesis. In HVS, the basis for further research into molecular oncology was the definition of an oncogenic, transformation-associated region in the HVS genome. Later, the complete genome sequences of the HVS prototype A11—the first sequenced rhadinovirus—and subsequently, the genome sequences of a series of other rhadinoviruses were completed, including the closely related HVA. A significant chapter of HVS (and HVA) research was initiated by the observation that the HVS strain C488 is capable of transforming human T lymphocytes to stable proliferation in culture. Thus, the transforming virus functions became relevant in the context of signal transduction in human T cells, such as the specific interaction between Tip of HVS C488 and the T-cellular tyrosine kinase Lck. Stp and Tip represent new classes of membrane-bound viral oncoproteins, most likely as small adaptor polypeptides that proficiently dominate T-cell signaling. Since HVS-transformed human T cells retain many essential features of T-cell signal transduction including MHC-restricted antigen specificity, they are promising tools for laboratory studies in T-cell immunology including immunodeficiencies. Recombinant HVS vectors can efficiently deliver foreign genes into primary human T lymphocytes as well as a broad spectrum of other primary human cell types. Therapeutic applications of persisting HVS vectors for T cells, possibly in adoptive immunotherapy, became

T-Cell Transformation and Oncogenesis

117

conceivable since transformed autologous T cells were well tolerated in macaque monkeys.

ACKNOWLEDGMENTS We thank Jens Albrecht for help with the artwork. Original work included in this chapter was supported by the Deutsche Forschungsgemeinschaft (Sonderforschungsbereich 466 and 643), the Wilhelm Sander-Stiftung, the Federal Ministry of Education and Research (BMBF), and the Interdisciplinary Center for Clinical Research (IZKF) at the University of Erlangen-Nuremberg.

REFERENCES Ablashi, D. V., Schirm, S., Fleckenstein, B., Faggioni, A., Dahlberg, J., Rabin, H., Loeb, W., Armstrong, G., Peng, J. W., Aulahk, G., and Torrisi, M. R. (1985). Herpesvirus saimiriinduced lymphoblastoid rabbit cell line: Growth characteristics, virus persistence, and oncogenic properties. J. Virol. 55, 623–633. Ahuja, S. K., and Murphy, P. M. (1993). Molecular piracy of mammalian interleukin-8 receptor type B by herpesvirus saimiri. J. Biol. Chem. 268, 20691–20694. Akari, H., Mori, K., Terao, K., Otani, I., Fukasawa, M., Mukai, R., and Yoshikawa, Y. (1996). In vitro immortalization of Old World monkey T lymphocytes with herpesvirus saimiri: Its susceptibility to infection with simian immunodeficiency viruses. Virology 218, 382–388. Albrecht, J. C. (2000). Primary structure of the herpesvirus ateles genome. J. Virol. 74, 1033–1037. Albrecht, J. C., Biesinger, B., Mu¨ ller-Fleckenstein, I., Lengenfelder, D., Schmidt, M., Fleckenstein, B., and Ensser, A. (2004). Herpesvirus ateles Tio can substitute for Herpesvirus saimiri StpC and Tip oncoproteinsin growth-transformation of monkey and human T cells. J. Virol., 78, 9814–9819. Albrecht, J. C., and Fleckenstein, B. (1990). Structural organization of the conserved gene block of Herpesvirus saimiri coding for DNA polymerase, glycoprotein B, and major DNA binding protein. Virology 174, 533–542. Albrecht, J. C., and Fleckenstein, B. (1992). New member of the multigene family of complement control proteins in herpesvirus saimiri. J. Virol. 66, 3937–3940. Albrecht, J. C., Friedrich, U., Kardinal, C., Koehn, J., Fleckenstein, B., Feller, S. M., and Biesinger, B. (1999). Herpesvirus ateles gene product Tio interacts with nonreceptor protein tyrosine kinases. J. Virol. 73, 4631–4639. Albrecht, J. C., Nicholas, J., Biller, D., Cameron, K. R., Biesinger, B., Newman, C., Wittmann, S., Craxton, M. A., Coleman, H., Fleckenstein, B., and Honess, R. W. (1992a). Primary structure of the herpesvirus saimiri genome. J. Virol. 66, 5047–5058. Albrecht, J. C., Nicholas, J., Cameron, K. R., Newman, C., Fleckenstein, B., and Honess, R. W. (1992b). Herpesvirus saimiri has a gene specifying a homologue of the cellular membrane glycoprotein CD59. Virology 190, 527–530. Alexander, L., Denekamp, L., Knapp, A., Auerbach, M. R., Damania, B., and Desrosiers, R. C. (2000). The primary sequence of rhesus monkey rhadinovirus isolate 26–95: Sequence

118

Armin Ensser and Bernhard Fleckenstein

similarities to Kaposi’s sarcoma-associated herpesvirus and rhesus monkey rhadinovirus isolate 17577. J. Virol. 74, 3388–3398. Alexander, L., Du, Z., Rosenzweig, M., Jung, J. U., and Desrosiers, R. C. (1997). A role for natural simian immunodeficiency virus and human immunodeficiency virus type 1 nef alleles in lymphocyte activation. J. Virol. 71, 6094–6099. Allende, L. M., Hernandez, M., Corell, A., Garcia-Perez, M. A., Varela, P., Moreno, A., Caragol, I., Garcia-Martin, F., Guillen-Perales, J., Olive, T., Espanol, T., and ArnaizVillena, A. (2000). A novel CD18 genomic deletion in a patient with severe leucocyte adhesion deficiency: A possible CD2/lymphocyte function-associated antigen-1 functional association in humans. Immunology 99, 440–450. Alt, M., Fleckenstein, B., and Grassmann, R. (1991). A pair of selectable herpesvirus vectors for simultaneous gene expression in human lymphoid cells. Gene 102, 265–269. Alt, M., and Grassmann, R. (1993). Resistance of human fibroblasts to c-fos mediated transformation. OncoGene 8, 1421–1427. Altare, F., Durandy, A., Lammas, D., Emile, J. F., Lamhamedi, S., Le Deist, F., Drysdale, P., Jouanguy, E., Doffinger, R., Bernaudin, F., Jeppsson, O., Gollob, J. A., Meinl, E., Segal, A. W., Fischer, A., Kumararatne, D., and Casanova, J. L. (1998). Impairment of mycobacterial immunity in human interleukin-12 receptor deficiency. Science 280, 1432–1435. Altare, F., Ensser, A., Breiman, A., Reichenbach, J., Baghdadi, J. E., Fischer, A., Emile, J. F., Gaillard, J. L., Meinl, E., and Casanova, J. L. (2001). Interleukin-12 receptor beta1 deficiency in a patient with abdominal tuberculosis. J. Infect. Dis. 184, 231–236. Alvarez-Zapata, D., de-Miguel, O. S., Fontan, G., Ferreira, A., Garcia, R. M., Madero, L., van-den, E. P., and Regueiro, J. R. (1998). Phenotypical and functional characterization of Herpesvirus saimiri-immortalized human major histocompatibility complex class II-deficient T lymphocytes. Tissue Antigens 51, 250–257. Bellows, D. S., Chau, B. N., Lee, P., Lazebnik, Y., Burns, W. H., and Hardwick, J. M. (2000). Antiapoptotic herpesvirus Bcl-2 homologs escape caspase-mediated conversion to proapoptotic proteins. J. Virol. 74, 5024–5031. Berend, K. R., Jung, J. U., Boyle, T. J., DiMaio, J. M., Mungal, S. A., Desrosiers, R. C., and Lyerly, H. K. (1993). Phenotypic and functional consequences of herpesvirus saimiri infection of human CD8þ cytotoxic T lymphocytes. J. Virol. 67, 6317–6321. Biesinger, B., Mu¨ ller-Fleckenstein, I., Simmer, B., Lang, G., Wittmann, S., Platzer, E., Desrosiers, R. C., and Fleckenstein, B. (1992). Stable growth transformation of human T lymphocytes by herpesvirus saimiri. Proc. Natl. Acad. Sci. USA 89, 3116–3119. Biesinger, B., Trimble, J. J., Desrosiers, R. C., and Fleckenstein, B. (1990). The divergence between two oncogenic herpesvirus saimiri strains in a genomic region related to the transforming phenotype. Virology 176, 505–514. Biesinger, B., Tsygankov, A. Y., Fickenscher, H., Emmrich, F., Fleckenstein, B., Bolen, J. B., and Bro¨ ker, B. M. (1995). The product of the herpesvirus saimiri open reading frame 1 (tip) interacts with T cell-specific kinase p56lck in transformed cells. J. Biol. Chem. 270, 4729–4734. Bro¨ ker, B. M., Kraft, M. S., Klauenberg, U., Le Deist, F., de Villartay, J. P., Fleckenstein, B., Fleischer, B., and Meinl, E. (1997). Activation induces apoptosis in Herpesvirus saimiritransformed T cells independent of CD95 (Fas, APO-1). Eur. J. Immunol. 27, 2774–2780. Bro¨ ker, B. M., Tsygankov, A. Y., Mu¨ ller-Fleckenstein, I., Guse, A. H., Chitaev, N. A., Biesinger, B., Fleckenstein, B., and Emmrich, F. (1993). Immortalization of human T cell clones by Herpesvirus saimiri. Signal transduction analysis reveals functional CD3, CD4, and IL-2 receptors. J. Immunol. 151, 1184–1192. Calderwood, M. A., Hall, K. T., Matthews, D. A., and Whitehouse, A. (2004). The herpesvirus saimiri ORF73 gene product interacts with host-cell mitotic chromosomes and selfassociates via its C terminus. J. Gen. Virol. 85, 147–153.

T-Cell Transformation and Oncogenesis

119

Choi, J. K., Ishido, S., and Jung, J. U. (2000). The collagen repeat sequence is a determinant of the degree of herpesvirus saimiri STP transforming activity. J. Virol. 74, 8102–8110. Chou, C. S., Medveczky, M. M., Geck, P., Vercelli, D., and Medveczky, P. G. (1995). Expression of IL-2 and IL-4 in T lymphocytes transformed by herpesvirus saimiri. Virology 208, 418–426. Chung, Y. H., Cho, N. H., Garcia, M. I., Lee, S. H., Feng, P., and Jung, J. U. (2004). Activation of Stat3 transcription factor by Herpesvirus saimiri STP-A oncoprotein. J. Virol. 78, 6489–6497. Collins, C. M., Medveczky, M. M., Lund, T., and Medveczky, P. G. (2002). The terminal repeats and latency-associated nuclear antigen of herpesvirus saimiri are essential for episomal persistence of the viral genome. J. Gen. Virol. 83, 2269–2278. Cooper, M., Goodwin, D. J., Hall, K. T., Stevenson, A. J., Meredith, D. M., Markham, A. F., and Whitehouse, A. (1999). The gene product encoded by ORF 57 of herpesvirus saimiri regulates the redistribution of the splicing factor SC-35. J. Gen. Virol. 80, 1311–1316. Coscoy, L., and Ganem, D. (2000). Kaposi’s sarcoma-associated herpesvirus encodes two proteins that block cell surface display of MHC class I chains by enhancing their endocytosis. Proc. Natl. Acad. Sci. USA 97, 8051–8056. Damania, B., Li, M., Choi, J. K., Alexander, L., Jung, J. U., and Desrosiers, R. C. (1999). Identification of the R1 oncogene and its protein product from the rhadinovirus of rhesus monkeys. J. Virol. 73, 5123–5131. Daniel, M. D., Silva, D., Jackman, D., Sehgal, P., Baggs, R. B., Hunt, R. D., King, N. W., and Melendez, L. V. (1975). Reactivation of squirrel monkey heart isolate (Herpesvirus saimiri strain) from latently infected human cell cultures and induction of malignant lymphoma in marmoset monkeys. Bibl. Haematol.392–395. Daniel, M. D., Silva, D., and Ma, N. (1976). Establishment of owl monkey kidney 210 cell line for virological studies. In Vitro. 12, 290. De Carli, M., Berthold, S., Fickenscher, H., Mu¨ ller-Fleckenstein, I., D’Elios, M. M., Gao, Q., Biagiotti, R., Giudizi, M. G., Kalden, J. R., Fleckenstein, B., Romagnani, S., and Del Prete, G. (1993). Immortalization with herpesvirus saimiri modulates the cytokine secretion profile of established Th1 and Th2 human T cell clones. J. Immunol. 151, 5022–5030. Deckert, M., Ticchioni, M., Mari, B., Mary, D., and Bernard, A. (1995). The glycosylphosphatidylinositol-anchored CD59 protein stimulates both T cell receptor zeta/ZAP-70dependent and -independent signaling pathways in T cells. Eur. J. Immunol. 25, 1815–1822. Derfuss, T., Fickenscher, H., Kraft, M. S., Henning, G., Lengenfelder, D., Fleckenstein, B., and Meinl, E. (1998). Antiapoptotic activity of the herpesvirus saimiri-encoded Bcl-2 homolog: Stabilization of mitochondria and inhibition of caspase-3-like activity. J. Virol. 72, 5897–5904. Desrosiers, R. C., Bakker, A., Kamine, J., Falk, L. A., Hunt, R. D., and King, N. W. (1985a). A region of the herpesvirus saimiri genome required for oncogenicity. Science 228, 184–187. Desrosiers, R. C., Burghoff, R. L., Bakker, A., and Kamine, J. (1984). Construction of replication-competent herpesvirus saimiri deletion mutants. J. Virol. 49, 343–348. Desrosiers, R. C., and Falk, L. A. (1982). Herpesvirus saimiri strain variability. J. Virol. 43, 352–356. Desrosiers, R. C., Kamine, J., Bakker, A., Silva, D., Woychik, R. P., Sakai, D. D., and Rottman, F. M. (1985b). Synthesis of bovine growth hormone in primates by using a herpesvirus vector. Mol. Cell Biol. 5, 2796–2803. Desrosiers, R. C., Mulder, C., and Fleckenstein, B. (1979). Methylation of herpesvirus saimiri DNA in lymphoid tumor cell lines. Proc. Natl. Acad. Sci. USA 76, 3839–3843. Desrosiers, R. C., Silva, D. P., Waldron, L. M., and Letvin, N. L. (1986). Nononcogenic deletion mutants of herpesvirus saimiri are defective for in vitro immortalization. J. Virol. 57, 701–705.

120

Armin Ensser and Bernhard Fleckenstein

Drexler, H. G., Dirks, W. G., Matsuo, Y., and MacLeod, R. A. (2003). False leukemialymphoma cell lines: An update on over 500 cell lines. Leukemia 17, 416–426. Duboise, M., Guo, J., Czajak, S., Lee, H., Veazey, R., Desrosiers, R. C., and Jung, J. U. (1998a). A role for herpesvirus saimiri orf14 in transformation and persistent infection. J. Virol. 72, 6770–6776. Duboise, S. M., Guo, J., Czajak, S., Desrosiers, R. C., and Jung, J. U. (1998b). STP and Tip are essential for herpesvirus saimiri oncogenicity. J. Virol. 72, 1308–1313. Duboise, S. M., Guo, J., Desrosiers, R. C., and Jung, J. U. (1996). Use of virion DNA as a cloning vector for the construction of mutant and recombinant herpesviruses. Proc. Natl. Acad. Sci. USA 93, 11389–11394. Ensser, A., and Fleckenstein, B. (2004). Herpesvirus saimiri transformation of human T lymphocytes. In ‘‘Current Protocols in Immunology’’ ( J. E. Coligan, B. E. Bierer, D. H. Margulies, E. M. Shevach, and W. Strober, Eds.), John Wiley & Sons, Inc., New York. DOI: 10.1002/0471142735.im0721s63. Ensser, A., Glykofrydes, D., Niphuis, H., Kuhn, E. M., Rosenwirth, B., Heeney, J. L., Niedobitek, G., Mu¨ ller-Fleckenstein, I., and Fleckenstein, B. (2001). Independence of herpesvirus-induced T cell lymphoma from viral cyclin D homologue. J. Exp. Med. 193, 637–642. Ensser, A., Neipel, F., and Fickenscher, H. (2002). Rhadinovirus pathogenesis. In ‘‘Structure– Function Relationships of Human Pathogenic Viruses’’ (E. Bogner and A. Holzenburg, Eds.), pp. 349–429. Kluwer Academic Publishers/Plenum Publishers, New York. Ensser, A., Pfinder, A., Mu¨ ller-Fleckenstein, I., and Fleckenstein, B. (1999). The URNA genes of herpesvirus saimiri (strain C488) are dispensable for transformation of human T cells in vitro. J. Virol. 73, 10551–10555. Ensser, A., Thurau, M., Wittmann, S., and Fickenscher, H. (2003). The primary structure of the herpesvirus saimiri strain C488 genome. Virology 314, 471–487. Falk, L. A., Johnson, D., and Deinhardt, F. (1978). Transformation of marmoset lymphocytes in vitro with Herpesvirus ateles. Int. J. Cancer 21, 652–657. Falk, L. A., Nigida, S. M., Deinhardt, F., Wolfe, L. G., Cooper, R. W., and HernandezCamacho, J. I. (1974). Herpesvirus ateles: Properties of an oncogenic herpesvirus isolated from circulating lymphocytes of spider monkeys (Ateles sp.). Int. J. Cancer 14, 473–482. Falk, L. A., Wolfe, L. G., and Deinhardt, F. (1972). Isolation of herpesvirus saimiri from blood of squirrel monkeys (Saimiri sciureus). J. Natl. Cancer Inst. 48, 1499–1505. Fickenscher, H., Biesinger, B., Knappe, A., Wittmann, S., and Fleckenstein, B. (1996). Regulation of the herpesvirus saimiri oncogene stpC, similar to that of T-cell activation genes, in growth-transformed human T lymphocytes. J. Virol. 70, 6012–6019. Fickenscher, H., Bo¨ kel, C., Knappe, A., Biesinger, B., Meinl, E., Fleischer, B., Fleckenstein, B., and Bro¨ ker, B. M. (1997). Functional phenotype of transformed human alphabeta and gammadelta T cells determined by different subgroup C strains of herpesvirus saimiri. J. Virol. 71, 2252–2263. Fickenscher, H., and Fleckenstein, B. (2001). Herpesvirus saimiri. Philos. Trans. R. Soc. Lond. B. Biol. Sci. 356, 545–567. Fickenscher, H., and Fleckenstein, B. (2002). Growth-transformation of human T cells. Meth. Microbiology 32, 657–692. Fleckenstein, B., Bornkamm, G. W., Mulder, C., Werner, F. J., Daniel, M. D., Falk, L. A., and Delius, H. (1978a). Herpesvirus ateles DNA and its homology with Herpesvirus saimiri nucleic acid. J. Virol. 25, 361–373. Fleckenstein, B., Daniel, M. D., Hunt, R. D., Werner, J., Falk, L. A., and Mulder, C. (1978b). Tumor induction with DNA of oncogenic primate herpesviruses. Nature 274, 57–59.

T-Cell Transformation and Oncogenesis

121

Fleckenstein, B., and Desrosiers, R. C. (1982). Herpesvirus saimiri and herpesvirus ateles. In ‘‘The Herpesviruses’’ (B. Roizman, Ed.), Vol. 1, pp. 253–332. Plenum Press, New York, London. Fodor, W. L., Rollins, S. A., Bianco Caron, S., Rother, R. P., Guilmette, E. R., Burton, W. V., Albrecht, J. C., Fleckenstein, B., and Squinto, S. P. (1995). The complement control protein homolog of herpesvirus saimiri regulates serum complement by inhibiting C3 convertase activity. J. Virol. 69, 3889–3892. Fossiez, F., Banchereau, J., Murray, R., Van, K. C., Garrone, P., and Lebecque, S. (1998). Interleukin-17. Int. Rev. Immunol. 16, 541–551. Frolova-Jones, E. A., Ensser, A., Stevenson, A. J., Kinsey, S. E., and Meredith, D. M. (2000). Stable marker gene transfer into human bone marrow stromal cells and their progenitors using novel herpesvirus saimiri-based vectors. J. Hematother. Stem Cell Res. 9, 573–581. Fruehling, S., and Longnecker, R. (1997). The immunoreceptor tyrosine-based activation motif of Epstein-Barr virus LMP2A is essential for blocking BCR-mediated signal transduction. Virology 235, 241–251. Fujimuro, M., and Hayward, S. D. (2003). The latency-associated nuclear antigen of Kaposi’s sarcoma-associated herpesvirus manipulates the activity of glycogen synthase kinase-3beta. J. Virol. 77, 8019–8030. Gallego, M. D., Santamaria, M., Pena, J., and Molina, I. J. (1997). Defective actin reorganization and polymerization of Wiskott-Aldrich T cells in response to CD3-mediated stimulation. Blood 90, 3089–3097. Glykofrydes, D., Niphuis, H., Kuhn, E. M., Rosenwirth, B., Heeney, J. L., Bruder, J., Niedobitek, G., Mu¨ ller-Fleckenstein, I., Fleckenstein, B., and Ensser, A. (2000). Herpesvirus saimiri vFLIP provides an antiapoptotic function but is not essential for viral replication, transformation, or pathogenicity. J. Virol. 74, 11919–11927. Gompels, U. A., Craxton, M. A., and Honess, R. W. (1988). Conservation of gene organization in the lymphotropic herpesviruses Herpesvirus saimiri and Epstein-Barr virus. J. Virol. 62, 757–767. Goodwin, D. J., Hall, K. T., Stevenson, A. J., Markham, A. F., and Whitehouse, A. (1999). The open reading frame 57 gene product of herpesvirus saimiri shuttles between the nucleus and cytoplasm and is involved in viral RNA nuclear export. J. Virol. 73, 10519–10524. Grassmann, R., Berchtold, S., Radant, I., Alt, M., Fleckenstein, B., Sodroski, J. G., Haseltine, W. A., and Ramstedt, U. (1992). Role of human T-cell leukemia virus type 1 X region proteins in immortalization of primary human lymphocytes in culture. J. Virol. 66, 4570–4575. Grassmann, R., Dengler, C., Mu¨ ller-Fleckenstein, I., Fleckenstein, B., McGuire, K., Dokhelar, M. C., Sodroski, J. G., and Haseltine, W. A. (1989). Transformation to continuous growth of primary human T lymphocytes by human T-cell leukemia virus type I X-region genes transduced by a Herpesvirus saimiri vector. Proc. Natl. Acad. Sci. USA 86, 3351–3355. Grassmann, R., and Fleckenstein, B. (1989). Selectable recombinant herpesvirus saimiri is capable of persisting in a human T-cell line. J. Virol. 63, 1818–1821. Greve, T., Tamguney, G., Fleischer, B., Fickenscher, H., and Bro¨ ker, B. M. (2001). Downregulation of p56(lck) tyrosine kinase activity in T cells of squirrel monkeys (Saimiri sciureus) correlates with the nontransforming and apathogenic properties of herpesvirus saimiri in its natural host. J. Virol. 75, 9252–9261. Grillho¨ sl, C., Bogner, E., Wucherpfennig, F., Lengenfelder, D., Heck, E., Birkmann, A., Neipel, F., Fleckenstein, B., Ensser, A. Mutagenesis of the Herpesvirus saimiri strain C488 genome cloned as a bacterial artificial chromosome. Manuscript in preparation. Groux, H., Cottrez, F., Montpellier, C., Quatannens, B., Coll, J., Stehelin, D., and Auriault, C. (1997). Isolation and characterization of transformed human T-cell lines infected by EpsteinBarr virus. Blood 89, 4521–4530.

122

Armin Ensser and Bernhard Fleckenstein

Gru¨ ter, P., Tabernero, C., von Kobbe, C., Schmitt, C., Saavedra, C., Bachi, A., Wilm, M., Felber, B. K., and Izaurralde, E. (1998). TAP, the human homolog of Mex67p, mediates CTE-dependent RNA export from the nucleus. Mol. Cell 1, 649–659. Guo, J., Duboise, M., Lee, H., Li, M., Choi, J. K., Rosenzweig, M., and Jung, J. U. (1997). Enhanced downregulation of Lck-mediated signal transduction by a Y114 mutation of herpesvirus saimiri tip. J. Virol. 71, 7092–7096. Guo, J., Williams, K., Duboise, S. M., Alexander, L., Veazey, R., and Jung, J. U. (1998). Substitution of ras for the herpesvirus saimiri STP oncogene in lymphocyte transformation. J. Virol. 72, 3698–3704. Hall, K. T., Giles, M. S., Calderwood, M. A., Goodwin, D. J., Matthews, D. A., and Whitehouse, A. (2002). The Herpesvirus saimiri open reading frame 73 gene product interacts with the cellular protein p32. J. Virol. 76, 11612–11622. Hall, K. T., Giles, M. S., Goodwin, D. J., Calderwood, M. A., Carr, I. M., Stevenson, A. J., Markham, A. F., and Whitehouse, A. (2000). Analysis of gene expression in a human cell line stably transduced with herpesvirus saimiri. J. Virol. 74, 7331–7337. Hall, K. T., Stevenson, A. J., Goodwin, D. J., Gibson, P. C., Markham, A. F., and Whitehouse, A. (1999). The activation domain of herpesvirus saimiri R protein interacts with the TATAbinding protein. J. Virol. 73, 9756–9763. Hartley, D. A., and Cooper, G. M. (2000). Direct binding and activation of STAT transcription factors by the herpesvirus saimiri protein tip. J. Biol. Chem. 275, 16925–16932. Hartley, D. A., Hurley, T. R., Hardwick, J. S., Lund, T. C., Medveczky, P. G., and Sefton, B. M. (1999). Activation of the lck tyrosine-protein kinase by the binding of the tip protein of herpesvirus saimiri in the absence of regulatory tyrosine phosphorylation. J. Biol. Chem. 274, 20056–20059. Hasham, M. G., and Tsygankov, A. Y. (2004). Tip, an Lck-interacting protein of Herpesvirus saimiri, causes Fas- and Lck-dependent apoptosis of T lymphocytes. Virology 320, 313–329. Heck, E., Grillho¨ sl, C., Lengenfelder, D., Schmidt, M., Mu¨ ller-Fleckenstein, I., Biesinger, B., Fleckenstein, B., and Ensser, A. (2004). STAT3 phosphorylation is not required for the transformation of human T-cells by Herpesvirus saimiri. Heck, E., Lengenfelder, D., Schmidt, M., Mu¨ ller-Fleckenstein, I., Fleckenstein, B., Biesinger, B., and Ensser, A. (2005). T-cell Growth Transformation by Herpesvirus saimiri is independent of STAT3-activation. J. Virol. in press. Henle, W., Diehl, V., Kohn, G., Zur, H. H., and Henle, G. (1967). Herpes-type virus and chromosome marker in normal leukocytes after growth with irradiated Burkitt cells. Science 157, 1064–1065. Hiller, C., Tamguney, G., Stolte, N., Matz-Rensing, K., Lorenzen, D., Hor, S., Thurau, M., Wittmann, S., Slavin, S., and Fickenscher, H. (2000a). Herpesvirus saimiri pathogenicity enhanced by thymidine kinase of herpes simplex virus. Virology 278, 445–455. Hiller, C., Wittmann, S., Slavin, S., and Fickenscher, H. (2000b). Functional long-term thymidine kinase suicide gene expression in human T cells using a herpesvirus saimiri vector. Gene Ther. 7, 664–674. Hoggarth, J. H., Jones, E., Ensser, A., and Meredith, D. M. (2004). Functional expression of thymidine kinase in human leukaemic and colorectal cells, delivered as EGFP fusion protein by herpesvirus saimiri-based vector. Cancer Gene Ther. 11, 613–624. Ho¨ r, S., Ensser, A., Reiss, C., Ballmer-Hofer, K., and Biesinger, B. (2001). Herpesvirus saimiri protein StpB associates with cellular Src. J. Gen. Virol. 82, 339–344. Ho¨ r, S., Pirzer, H., Dumoutier, L., Bauer, F., Wittmann, S., Sticht, H., Renauld, J. C., De Waal, M. R., and Fickenscher, H. (2004). The T-cell lymphokine interleukin-26 targets epithelial cells through the interleukin-20 receptor 1 and interleukin-10 receptor 2 chains. J. Biol. Chem. 279, 33343–33351.

T-Cell Transformation and Oncogenesis

123

Hunt, R. D., Melendez, L. V., Garcia, F. G., and Trum, B. F. (1972). Pathologic features of Herpesvirus ateles lymphoma in cotton-topped marmosets (Saguinus oedipus). J. Natl. Cancer Inst. 49, 1631–1639. Huppes, W., Fickenscher, H., ’t Hart, B. A., and Fleckenstein, B. (1994). Cytokine dependence of human to mouse graft-versus-host disease. Scand. J. Immunol. 40, 26–36. Inatsuki, A., Yasukawa, M., and Kobayashi, Y. (1989). Functional alterations of herpes simplex virus-specific CD4þ multifunctional T cell clones following infection with human T lymphotropic virus type I. J. Immunol. 143, 1327–1333. Ishido, S., Choi, J. K., Lee, B. S., Wang, C., DeMaria, M., Johnson, R. P., Cohen, G. B., and Jung, J. U. (2000a). Inhibition of natural killer cell-mediated cytotoxicity by Kaposi’s sarcoma-associated herpesvirus K5 protein. Immunity 13, 365–374. Ishido, S., Wang, C., Lee, B. S., Cohen, G. B., and Jung, J. U. (2000b). Downregulation of major histocompatibility complex class I molecules by Kaposi’s sarcoma-associated herpesvirus K3 and K5 proteins. J. Virol. 74, 5300–5309. Jeong, J. H., Orvis, J., Kim, J. W., McMurtrey, C. P., Renne, R., and Dittmer, D. P. (2004). Regulation and autoregulation of the promoter for the latency-associated nuclear antigen of Kaposi’s sarcoma-associated herpesvirus. J. Biol. Chem. 279, 16822–16831. Jung, J. U., and Desrosiers, R. C. (1991). Identification and characterization of the herpesvirus saimiri oncoprotein STP-C488. J. Virol. 65, 6953–6960. Jung, J. U., and Desrosiers, R. C. (1992). Herpesvirus saimiri oncogene STP-C488 encodes a phosphoprotein. J. Virol. 66, 1777–1780. Jung, J. U., and Desrosiers, R. C. (1994). Distinct functional domains of STP-C488 of herpesvirus saimiri. Virology 204, 751–758. Jung, J. U., and Desrosiers, R. C. (1995). Association of the viral oncoprotein STP-C488 with cellular ras. Mol. Cell. Biol. 15, 6506–6512. Jung, J. U., Lang, S. M., Friedrich, U., Jun, T., Roberts, T. M., Desrosiers, R. C., and Biesinger, B. (1995a). Identification of Lck-binding elements in tip of herpesvirus saimiri. J. Biol. Chem. 270, 20660–20667. Jung, J. U., Lang, S. M., Jun, T., Roberts, T. M., Veillette, A., and Desrosiers, R. C. (1995b). Downregulation of Lck-mediated signal transduction by tip of herpesvirus saimiri. J. Virol. 69, 7814–7822. Jung, J. U., Stager, M., and Desrosiers, R. C. (1994). Virus-encoded cyclin. Mol. Cell. Biol. 14, 7235–7244. Jung, J. U., Trimble, J. J., King, N. W., Biesinger, B., Fleckenstein, B. W., and Desrosiers, R. C. (1991). Identification of transforming genes of subgroup A and C strains of herpesvirus saimiri. Proc. Natl. Acad. Sci. USA 88, 7051–7055. Kaschka-Dierich, C., Werner, F. J., Bauer, I., and Fleckenstein, B. (1982). Structure of nonintegrated, circular Herpesvirus saimiri and Herpesvirus ateles genomes in tumor cell lines and in vitro-transformed cells. J. Virol. 44, 295–310. Kiyotaki, M., Desrosiers, R. C., and Letvin, N. L. (1986). Herpesvirus saimiri strain 11 immortalizes a restricted marmoset T8 lymphocyte subpopulation in vitro. J. Exp. Med. 164, 926–931. Knappe, A., Feldmann, G., Dittmer, U., Meinl, E., Nisslein, T., Wittmann, S., Ma¨ tz-Rensing, K., Kirchner, T., Bodemer, W., and Fickenscher, H. (2000a). Herpesvirus saimiri-transformed macaque T cells are tolerated and do not cause lymphoma after autologous reinfusion. Blood 95, 3256–3261. Knappe, A., Hiller, C., Niphuis, H., Fossiez, F., Thurau, M., Wittmann, S., Kuhn, E. M., Lebecque, S., Banchereau, J., Rosenwirth, B., Fleckenstein, B., Heeney, J., and Fickenscher, H. (1998a). The interleukin-17 gene of herpesvirus saimiri. J. Virol. 72, 5797–5801.

124

Armin Ensser and Bernhard Fleckenstein

Knappe, A., Hiller, C., Thurau, M., Wittmann, S., Hofmann, H., Fleckenstein, B., and Fickenscher, H. (1997). The superantigen-homologous viral immediate-early gene ie14/vsag in herpesvirus saimiri-transformed human T cells. J. Virol. 71, 9124–9133. Knappe, A., Ho¨ r, S., Wittmann, S., and Fickenscher, H. (2000b). Induction of a novel cellular homolog of interleukin-10, AK155, by transformation of T lymphocytes with herpesvirus saimiri. J. Virol. 74, 3881–3887. Knappe, A., Thurau, M., Niphuis, H., Hiller, C., Wittmann, S., Kuhn, E. M., Rosenwirth, B., Fleckenstein, B., Heeney, J., and Fickenscher, H. (1998b). T-cell lymphoma caused by herpesvirus saimiri C488 independently of ie14/vsag, a viral gene with superantigen homology. J. Virol. 72, 3469–3471. Koomey, J. M., Mulder, C., Burghoff, R. L., Fleckenstein, B., and Desrosiers, R. C. (1984). Deletion of DNA sequence in a nononcogenic variant of Herpesvirus saimiri. J. Virol. 50, 662–665. Korty, P. E., Brando, C., and Shevach, E. M. (1991). CD59 functions as a signal-transducing molecule for human T cell activation. J. Immunol. 146, 4092–4098. Kretschmer, C., Murphy, C., Biesinger, B., Beckers, J., Fickenscher, H., Kirchner, T., Fleckenstein, B., and Ru¨ ther, U. (1996). A herpes saimiri oncogene causing peripheral T-cell lymphoma in transgenic mice. OncoGene 12, 1609–1616. Kung, S. H., and Medveczky, P. G. (1996). Identification of a herpesvirus saimiri cis-acting DNA fragment that permits stable replication of episomes in transformed T-cells. J. Virol. 70, 1738–1744. Lan, K., Kuppers, D. A., Verma, S. C., and Robertson, E. S. (2004). Kaposi’s sarcomaassociated herpesvirus-encoded latency-associated nuclear antigen inhibits lytic replication by targeting Rta: A potential mechanism for virus-mediated control of latency. J. Virol. 78, 6585–6594. Lang, G., and Fleckenstein, B. (1990). Trans-activation of the thymidylate synthase promoter of herpesvirus saimiri. J. Virol. 64, 5333–5341. Lee, H., Choi, J. K., Li, M., Kaye, K., Kieff, E., and Jung, J. U. (1999). Role of cellular tumor necrosis factor receptor-associated factors in NF-kappaB activation and lymphocyte transformation by herpesvirus saimiri STP. J. Virol. 73, 3913–3919. Lee, H., Trimble, J. J., Yoon, D. W., Regier, D., Desrosiers, R. C., and Jung, J. U. (1997). Genetic variation of herpesvirus saimiri subgroup A transforming protein and its association with cellular src. J. Virol. 71, 3817–3825. Lee, H., Veazey, R., Williams, K., Li, M., Guo, J., Neipel, F., Fleckenstein, B., Lackner, A., Desrosiers, R. C., and Jung, J. U. (1998). Deregulation of cell growth by the K1 gene of Kaposi’s sarcoma-associated herpesvirus. Nat. Med. 4, 435–440. Lee, S. H., Chung, Y. H., Cho, N. H., Gwack, Y., Feng, P., and Jung, J. U. (2004). Modulation of T-cell receptor signal transduction by herpesvirus signaling adaptor protein. Mol. Cell. Biol. 24, 5369–5382. Lund, T., Medveczky, M. M., and Medveczky, P. G. (1997a). Herpesvirus saimiri Tip-484 membrane protein markedly increases p56lck activity in T cells. J. Virol. 71, 378–382. Lund, T. C., Garcia, R., Medveczky, M. M., Jove, R., and Medveczky, P. G. (1997b). Activation of STAT transcription factors by herpesvirus saimiri Tip-484 requires p56lck. J. Virol. 71, 6677–6682. Lund, T. C., Prator, P. C., Medveczky, M. M., and Medveczky, P. G. (1999). The Lck binding domain of herpesvirus saimiri tip-484 constitutively activates Lck and STAT3 in T cells. J. Virol. 73, 1689–1694. Means, R. E., Ishido, S., Alvarez, X., and Jung, J. U. (2002). Multiple endocytic trafficking pathways of MHC class I molecules induced by a Herpesvirus protein. EMBO J. 21, 1638–1649.

T-Cell Transformation and Oncogenesis

125

Medveczky, M. M., Geck, P., Sullivan, J. L., Serbousek, D., Djeu, J. Y., and Medveczky, P. G. (1993). IL-2 independent growth and cytotoxicity of herpesvirus saimiri-infected human CD8 cells and involvement of two open reading frame sequences of the virus. Virology 196, 402–412. Medveczky, M. M., Szomolanyi, E., Hesselton, R., DeGrand, D., Geck, P., and Medveczky, P. G. (1989). Herpesvirus saimiri strains from three DNA subgroups have different oncogenic potentials in New Zealand white rabbits. J. Virol. 63, 3601–3611. Medveczky, P., Szomolanyi, E., Desrosiers, R. C., and Mulder, C. (1984). Classification of herpesvirus saimiri into three groups based on extreme variation in a DNA region required for oncogenicity. J. Virol. 52, 938–944. Meinl, E., Fickenscher, H., Hoch, R. M., de Waal Malefyt, R., ’t Hart, B. A., Wekerle, H., Hohlfeld, R., and Fleckenstein, B. (1997). Growth transformation of antigen-specific T cell lines from rhesus monkeys by herpesvirus saimiri. Virology 229, 175–182. Meinl, E., Hohlfeld, R., Wekerle, H., and Fleckenstein, B. (1995). Immortalization of human T cells by Herpesvirus saimiri. Immunol. Today 16, 55–58. Melendez, L. V., Daniel, M. D., Hunt, R. D., and Garcia, F. G. (1968). An apparently new herpesvirus from primary kidney cultures of the squirrel monkey (Saimiri sciureus). Lab. Anim. Care 18, 374–381. Melendez, L. V., Hunt, R. D., Daniel, M. D., Garcia, F. G., and Fraser, C. E. (1969). Herpesvirus saimiri. II. Experimentally induced malignant lymphoma in primates. Lab. Anim. Sci. 19, 378–386. Melendez, L. V., Hunt, R. D., King, N. W., Barahona, H. H., Daniel, M. D., Fraser, C. E., and Garcia, F. G. (1972). Herpesvirus ateles, a new lymphoma virus of monkeys. Nature New Biol. 235, 182–184. Meuer, S. C., Hussey, R. E., Fabbi, M., Fox, D., Acuto, O., Fitzgerald, K. A., Hodgdon, J. C., Protentis, J. P., Schlossman, S. F., and Reinherz, E. L. (1984). An alternative pathway of T-cell activation: A functional role for the 50 kd T11 sheep erythrocyte receptor protein. Cell 36, 897–906. Miller, C. L., Lee, J. H., Kieff, E., and Longnecker, R. (1994). An integral membrane protein (LMP2) blocks reactivation of Epstein-Barr virus from latency following surface immunoglobulin crosslinking. Proc. Natl. Acad. Sci. USA 91, 772–776. Mittru¨ cker, H. W., Mu¨ ller-Fleckenstein, I., Fleckenstein, B., and Fleischer, B. (1992). CD2mediated autocrine growth of herpes virus saimiri-transformed human T lymphocytes. J. Exp. Med. 176, 909–913. Miyoshi, I., Kubonishi, I., Yoshimoto, S., Akagi, T., Ohtsuki, Y., Shiraishi, Y., Nagata, K., and Hinuma, Y. (1981). Type C virus particles in a cord T-cell line derived by co-cultivating normal human cord leukocytes and human leukemic T cells. Nature 294, 770–771. Murphy, C., Kretschmer, C., Biesinger, B., Beckers, J., Jung, J., Desrosiers, R. C., Mu¨ llerHermelink, H. K., Fleckenstein, B. W., and Ru¨ ther, U. (1994). Epithelial tumours induced by a herpesvirus oncogene in transgenic mice. OncoGene 9, 221–226. Murphy, P. M. (1994). The molecular biology of leukocyte chemoattractant receptors. Annu. Rev. Immunol. 12, 593–633. Murthy, S. C., Trimble, J. J., and Desrosiers, R. C. (1989). Deletion mutants of herpesvirus saimiri define an open reading frame necessary for transformation. J. Virol. 63, 3307–3314. Nava, V. E., Cheng, E. H. Y., Veliuona, M., Zou, S., Clem, R. J., Mayer, M. L., and Hardwick, J. M. (1997). Herpesvirus saimiri encodes a functional homolog of the human bcl-2 oncogene. J. Virol. 71, 4118–4122. Neipel, F., Albrecht, J. C., and Fleckenstein, B. (1997). Cell-homologous genes in the Kaposi’s sarcoma associated rhadinovirus human herpesvirus 8: Determinants of its pathogenicity? J. Virol. 71, 4187–4192.

126

Armin Ensser and Bernhard Fleckenstein

Nicholas, J., Cameron, K. R., and Honess, R. W. (1992). Herpesvirus saimiri encodes homologues of G protein-coupled receptors and cyclins. Nature 355, 362–365. Nicholas, J., Coles, L. S., Newman, C., and Honess, R. W. (1991). Regulation of the herpesvirus saimiri (HVS) delayed-early 110- kilodalton promoter by HVS immediate-early gene products and a homolog of the Epstein-Barr virus R trans activator. J. Virol. 65, 2457–2466. Nicholas, J., Gompels, U. A., Craxton, M. A., and Honess, R. W. (1988). Conservation of sequence and function between the product of the 52-kilodalton immediate-early gene of herpesvirus saimiri and the BMLF1-encoded transcriptional effector (EB2) of Epstein-Barr virus. J. Virol. 62, 3250–3257. Nick, S., Fickenscher, H., Biesinger, B., Born, G., Jahn, G., and Fleckenstein, B. (1993). Herpesvirus saimiri transformed human T cell lines: A permissive system for human immunodeficiency viruses. Virology 194, 875–877. Park, J., Cho, N. H., Choi, J. K., Feng, P., Choe, J., and Jung, J. U. (2003). Distinct roles of cellular Lck and p80 proteins in herpesvirus saimiri Tip function on lipid rafts. J. Virol. 77, 9041–9051. Park, J., Lee, B. S., Choi, J. K., Means, R. E., Choe, J., and Jung, J. U. (2002). Herpesviral protein targets a cellular WD repeat endosomal protein to downregulate T lymphocyte receptor expression. Immunity 17, 221–233. Randall, R. E., Newman, C., and Honess, R. W. (1984). A single major immediate-early virus gene product is synthesized in cells productively infected with herpesvirus saimiri. J. Gen. Virol. 65, 1215–1219. Randall, R. E., Newman, C., and Honess, R. W. (1985). Asynchronous expression of the immediate-early protein of herpesvirus saimiri in populations of productively infected cells. J. Gen. Virol. 66, 2199–2213. Rodriguez-Gallego, C., Corell, A., Pacheco, A., Timon, M., Regueiro, J. R., Allende, L. M., Madrono, A., and Arnaiz, V. A. (1996). Herpes virus saimiri transformation of T cells in CD3 gamma immunodeficiency: Phenotypic and functional characterization. J. Immunol. Methods 198, 177–186. Roizman, B., Desrosiers, R. C., Fleckenstein, B., Lopez, C., Minson, A. C., and Studdert, M. J. (1992). The family Herpesviridae: An update. The Herpesvirus Study Group of the International Committee on Taxonomy of Viruses. Arch. Virol. 123, 425–449. Rosenkilde, M. M., McLean, K. A., and Schwartz, T. W. (2004). The CXC-chemokine receptor encoded by Herpesvirus saimiri, ECRF3, shows ligand-regulated signaling through Gi, Gq, and G12/13 proteins—but constitutive signaling through only Gi and G12/13 proteins. J. Biol. Chem. 279, 32524–32533. Rother, R. P., Rollins, S. A., Fodor, W. L., Albrecht, J. C., Setter, E., Fleckenstein, B., and Squinto, S. P. (1994). Inhibition of complement-mediated cytolysis by the terminal complement inhibitor of herpesvirus saimiri. J. Virol. 68, 730–737. Rouvier, E., Luciani, M. F., Mattei, M. G., Denizot, F., and Golstein, P. (1993). CTLA-8, cloned from an activated T cell, bearing AU-rich messenger RNA instability sequences, and homologous to a herpesvirus saimiri gene. J. Immunol. 150, 5445–5456. Russo, J. J., Bohenzky, R. A., Chien, M.-C., Chen, J., Yan, M., Maddalena, D., Parry, J. P., Peruzzi, D., Edelman, I. S., Chang, Y., and Moore, P. S. (1996). Nucleotide sequence of the Kaposi’s sarcoma associated herpesvirus (HHV8). Proc. Natl. Acad. Sci. USA 93, 14862–14867. Scha¨ fer, A., Lengenfelder, D., Grillho¨ sl, C., Wieser, C., Fleckenstein, B., and Ensser, A. (2003). The latency-associated nuclear antigen homolog of herpesvirus saimiri inhibits lytic virus replication. J. Virol. 77, 5911–5925. Schirm, S., Mu¨ ller, I., Desrosiers, R. C., and Fleckenstein, B. (1984). Herpesvirus saimiri DNA in a lymphoid cell line established by in vitro transformation. J. Virol. 49, 938–946.

T-Cell Transformation and Oncogenesis

127

Schneider, U., Schwenk, H. U., and Bornkamm, G. (1977). Characterization of EBV-genome negative ‘‘null’’ and ‘‘T’’ cell lines derived from children with acute lymphoblastic leukemia and leukemic transformed non-Hodgkin lymphoma. Int. J. Cancer 19, 621–626. Schofield, A. (1994). Investigations of the origins of replication of herpesvirus saimiri, pp. 1–220. Open University, Ph.D. Thesis. Schweimer, K., Hoffmann, S., Bauer, F., Friedrich, U., Kardinal, C., Feller, S. M., Biesinger, B., and Sticht, H. (2002). Structural investigation of the binding of a herpesviral protein to the SH3 domain of tyrosine kinase Lck. Biochemistry 41, 5120–5130. Searles, R. P., Bergquam, E. P., Axthelm, M. K., and Wong, S. W. (1999). Sequence and genomic analysis of a Rhesus macaque rhadinovirus with similarity to Kaposi’s sarcoma-associated herpesvirus/human herpesvirus 8. J. Virol. 73, 3040–3053. Simmer, B., Alt, M., Buckreus, I., Berthold, S., Fleckenstein, B., Platzer, E., and Grassmann, R. (1991). Persistence of selectable herpesvirus saimiri in various human hematopoietic and epithelial cell lines. J. Gen. Virol. 72, 1953–1958. Sinclair, A. J. (2003). bZIP proteins of human gammaherpesviruses. J. Gen. Virol. 84, 1941–1949. Sorokina, E. M., Merlo, J. J., Jr., and Tsygankov, A. Y. (2004). Molecular mechanisms of the effect of Herpesvirus saimiri protein StpC on the signaling pathway leading to NF-kappa B activation. J. Biol. Chem. 279, 13469–13477. Stephan, V., Wahn, V., Le Deist, F., Dirksen, U., Bro¨ ker, B. M., Mu¨ ller-Fleckenstein, I., Horneff, G., Schroten, H., Fischer, A., and de Saint Basile, G. (1996). Atypical X-linked severe combined immunodeficiency due to possible spontaneous reversion of the genetic defect in T cells. N. Engl. J. Med. 335, 1563–1567. Stevenson, A. J., Clarke, D., Meredith, D. M., Kinsey, S. E., Whitehouse, A., and Bonifer, C. (2000a). Herpesvirus saimiri-based gene delivery vectors maintain heterologous expression throughout mouse embryonic stem cell differentiation in vitro. Gene Ther. 7, 464–471. Stevenson, A. J., Cooper, M., Griffiths, J. C., Gibson, P. C., Whitehouse, A., Jones, E. F., Markham, A. F., Kinsey, S. E., and Meredith, D. M. (1999). Assessment of Herpesvirus saimiri as a potential human gene therapy vector. J. Med. Virol. 57, 269–277. Stevenson, A. J., Giles, M. S., Hall, K. T., Goodwin, D. J., Calderwood, M. A., Markham, A. F., and Whitehouse, A. (2000b). Specific oncolytic activity of herpesvirus saimiri in pancreatic cancer cells. Br. J. Cancer 83, 329–332. Swanton, C., Mann, D. J., Fleckenstein, B., Neipel, F., Peters, G., and Jones, N. (1997). Herpes viral cyclin/Cdk6 complexes evade inhibition by CDK inhibitor proteins. Nature 390, 184–187. Szomolanyi, E., Medveczky, P., and Mulder, C. (1987). In vitro immortalization of marmoset cells with three subgroups of herpesvirus saimiri. J. Virol. 61, 3485–3490. Thome, M., Schneider, P., Hofmann, K., Fickenscher, H., Meinl, E., Neipel, F., Mattmann, C., Burns, K., Bodmer, J. L., Schroter, M., Scaffidi, C., Krammer, P. H., Peter, M. E., and Tschopp, J. (1997). Viral FLICE-inhibitory proteins (FLIPs) prevent apoptosis induced by death receptors. Nature 386, 517–521. Thomson, B. J., and Nicholas, J. (1991). Superantigen function. Nature 351, 530. Thurau, M., Whitehouse, A., Wittmann, S., Meredith, D., and Fickenscher, H. (2000). Distinct transcriptional and functional properties of the R transactivator gene orf50 of the transforming herpesvirus saimiri strain C488. Virology 268, 167–177. Troidl, B., Simmer, B., Fickenscher, H., Mu¨ ller-Fleckenstein, I., Emmrich, F., Fleckenstein, B., and Gebhart, E. (1994). Karyotypic characterization of human T-cell lines immortalized by Herpesvirus saimiri. Int. J. Cancer 56, 433–438. van Dyk, L. F., Hess, J. L., Katz, J. D., Jacoby, M., Speck, S. H., and Virgin, H. W., IV (1999). The murine gammaherpesvirus 68 v-cyclin gene is an oncogene that promotes cell cycle progression in primary lymphocytes. J. Virol. 73, 5110–5122.

128

Armin Ensser and Bernhard Fleckenstein

van Dyk, L. F., Virgin, H. W., IV, and Speck, S. H. (2000). The murine gammaherpesvirus 68 v-cyclin is a critical regulator of reactivation from latency. J. Virol. 74, 7451–7461. Verma, S. C., and Robertson, E. S. (2003). ORF73 of herpesvirus Saimiri strain C488 tethers the viral genome to metaphase chromosomes and binds to cis-acting DNA sequences in the terminal repeats. J. Virol. 77, 12494–12506. Weber, F., Meinl, E., Drexler, K., Czlonkowska, A., Huber, S., Fickenscher, H., Mu¨ ller-Fleckenstein, I., Fleckenstein, B., Wekerle, H., and Hohlfeld, R. (1993). Transformation of human T-cell clones by Herpesvirus saimiri: Intact antigen recognition by autonomously growing myelin basic protein-specific T cells. Proc. Natl. Acad. Sci. USA 90, 11049–11053. White, R. E., Calderwood, M. A., and Whitehouse, A. (2003). Generation and precise modification of a herpesvirus saimiri bacterial artificial chromosome demonstrates that the terminal repeats are required for both virus production and episomal persistence. J. Gen. Virol. 84, 3393–3403. Whitehouse, A., Cooper, M., and Meredith, D. M. (1998). The immediate-early gene product encoded by open reading frame 57 of herpesvirus saimiri modulates gene expression at a posttranscriptional level. J. Virol. 72, 857–861. Whitehouse, A., Stevenson, A. J., Cooper, M., and Meredith, D. M. (1997). Identification of cisacting element within the herpesvirus saimiri ORF 6 promoter that is responsive to the HVS. R transactivator. J. Gen. Virol. 78, 1411–1415. Wiese, N., Tsygankov, A. Y., Klauenberg, U., Bolen, J. B., Fleischer, B., and Bro¨ ker, B. M. (1996). Selective activation of T cell kinase p56lck by Herpesvirus saimiri protein tip. J. Biol. Chem. 271, 847–852. Wright, J., Falk, L. A., Collins, D., and Deinhardt, F. (1976). Mononuclear cell fraction carrying Herpesvirus saimiri in persistently infected squirrel monkeys. J. Natl. Cancer Inst. 57, 959–962. Yamamoto, N., Okada, M., Koyanagi, Y., Kannagi, M., and Hinuma, Y. (1982). Transformation of human leukocytes by cocultivation with an adult T cell leukemia virus producer cell line. Science 217, 737–739. Yamanashi, Y., Mori, S., Yoshida, M., Kishimoto, T., Inoue, K., Yamamoto, T., and Toyoshima, K. (1989). Selective expression of a protein-tyrosine kinase, p56lyn, in hematopoietic cells and association with production of human T-cell lymphotropic virus type I. Proc. Natl. Acad. Sci. USA 86, 6538–6542. Yao, Z. B., Fanslow, W. C., Seldin, M. F., Rousseau, A. M., Painter, S. L., Comeau, M. R., Cohen, J. I., and Spriggs, M. K. (1995). Herpesvirus saimiri encodes a new cytokine, IL-17, which binds to a novel cytokine receptor. Immunity 3, 811–821. Yao, Z. B., Maraskovsky, E., Spriggs, M. K., Cohen, J. I., Armitage, R. J., and Alderson, M. R. (1996). Herpesvirus saimiri open reading frame 14, a protein encoded by a T-lymphotropic herpesvirus, binds to MHC class-II molecules and stimulates T-cell proliferation. J. Immunol. 156, 3260–3266. Yoon, D. W., Lee, H., Seol, W., DeMaria, M., Rosenzweig, M., and Jung, J. U. (1997). Tap: A novel cellular protein that interacts with tip of herpesvirus saimiri and induces lymphocyte aggregation. Immunity 6, 571–582. Yssel, H., De Waal, M. R., Duc, D., Blanchard, D., Gazzolo, L., de Vries, J. E., and Spits, H. (1989). Human T cell leukemia/lymphoma virus type I infection of a CD4þ proliferative/ cytotoxic T cell clone progresses in at least two distinct phases based on changes in function and phenotype of the infected cells. J. Immunol. 142, 2279–2289. Zapata, D. A., Pacheco-Castro, A., Torres, P. S., Ramiro, A. R., Jose, E. S., Alarcon, B., Alibaud, L., Rubin, B., Toribio, M. L., and Regueiro, J. R. (1999). Conformational and biochemical differences in the TCR.CD3 complex of CD8(þ) versus CD4(þ) mature lymphocytes revealed in the absence of CD3gamma. J. Biol. Chem. 274, 35119–35128.

Chaperoning Antigen Presentation by MHC Class II Molecules and Their Role in Oncogenesis Marije Marsman,* Ingrid Jordens,* Alexander Griekspoor, and Jacques Neefjes Division of Tumor Biology, The Netherlands Cancer Institute, 1066 CX Amsterdam, The Netherlands

I. Introduction II. Multiple Steps in MHC Class II Antigen Presentation A. A Brief Introduction to the Process of Antigen Presentation B. The Invariant Chain, a Transporting Pseudopeptide C. Processing of the Invariant Chain D. The Invariant Chain, More Than Just an MHC Class II Chaperone E. HLA-DM, the Editor for Antigenic Peptide Loading of MHC Class II Molecules F. HLA-DO, the Chaperone of Chaperones G. From the MIIC to the Plasma Membrane III. Interfering with Antigen Presentation by MHC Class II Molecules A. Promoting Antigen Presentation B. Inhibiting Antigen Presentation IV. MHC Class II Molecules in Oncogenesis A. Immune System Involved in Tumor Surveillance B. MHC Class II Expression and Tumor Development C. Toward MHC Class II–Restricted Tumor Immunotherapy V. Conclusions References Tumor vaccine development aimed at stimulating the cellular immune response focuses mainly on MHC class I molecules. This is not surprising since most tumors do not express MHC class II or CD1 molecules. Nevertheless, the most successful targets for cancer immunotherapy, leukemia and melanoma, often do express MHC class II molecules, which leaves no obvious reason to ignore MHC class II molecules as a mediator in anticancer immune therapy. We review the current state of knowledge on the process of MHC class II–restricted antigen presentation and subsequently discuss the consequences of MHC class II expression on tumor surveillance and the induction of an efficient MHC class II mediated antitumor response in vivo and after vaccination. ß 2005 Elsevier Inc.

I. INTRODUCTION The MHC class I pathway is the only known system able to present intracell antigens to the immune system. CD8þ cytotoxic T cells recognize the *Equal Contribution Advances in CANCER RESEARCH 0065-230X/05 $35.00

129

Copyright 2005, Elsevier Inc. All rights reserved

130

Marije Marsman et al.

Major Histocompatibility Complex Class I–peptide (MHC I–peptide) combination and subsequently eliminate cells presenting altered self fragments. As a consequence, this mechanism efficiently clears cells with viruses or other intracellular pathogens. This could include tumor-causing viruses like human papilloma virus, Epstein-Barr virus (EBV) and hepatitis virus. Moreover, cells with mutated self-proteins will also present antigenic fragments of these proteins to CD8þ cytotoxic T cells. Since various mutated proteins can cause cancer, the resulting tumors can be targets for immune surveillance by cytotoxic T cells (Melief, 1992). This, plus the fact that MHC class I molecules are expressed on virtually all cells, makes them the most intensively studied targets in the development of strategies for tumor vaccination (Dudley et al., 2002). Unlike MHC class I molecules, MHC class II molecules are expressed mainly in the hemapoietic system but can also be expressed in other cell types after stimulation by, for example, interferon . Due to this restricted tissue distribution, MHC class II molecules are less popular as mediators in tumor vaccination strategies. Interestingly, the tumors that are currently treated by various vaccination protocols are mainly leukemia and melanoma, tumors that often express MHC class II molecules. These tumors express the accessory panel of proteins necessary for successful loading of MHC class II molecules with antigenic peptides. Since the proteases expressed in tumors may be different from those expressed in normal hemapoietic cells, different fragments of a normal antigen can be generated and presented, rendering novel antigenic peptides and thus different T cells responses. The resulting CD4þ T cell response may directly eliminate such tumor cells and/or stimulate surrounding CD8þ or NK killer T cells to do so. The fact is that this part of the cellular immune response against tumors is still largely ignored. Here, we first describe the system required for successful antigen presentation by MHC class II molecules and how this is regulated with the help of three dedicated chaperones. We then discuss the current state of knowledge on the relationship between MHC class II expression and cancer. Finally, the potential application of MHC class II–restricted antigen presentation in the development of tumor vaccination strategies will be discussed.

II. MULTIPLE STEPS IN MHC CLASS II ANTIGEN PRESENTATION A. A Brief Introduction to the Process of Antigen Presentation Although MHC class I and II molecules are very similar in structure and both present peptide fragments (Fig. 1), they differ in almost all other

MHC Class II Molecules in Cancer

131

aspects. The major difference is the source of antigens sampled by these molecules. MHC class I presents fragments from cytoplasmic or nuclear antigens. MHC class II molecules present fragments from proteins degraded in the endocytic pathway (Cresswell, 2000; Wubbolts and Neefjes, 1999). This implies that many steps in the process of successful peptide loading of MHC class II molecules differ from that of MHC class I molecules (Fig. 1). MHC class II molecules are composed of an  and a chain that assemble in the endoplasmic reticulum (ER) into an  heterodimer. Subsequently, a third chain, the invariant chain or Ii, interacts with this heterodimer to form a heterotrimer. In fact, a trimer of this heterotrimer is formed, resulting in a nonameric complex (Cresswell, 1994; Marks et al., 1990; Roche et al., 1991). Ii acts as a sort of a pseudosubstrate by allowing a small segment (called CLIP, for Class II-associated Ii peptide) to enter the peptide-binding groove of MHC class II. Moreover, Ii is necessary for transport out of the ER, as illustrated in mice lacking Ii, which show a reduced surface expression of MHC class II (Bikoff et al., 1993; Viville et al., 1993). Whereas most proteins, including MHC class I, are transported via the Golgi directly to the plasma membrane, Ii targets MHC class II molecules from the trans-Golgi network to late endosomal structures, collectively called MIIC for ‘‘MHC class II-containing compartments’’ (Neefjes et al., 1990). In the MIIC, all the requirements for efficient peptide loading of MHC class II are concentrated: First, proteases that degrade Ii until only the CLIP fragment is left in the peptide-binding groove of MHC class II molecules (Neefjes and Ploegh, 1992). Second, proteases, reductases, and unfoldases that process antigenic fragments which have entered the cell by receptor-mediated or fluid-phase endocytosis (Lennon-Dumenil et al., 2002). Finally, HLA-DM mediating the exchange of the CLIP fragment for fragments generated from the endocytosed antigens (Sanderson et al., 1994). The activity of HLA-DM, in turn, can be controlled by a chaperone-of-chaperones called HLA-DO (Denzin et al., 1997; Liljedahl et al., 1996; van Ham et al., 1997). Thus, in these specialized MIIC, the unique combination of dedicated (endosomal) chaperones and proteolytic activity supports proper antigen loading of MHC class II molecules. Loaded MHC class II molecules are subsequently transported to the plasma membrane for presentation of endocytosed antigens to CD4þ T cells (Germain and Rinker, 1993; Neefjes and Ploegh, 1992).

B. The Invariant Chain, a Transporting Pseudopeptide As has been discussed, MHC class II molecules can encounter at least three different specialized chaperones during their existence. The formation of the MHC class II  heterodimer is the first step in the formation of the MHC

132 Fig. 1 Peptide loading of MHC class II and MHC class I molecules. MHC class II molecules (MHCII) are assembled as dimers in the endoplasmic reticulum (ER) with the help of the specialized chaperone invariant chain (Ii), which, in addition, occupies the peptide-binding groove (upper panel). Three of these MHC class II/Ii complexes together form a nonameric complex that is transported to the MHC class II–containing compartments (MIIC). Here, the invariant chain is degraded by cathepsins and proteases until only the part occupying the peptide-binding groove, which is called CLIP, is left. In these compartments, MHCII also encounters antigenic peptide fragments derived from proteins degraded in the endocytic track. CLIP is then exchanged for one of these fragments with help of the chaperone HLA-DM, and the peptide-loaded MHCII is transported to the plasma membrane for presentation to the immune system. Peptide loading of MHC class I molecules follows a different route (lower panel). After assembly in

MHC Class II Molecules in Cancer

133

class II complex in the ER and is assisted by various common chaperones, such as BiP and PDI (Bonnerot et al., 1994; Cotner and Pious, 1995; Nijenhuis and Neefjes, 1994; Schaiff et al., 1992). Rapidly after this assembly step, the first dedicated chaperone Ii is co-assembled (Cresswell, 1994; Marks et al., 1990; Roche et al., 1991). Ii is usually expressed in molar excess over MHC class II  heterodimers and is retained in the ER. Whereas Ii supports exit of MHC class II  heterodimers from the ER, the reverse is also the case. MHC class II  heterodimers are required for release of Ii from the ER (Bikoff et al., 1993; Marks et al., 1990; Viville et al., 1993). Ii is not as invariant as the name suggests since multiple splice variants exist that are usually co-expressed. The p31/p33, p35, p41, and p43 forms of Ii have been described in humans, whereas mice contain the p31 and p41 form (Fineschi et al., 1995; O’Sullivan et al., 1987; Strubin et al., 1986). Ii is a type II transmembrane glycoprotein containing a short amino-terminal cytoplasmic domain, a single transmembrane domain, a domain that occupies the peptide-binding groove of MHC class II (called Class-II-associated Ii peptide, or CLIP), and a carboxyl-terminal trimerization motif (Fig. 2). Ii forms trimers that interact with dimers of MHC class II in the ER (Cresswell, 1994; Marks et al., 1990; Roche et al., 1991) (Fig. 1). The interaction of Ii with MHC class II is necessary for proper folding and supports transport of MHC class II molecules (Anderson and Miller, 1992; Cresswell, 1996; Marks et al., 1995a; Viville et al., 1993) from the ER to endosomal structures (Bikoff et al., 1993). Ii, however, not only functions as a mediator in transport. Early in assembly, the CLIP segment of Ii enters the MHC class II peptide-binding groove. This acts as a pseudopeptide, preventing premature loading of MHC class II with peptides that have entered the ER for binding to MHC class I molecules (Busch et al., 1996). In addition, peptide (or pseudopeptide) binding is required to pass the ‘‘ER quality control system’’ for transport to the endocytic pathway (Bikoff et al., 1993; Viville et al., 1993), a situation strongly resembling that of MHC class I molecules, which also require peptide binding for transport out of the ER (Schumacher et al., 1990; Townsend et al., 1989). After ER exit, the vast majority of MHC class II/Ii complexes enters the endocytic pathway (Neefjes et al., 1990), although a small percentage is transported directly to the plasma membrane via the secretory pathway

the ER and with the subsequent help of the chaperones calnexin, calreticulin, and ERp57, the MHC class I molecules dock onto the ER-resident peptide transporter TAP. This process is facilitated by the specialized chaperone tapasin. TAP pumps antigenic peptides from cellular origin that are produced in the cytosol into the ER lumen. These can then bind the ER-retained MHC class I molecules. Peptide binding stabilizes the recipient molecules and allows their transport to the plasma membrane.

134

Marije Marsman et al.

Fig. 2 Structural overview of the p31 isoform of the human invariant chain. The invariant chain is a type II transmembrane glycoprotein containing a short amino-terminal cytoplasmic domain, a single transmembrane domain (TM), a domain that occupies the peptide-binding groove of MHC class II (called class II–associated Ii peptide, or CLIP), and a carboxyl-terminal trimerization motif. The cytoplasmic domain contains two di-leucine motifs that are essential for sorting and intracellular trafficking. A destruction box motif (RXXL) near the transmembrane region appears to be involved in nuclear NF-B signaling. Also depicted is the p41 isoform that harbors an additional 64 amino acid Thyroglobulin domain (TGD) known to regulate cathepsin L activity.

(Lamb et al., 1991), followed by rapid internalization and sorting into the endocytic pathway (Brachet et al., 1999; Roche et al., 1993). The cytoplasmic domain of Ii contains two di-leucine motifs (Fig. 2), which are necessary for sorting of MHC class II molecules to the MIIC and also for internalization from the plasma membrane (Bakke and Dobberstein, 1990; Lotteau et al., 1990; Nijenhuis et al., 1994; Odorizzi et al., 1994; Pieters et al., 1993; Pond et al., 1995). There is some debate about how the MHC class II complexes traffic to the MIIC. Some reports describe trafficking via early endosomes to MIIC (Brachet et al., 1999; Pond and Watts, 1999), but most show direct targeting from the trans-Golgi network to the MIIC (Benaroch et al., 1995; Davidson, 1999; Neefjes et al., 1990; Peters et al., 1991). In all cases, the MHC class II complexes are targeted to the pre-lysosomal MIIC compartment.

C. Processing of the Invariant Chain Once inside the endocytic pathway, Ii is degraded by various proteases in a processive manner. The Ii degradation is essential for transport of antigenloaded MHC class II molecules from the MIIC to the plasma membrane. A 22 kD fragment (P22 LIP, for leupeptin-induced protein), a 10 kD fragment (P10 SLIP, for small LIP), and as has been described, the 2.5 kD fragment CLIP have been defined as Ii degradation intermediates associated to MHC class II molecules (Riese and Chapman, 2000). Although cathepsin B and D were originally claimed to be responsible for degradation of Ii, this

MHC Class II Molecules in Cancer

135

concept was abandoned when it was observed that Ii degradation proceeded normally in mice lacking these proteases (Deussing et al., 1998; Villadangos et al., 1997). Instead, Ii degradation involves more specific proteases, such as cathepsin S and L. Inhibition of cathepsin S impaired MHC class II antigen presentation and Ii degradation (Riese et al., 1996, 1998). A marked tissue specific expression profile is observed for the cathepsins. In macrophages, cathepsin S is up-regulated and cathepsin L is down-regulated upon interferon stimulation (Beers et al., 2003). Where peripheral antigenpresenting cells contain relatively high cathepsin S levels (Beers et al., 2003; Honey et al., 2001), cathepsin L appears to be crucial for Ii degradation in cortical tissue endothelial cells (Honey et al., 2002; Nakagawa and Rudensky, 1999). Thus, cathepsin S and L are both important with partially overlapping activities for Ii degradation, although other (currently unknown) proteases will be involved in other steps of this degradation. Further studies found cathepsin L tightly bound to the p41 form of invariant chain (Ogrinc et al., 1993). By interacting with the thyroglobulin domain (TGD) of the invariant chain (Fig. 2), proteolytic activity of cathepsin L is inhibited (Guncar et al., 1999). On the other hand, cathepsin L is protected from premature destruction by binding to the p41 isoform (Lennon-Dumenil et al., 2001). Collectively, this may result in regulation of antigen processing and loading of MHC class II and could explain the enhanced antigen presentation observed for the p41 form (Peterson and Miller, 1992), but this has to be further elucidated. In conclusion, several proteases are required for one simple but crucial act, the removal of a transporting chaperone 1 to 3 hours after assembly of the MHC class II/Ii nonamer. This is critical for the exchange of remaining CLIP fragments for antigenic peptides and for the transport of MHC class II to the plasma membrane.

D. The Invariant Chain, More than Just an MHC Class II Chaperone Invariant chain does not merely function as an MHC class II chaperone preventing peptide loading in the ER, stimulating exit from the ER, and modulating antigenic peptide loading, but it may have additional functions as well. The development from immature to mature B cells is impaired in Iideficient mice. These B cells arrest in an immature stage with low IgD and CD23 levels (Kenty and Bikoff, 1999; Shachar and Flavell, 1996). The N-terminal cytoplasmic domain of Ii is required for B-cell maturation, since expression of only this domain suffices to stimulate B-cell maturation (Matza et al., 2002b). The Ii cytosolic domain diffuses from MIIC into the nucleus, where it is supposed to activate NF-B signaling which

136

Marije Marsman et al.

then results in B-cell maturation (Matza et al., 2002a). A destruction box RXXL motif (Fig. 2) in the released cytosolic domain of Ii appears to be important in down-regulation of the signaling cascade, by inducing degradation of the signaling peptide. Mutations in this domain block degradation of the cytosolic Ii fragment, inducing NFB activation. This process resembles that found for various integral membrane proteins where a short-lived soluble fragment is released after proteolysis that migrates to the nucleus to induce transcription, a process named RIP for Regulated Intramembrane Proteolysis (Matza et al., 2003).

E. HLA-DM, the Editor for Antigenic Peptide Loading of MHC Class II Molecules The need for additional chaperones in the loading of MHC class II molecules with peptides was not obvious. Only after a thorough analysis of B cell lines with deletions in the MHC locus at chromosome 6 did it become apparent that additional proteins were required for successful peptide loading of MHC class II molecules. The genes were subsequently identified as HLA-DMA and HLA-DMB that assemble into HLA-DM (H2-M in mice), a nonpolymorphic type I membrane protein with high similarity in sequence and structure to MHC class II molecules (Cho et al., 1991; Fling et al., 1994; Kelly et al., 1991; Mellins et al., 1991; Morris et al., 1994). HLA-DM and MHC class II genes probably arose by gene duplications of a shared ancestor gene. After assembly in the ER, HLA-DM is transported to MIIC through a tyrosine-based targeting signal in the cytoplasmic tail of HLADMB (Copier et al., 1996; Marks et al., 1995b). Although HLA-DM accumulates in MIIC, it probably recycles via the plasma membrane by efficient re-internalization mediated by the tyrosine-motif in the HLA-DMB tail (Arndt et al., 2000; van Lith et al., 2001). HLA-DM deficient cells and mice express surface MHC class II molecules loaded with the Ii-degradation fragment CLIP instead of antigenic peptides (Ceman et al., 1994; Fling et al., 1994; Fung-Leung et al., 1996; Martin et al., 1996; Miyazaki et al., 1996; Morris et al., 1994; Riberdy et al., 1992). Thus, although some spontaneous exchange of CLIP for antigenic fragments can occur in the acidic MIIC (Avva and Cresswell, 1994), efficient exchange requires HLA-DM to release CLIP and low-affinity peptides, while allowing high-affinity peptides to remain associated (Denzin and Cresswell, 1995; Kropshofer et al., 1996; Lovitch et al., 2003; Sherman et al., 1995; Sloan et al., 1995; van Ham et al., 1996; Weber et al., 1996). Further in vitro experiments showed that the interaction between HLA-DM and MHC class II molecules and the ‘‘activity’’ of HLA-DM were facilitated by acidic pH, as found in the MIIC (Kropshofer et al., 1997; Sanderson

MHC Class II Molecules in Cancer

137

et al., 1996; Ullrich et al., 1997; Vogt et al., 1997). HLA-DM appears to stabilize MHC class II molecules devoid of peptide to allow binding of high-affinity peptides and, at the same time, as a true chaperone, prevents the aggregation of ‘‘empty’’ MHC class II molecules (Denzin et al., 1996; Kropshofer et al., 1997). The structure of HLA-DM reveals a molecule with a high structural identity to MHC class II molecules (Fig. 3). One major difference is the absence of an MHC class II peptide-binding groove in HLA-DM, which renders it unable to bind peptides (and the invariant chain) (Fremont et al., 1998; Mosyak et al., 1998). A co-crystal of HLA-DM and MHC class II molecules has not been generated, but mutational studies have revealed areas in the top part (peptide-binding groove) of MHC class II and HLADM as interacting segments (Pashine et al., 2003; Stratikos et al., 2002). MHC class II molecules are highly polymorphic and different MHC class II alleles present different fragments from the same antigen. Still, Ii as well as HLA-DM are nonpolymorphic and interact with all polymorphic MHC class II alleles (Robinson et al., 2001). As a consequence, the binding affinity of CLIP for MHC class II molecules differs for the different MHC class II haplotypes, possibly resulting in a different dependency on HLA-DM

Fig. 3 The structure of HLA-DM shows high structural identity to MHC class II molecules. The structures of a MHC class II molecule (in this case, HLA-DR3) and HLA-DM are projected above the membrane. The transmembrane regions of both molecules are undefined and not depicted. Note the absence of an open peptide-binding groove in HLA-DM, which prohibits peptide binding.

138

Marije Marsman et al.

(Doebele et al., 2003). Whether this results in differences in peptide loading of MHC class II remains unclear.

F. HLA-DO, the Chaperone of Chaperones As described, efficient loading of MHC class II with specific antigenic peptides is tightly regulated by the chaperones Ii and HLA-DM. More recently, attention has shifted to another MHC class II look-alike, also encoded in the MHC locus. Two genes encoding for HLA-DOA and HLA-DOB were identified that assemble into a HLA-DO (or H2-O the murine homologue of HLA-DO) heterodimer (Tonnelle et al., 1985; Young and Trowsdale, 1990). Like HLA-DM, HLA-DO has a very high sequence identity to HLA-DR molecules, which suggests that they arose from recent gene duplication (van Lith et al., 2002). HLA-DO is also a nonpolymorphic heterodimer with lysosomal targeting sequences located in the cytoplasmic tail of HLA-DOB (van Lith et al., 2001). Unlike HLA-DMB, HLADOB contains two putative targeting signals, a di-leucine motif and a tyrosine-based motif (van Lith et al., 2001). Whereas HLA-DM is always co-expressed with MHC class II molecules in APCs, HLA-DO is only expressed on a subset of thymic medullary epithelium and in immature B cells (Douek and Altmann, 1997; Karlsson et al., 1991; Tonnelle et al., 1985). Moreover, both HLA-DO and HLA-DM are rapidly down-regulated upon activation of B cells (Roucard et al., 2001). A stable interaction is formed between HLA-DO and HLA-DM, and targets HLA-DO to the MIIC (Liljedahl et al., 1996). Upon deletion of its targeting signals, HLA-DO is still targeted to the MIIC via HLA-DM (van Lith et al., 2001). Subsequently, the HLA-DM/DO heterotetramer recycles between MIIC and the plasma membrane, although it accumulates in MIIC (van Lith et al., 2001). HLA-DO acts as a negative regulator of HLA-DM since MHC class II molecules loaded with the CLIP fragment appeared at the plasma membrane in response to ectopic expression of HLA-DO or overexpression of H2-O in transgenic mice (Brocke et al., 2003; Denzin et al., 1997; van Ham et al., 1997, 2000). Mice deficient for H2-O have only mild phenotypes, including an increase in antibody titer in plasma, suggesting that B cell proliferation is less tightly controlled (Liljedahl et al., 1998; Perraudeau et al., 2000). Further studies revealed that HLA-DO altered the pH sensitivity of HLA-DM in supporting peptide loading of MHC class II molecules in vitro. Apparently, HLA-DO acts as a pH sensor restricting HLA-DM activity to more acidic (late endosomal) structures (van Ham et al., 2000). The function of HLA-DO is not fully understood, the assumption is that the activity of B cells should be tightly controlled, implying that antigenic

MHC Class II Molecules in Cancer

139

peptide loading of MHC class II should primarily occur with antigens recognized by surface immunoglobulins. The activity of HLA-DO skews peptide loading to the late endosomal structures where B cell receptormediated antigens are processed. HLA-DO may thus play a critical role in controlling B cell activation by regulating the activity of HLA-DM.

G. From the MIIC to the Plasma Membrane Obviously, proper loading with antigenic peptides in the MIIC is not sufficient for successful MHC class II antigen presentation. Therefore, MHC class II molecules first have to be transported to the plasma membrane. Transport of MHC class II has been studied using combinations of electron microscopy and real-time imaging of GFP-tagged MHC class II molecules. MIIC move along microtubules from the Golgi area around the microtubule organizing center (MTOC) toward the plasma membrane (Wubbolts et al., 1996). This transport is similar to lysosomal transport and occurs in a bi-directional manner and in a stop-and-go fashion, mediated by the alternate activities of the dynein/dynactin and kinesin motor proteins (Wubbolts et al., 1999). How the motor protein activities are controlled is largely unclear, but the dynein/dynactin-mediated transport toward the minus-end involves at least the activity of the small GTPase Rab7 and its effector protein RILP (Jordens et al., 2001). Finally, the MIIC reaches the end of the microtubule at the cortical actin cytoskeleton just underneath the plasma membrane. How the last step occurs is unclear, but ultimately the MIIC fuses with the plasma membrane, as shown by electron microscopy (Raposo et al., 1996; Wubbolts et al., 1996). At the plasma membrane, part of the intracellular content (the internal vesicles in a multivesicular body) is secreted in the form of so-called exosomes (Raposo et al., 1996; Zitvogel et al., 1998). This is probably a small fraction because otherwise many internally residing proteins like HLA-DM and the tetraspans would be depleted from cells within 1 to 2 hours (which is the turnover time of MIIC in most cells (Neefjes et al., 1990)). The majority of the internal structures of the MIIC probably fuse back to the plasma membrane, followed by rapid internalization of the late endocytic MIIC proteins via their internalization signals (Arndt et al., 2000; van Lith et al., 2001). Subsequently, these proteins are transported back to the MIIC through the endocytic pathway. Since only a fraction of the MHC class II molecules can be internalized, MHC class II accumulates at the plasma membrane (Pinet et al., 1995). Interestingly, surface MHC class II molecules do not behave identically in all cell types. In fact, the half-life of MHC class II molecules differs considerably among different cell types. It is relatively long in melanoma

140

Marije Marsman et al.

cells and B cells compared to primary monocytes and dendritic cells. The half-life in dendritic cells increases (up to 100 h) after activation, which is possibly due to an increase in stable MHC class II-peptide complexes (Cella et al., 1997; Pierre et al., 1997). A fraction of the MHC class II molecules can be internalized (Pinet et al., 1995; Reid and Watts, 1990) and recycled back to the plasma membrane. In monocytes, the reappearance of MHC class II at the plasma membrane is controlled by interleukin 10 (Koppelman et al., 1997). Treatment with IL-10 results in a strong reduction of cell surface MHC class II molecules, possibly by affecting the Rab7 pathway, which, in turn, controls dynein motor-mediated MIIC transport (our unpublished results). This is the first example of regulation of MHC class II responses by manipulation of the last step in intracellular transport of MHC class II molecules to the cell surface. An alternative route of MHC class II molecules from the MIIC to the plasma membrane has been proposed for dendritic cells (Boes et al., 2002; Chow et al., 2002; Kleijmeer et al., 2001). Upon activation of DCs, the MIIC appears to alter its morphology, resulting in the formation of long tubular structures extending into the periphery. Live imaging of these cells revealed that these class II-positive structures, similar to the conventional MIIC, move in a microtubule-dependent manner (Boes et al., 2002). Probably in response to the gross alteration of the cytoskeleton of activated dendritic cells, the MHC class II molecules move more in the direction of the contact site with a specific T cell. Surprisingly, no accumulation of GFP-tagged MHC class II molecules was observed in the ‘‘immunological synapse’’ between the DC and the T cell (Boes et al., 2002). The exact function of this directed transport of MIIC-derived tubules after DC activation still has to be revealed. Thus, two modes of transport of MHC class II molecules from MIIC to the plasma membrane have been reported: direct transport and fusion of MIIC with the plasma membrane and the formation of tubular structures. In both cases, transport requires motor-based microtubule transport, likely to be mediated by dynein/dynactin and kinesin motor proteins, with the small GTPase Rab7 as one of the controllers of this transport step.

III. INTERFERING WITH ANTIGEN PRESENTATION BY MHC CLASS II MOLECULES A. Promoting Antigen Presentation The pathway of antigen presentation by MHC class II molecules which has been outlined shows that it requires a ‘‘multi-enzyme’’ process involving various chaperones, acidic pH, and proteases at different stages of

MHC Class II Molecules in Cancer

141

biosynthesis. Obviously, affecting one or more of these enzymes can positively or negatively influence antigen presentation. For instance, MHC class II antigen presentation can be improved when antigens are more efficiently acquired and targeted to MIIC. Macrophages and monocytes, in contrast to B cells, are able to internalize large volumes. This implies that many antigenic fragments have to compete for access to MHC class II molecules in the MIIC. More selective uptake of antigen-using surface immunoglobulins in B cells (Lanzavecchia, 1985; Rock et al., 1984; Siemasko and Clark, 2001), Fc-receptors on macrophages and DCs (Fanger et al., 1997; Guyre et al., 1997), or mannose-receptors on DCs (Engering et al., 1997) will strongly improve antigen presentation by MHC class II molecules (Lanzavecchia, 1996). Cells may also alter the conditions for antigen presentation in MIIC. Best studied are DCs that acidify the MIIC upon activation (Trombetta et al., 2003), and B cells that down-regulate HLA-DO upon activation (Roucard et al., 2001). In both cases, antigen presentation by MHC class II molecules is more efficient.

B. Inhibiting Antigen Presentation If activation of MHC class II antigen presentation is an option, the reverse is almost certainly true as well. For instance, Th2 cell activity controlling immune responses can inhibit class II antigen presentation. Another means of inhibiting MHC class II responses is by interfering with endosomal proteases, as was first shown by using leupeptin. Leupeptin is a protease inhibitor that inhibits complete degradation of Ii (Cresswell et al., 1990; Neefjes and Ploegh, 1992). Since Ii degradation is a prerequisite for transport of MHC class II molecules to the cell surface, these inhibitors are negative regulators of class II presentation (Amigorena et al., 1995; Brachet et al., 1997; Neefjes and Ploegh, 1992). Naturally occurring protease inhibitors exist as well. Cystatin is a reversible inhibitor of cysteine proteases like the cathepsins. Cystatin family members are expressed in a tissuespecific manner and can modulate cathepsin activities and thereby inhibit antigen presentation by MHC class II molecules (Pierre and Mellman, 1998), although this is still somewhat controversial. Pathogens are known to use a similar system to inhibit MHC class II presentation, as has been reported for two filarial nematodes, Onchocerca volvulus and Acanthocheilonema viteae (Hartmann et al., 1997; Schonemeyer et al., 2001). Both nematodes produce cystatin-like molecules that have an immunosuppressive activity by inhibiting cathepsin S and L. Moreover, Bm-CPI-2 is a cystatin homologue secreted by the parasite Brugia malayi that also interferes with MHC class II processing by inhibiting multiple cysteine proteases (Manoury et al., 2001).

142

Marije Marsman et al.

Intracellularly growing bacteria use different strategies to prevent presentation by MHC class II molecules. These pathogens include Salmonella Typhimurium, Mycobacterium Tuberculosis, and Mycobacterium Leprae, which usually reside in the endo/phagosomal pathway. Some of these pathogens are able to modulate the endo/lysosomal compartments. Salmonella injects several effector proteins into the host cytosol, which prevents fusion of the phagosome with mature lysosomes (Holden, 2002). It has been found that one of the bacterial effectors is a PI3P phosphatase (Hernandez et al., 2004) that has been implicated in inhibition of the formation of internal structures within the MIIC (Fernandez-Borja et al., 1999) and antigen presentation by MHC class II molecules (Song et al., 1997). Thus, Salmonella might interfere with MHC class II presentation in multiple ways. It escapes the degradation in the mature lysosomes, thereby limiting the amount of Salmonella-derived antigens. Secondly, by preventing the formation of internal membranes, Salmonella might reduce the peptideloading efficiency by preventing efficient interactions between MHC class II and HLA-DM (Zwart, W., in preparation). Nature has thus developed a complicated system to allow presentation of antigenic fragments generated in the endosomal track. It has also developed multiple ways to manipulate this process, exploited not only by pathogens, but also by tumors, as will be discussed in the following text.

IV. MHC CLASS II MOLECULES IN ONCOGENESIS A. Immune System Involved in Tumor Surveillance Various tumors down-regulate MHC class I expression by inactivating transcription of the MHC locus (Garcia-Lora et al., 2003). Specific MHC class I alleles can be down-regulated in rarer cases as well (Browning et al., 1996; Cabrera et al., 2003; Masucci et al., 1989). The observation that down-regulation of MHC class I expression correlates with an aggressive or more advanced tumor phenotype (Bubenik, 2003) suggests that the immune system is involved in controlling tumor outgrowth. Still, the exact role of the immune system in tumor surveillance is not fully understood. Transplantation patients that receive immunosuppressive drugs rarely develop tumors other than leukemia after use of cyclosporine for 10 to 12 years. Also, patients with genetic defects in components of the MHC system (CIITA, TAP) do not show a higher tumor incidence but merely a higher susceptibility to bacterial and viral infections (Bontron et al., 1997; Schultz et al., 2003; Steimle et al., 1993).

MHC Class II Molecules in Cancer

143

Whereas MHC class I and II molecules may be not highly important in active immune surveillance to various tumors, there are a number of notorious exceptions. As has been mentioned, leukemias appear with high incidence in immunosuppressed patients (Kinlen, 1992, 2000). Furthermore, melanomas and renal cell carcinomas are known to spontaneously regress in some patients and it is thought that this is due to tumor recognition by the immune system (Balch, 1992; Gromet et al., 1978). Especially for melanoma, many tumor-specific CTLs have been isolated, often recognizing melanoma proteins presented in the context of MHC class I, or activated with the help of MHC class II molecules that present tumorspecific antigens (Jager et al., 2000; Kirkin et al., 1999; Renkvist et al., 2001; Saleh et al., 2001, 2003; Slager et al., 2003; Tatsumi et al., 2002; van der Bruggen et al., 1991; Zeng et al., 2001). It is, therefore, not surprising that most attempts to use immunotherapy for tumor eradication concentrate not only on the virally induced tumors, but also on leukemia (using minor histocompatibility antigens), melanoma (using dendritic cells pulsed with melanoma extracts or long antigenic peptides), and renal cell carcinomas, with some impressive successes (Dudley et al., 2002; Mutis and Goulmy, 2002; Spierings et al., 2004; Thurner et al., 1999). Of note is that usually only MHC class I–restricted responses are considered in these therapies.

B. MHC Class II Expression and Tumor Development Why are MHC class II–restricted responses usually not considered in tumor development? One reason could be that most tumors do not express MHC class II molecules. Exceptions are—obviously—B cell leukemias like chronic lymphocytic leukemia, Burkitt lymphoma, EBV-induced B cell nonHodgkin lymphoma, follicular lymphoma, and Kahler’s disease (Guy et al., 1986). In addition, melanoma can express MHC class II molecules (Zeng et al., 2000), although often in a rather heterologous manner, and Glioma type 1 tumors also constitutively express MHC class II (Takamura et al., 2004). Furthermore, interferon and possibly other factors can induce MHC class II expression on Glioma type 2 and many other tumors, including cervix tumors (Santin et al., 1998) and bladder cancer (Champelovier et al., 2003). In most cases, not only MHC class II molecules are expressed but also the accessory proteins required for efficient transport (li) and peptide loading (HLA-DM) (Siegrist et al., 1995). Various studies report on a correlation between expression of MHC class II molecules and prognosis (Concha et al., 1991; Ostmeier et al., 1999; Passlick et al., 1994; Rimsza et al., 2004), but whether this is the result of an anti-tumor immune response or simply because MHC class II expression is a

144

Marije Marsman et al.

marker for another differentiation state of the tumor (and thus different growth and invading properties) remains unclear. MHC class II–restricted presentation of exogenous antigens can easily be observed ex vivo by EBV-transformed B cells (Nijenhuis et al., 1994) and melanoma cells (van Ham et al., 2000), implying that the MHC class II system ‘‘works’’ correctly and efficiently in these tumors. In addition, MHC class II molecules can present ‘‘tumor-specific antigens’’ like Epstein Barr viral (EBV) antigen EBNA-1 (Voo et al., 2002) and melanoma-specific antigens like tyrosinase and gp 100 (Parkhurst et al., 2003; Topalian et al., 1994). Still, although class II presentation is occurring, an efficient host response is obviously lacking when a tumor appears and selective outgrowth of tumor-specific CD4þ T cells has not been reported in patients, even though these T cells can be expanded in vitro. Apparently, tumor factors prevent expansion of these cells. Interestingly, some tumors that express MHC class II also express inhibitors for the MHC class II antigen presentation process. Most notably is the inhibitor for HLA-DM, HLA-DO, in B cell leukemia. Whether this again reflects a differentiation difference (HLA-DO is best expressed in immature B cells) or an active attempt to inhibit antigen presentation by MHC class II is unclear. Other tumors secrete inhibitory cytokines like IL-10 to suppress MHC class II and other responses, which is particularly clear for melanomas (Dummer et al., 1996) and EBV-induced B cell tumors (Benjamin et al., 1992). The EBV genome encodes a homologous protein for IL-10 (BCRF1) (Moore et al., 1991; Vieira et al., 1991) that may also inhibit T cell responses, although this is not fully established. Other tumors, like neuroblastoma, actively down-regulate expression of MHC class II by silencing the CTIIA promoter (Croce et al., 2003). Patients selectively lacking expression of MHC class II molecules (and not MHC class I) exist. These bare lymphocyte syndrome patients usually have genetic defects in the transcription machinery regulating expression of MHC class II molecules and its accessory proteins HLA-DM and the invariant chain (Mach et al., 1994). Defects have been reported for the transcription factor CIITA (Steimle et al., 1993). Although these patients are prone to many infections, no increased rate of tumor formation has been reported. Similarly, no increased tumor incidence is observed in mice deficient for MHC class II molecules, DM, or li. These observations suggest that class II plays no active role in the tumor formation. However, this does not exclude a relevant role of MHC class II molecules in anti-tumor responses. First, MHC class II molecules are essential for the induction of proper CTL responses (Ossendorp et al., 1998). These are most likely not the MHC class II molecules expressed on tumors but merely MHC class II molecules expressed on professional antigen-presenting cells like DCs. These cells probably internalize apoptotic bodies or necrotic

MHC Class II Molecules in Cancer

145

debris from tumor cells and present fragments to the CD4þ T cells, which then stimulate cytotoxic T cell proliferation (Ossendorp et al., 1998). Stateof-the-art tumor vaccination strategies therefore include—besides antigens for MHC class I molecules—antigens for presentation by MHC class II molecules. These antigens can be targeted into the MHC class II pathway using Fc-receptor-mediated uptake (You et al., 2001), via the mannose receptor (van Bergen et al., 1999), or (although this pathway is more undefined) used directly in the form of exosomes (Zitvogel et al., 1998). The stimulation of both the MHC class I and MHC class II pathways may ensure a better stimulation of tumor-specific cytotoxic T cells, and thus a better anti-tumor response.

C. Toward MHC Class II–Restricted Tumor Immunotherapy Like MHC class I molecules, MHC class II molecules also require specific antigens as targets for immunotherapy. Three types of targets are available: 1. Viral antigens expressed in virally induced tumors. An obvious candidate is EBV in B–cell non-Hodgkins lymphoma. Especially the EBV coat proteins (Cohen et al., 1984) and the EBV protein EBNA-1 (Munz et al., 2000) should reveal good fragments for MHC class II molecules. Whether v-IL 10 expressed by EBV prevents an efficient response is unclear (Sairenji et al., 1998). If so, a combination of vaccination with EBV protein or peptide antigen with simultaneous neutralization of v-IL 10 could mount an efficient MHC class II–restricted anti-tumor response. 2. Tumor-specific proteins. These can be mutated proteins or proteins expressed in a strong tissue-specific manner. Examples of the first are mutated growth factor receptors that could become constitutively activated and lead to constitutive cell growth and tumorigenesis, such as mutated EGF receptor and the ERB2 (neu) receptor (Hynes and Stern, 1994; Salomon et al., 1995). These receptors are degraded in the endosomal pathway generating peptides for MHC class II molecules. However, no CTLs specific for these mutated antigens have been identified yet. Examples of normal antigens expressed in a tissue-specific manner are tyrosinase and gp 100, two proteins expressed in cells of the melanocytic lineage (Brouwenstijn et al., 1997). These antigens are currently tested as antigens for MHC class I–restricted tumor therapy against melanoma (Dunbar et al., 1999; Meidenbauer et al., 2003; Sutmuller et al., 2000; Wang et al., 2001). Since these antigens are mostly degraded in the lysosomal pathway, strong MHC class II responses could be expected. Indeed, MHC class II–restricted T cells against some of these antigens can be isolated and may be used to

eradicate MHC class II expressing melanoma (Topalian et al., 1994, 1996). Obviously, tissue-specific antigens are also present in B cells tumors. These include CD20, a cell surface protein that is currently targeted in antibodybased immune therapy against B cell non-Hodgkin’s lymphoma (Milenic et al., 2004). In principle, fragments of CD20 presented in the context of MHC class II molecules could be used for cancer therapy as well. 3. Minor histocompatibility antigens. Minor histocompatibility antigens

MHC Class II Molecules in Cancer

147

ACKNOWLEDGMENTS This work was supported by grants from KWF Kankerbestrijding and the Netherlands Organization for Scientific Research (NWO).

REFERENCES Amigorena, S., Webster, P., Drake, J., Newcomb, J., Cresswell, P., and Mellman, I. (1995). Invariant chain cleavage and peptide loading in major histocompatibility complex class II vesicles. J. Exp. Med. 181, 1729–1741. Anderson, M. S., and Miller, J. (1992). Invariant chain can function as a chaperone protein for class II major histocompatibility complex molecules. Proc. Natl. Acad. Sci. USA 89, 2282–2286. Arndt, S. O., Vogt, A. B., Markovic-Plese, S., Martin, R., Moldenhauer, G., Wolpl, A., Sun, Y., Schadendorf, D., Hammerling, G. J., and Kropshofer, H. (2000). Functional HLA-DM on the surface of B cells and immature dendritic cells. EMBO J. 19, 1241–1251. Avva, R. R., and Cresswell, P. (1994). In vivo and in vitro formation and dissociation of HLADR complexes with invariant chain-derived peptides. Immunity 1, 763–774. Bakke, O., and Dobberstein, B. (1990). MHC class II-associated invariant chain contains a sorting signal for endosomal compartments. Cell 63, 707–716. Balch, C. M. (1992). Cutaneous melanoma: Prognosis and treatment results worldwide. Semin. Surg. Oncol. 8, 400–414. Beers, C., Honey, K., Fink, S., Forbush, K., and Rudensky, A. (2003). Differential regulation of cathepsin S and cathepsin L in interferon gamma-treated macrophages. J. Exp. Med. 197, 169–179. Benaroch, P., Yilla, M., Raposo, G., Ito, K., Miwa, K., Geuze, H. J., and Ploegh, H. L. (1995). How MHC class II molecules reach the endocytic pathway. EMBO J. 14, 37–49. Benjamin, D., Knobloch, T. J., and Dayton, M. A. (1992). Human B-cell interleukin-10: B-cell lines derived from patients with acquired immunodeficiency syndrome and Burkitt’s lymphoma constitutively secrete large quantities of interleukin-10. Blood 80, 1289–1298. Bikoff, E. K., Huang, L. Y., Episkopou, V., van Meerwijk, J., Germain, R. N., and Robertson, E. J. (1993). Defective major histocompatibility complex class II assembly, transport, peptide acquisition, and CD4þ T cell selection in mice lacking invariant chain expression. J. Exp. Med. 177, 1699–1712. Boes, M., Cerny, J., Massol, R., Op den Brouw, M., Kirchhausen, T., Chen, J., and Ploegh, H. L. (2002). T-cell engagement of dendritic cells rapidly rearranges MHC class II transport. Nature 418, 983–988. Bonnerot, C., Marks, M. S., Cosson, P., Robertson, E. J., Bikoff, E. K., Germain, R. N., and Bonifacino, J. S. (1994). Association with BiP and aggregation of class II MHC molecules synthesized in the absence of invariant chain. EMBO J. 13, 934–944. Bontron, S., Steimle, V., Ucla, C., Eibl, M. M., and Mach, B. (1997). Two novel mutations in the MHC class II transactivator CIITA in a second patient from MHC class II deficiency complementation group A. Hum. Genet. 99, 541–546. Brachet, V., Pehau-Arnaudet, G., Desaymard, C., Raposo, G., and Amigorena, S. (1999). Early endosomes are required for major histocompatibility complex class II transport to peptideloading compartments. Mol. Biol. Cell 10, 2891–2904.

148

Marije Marsman et al.

Brachet, V., Raposo, G., Amigorena, S., and Mellman, I. (1997). Ii chain controls the transport of major histocompatibility complex class II molecules to and from lysosomes. J. Cell Biol. 137, 51–65. Brocke, P., Armandola, E., Garbi, N., and Hammerling, G. J. (2003). Downmodulation of antigen presentation by H2-O in B cell lines and primary B lymphocytes. Eur. J. Immunol. 33, 411–421. Brouwenstijn, N., Slager, E. H., Bakker, A. B., Schreurs, M. W., Van der Spek, C. W., Adema, G. J., Schrier, P. I., and Figdor, C. G. (1997). Transcription of the gene encoding melanomaassociated antigen gp 100 in tissues and cell lines other than those of the melanocytic lineage. Br. J. Cancer 76, 1562–1566. Browning, M., Petronzelli, F., Bicknell, D., Krausa, P., Rowan, A., Tonks, S., Murray, N., Bodmer, J., and Bodmer, W. (1996). Mechanisms of loss of HLA class I expression on colorectal tumor cells. Tissue Antigens 47, 364–371. Bubenik, J. (2003). Tumour MHC class I downregulation and immunotherapy (Review). Oncol. Rep. 10, 2005–2008. Busch, R., Cloutier, I., Sekaly, R. P., and Hammerling, G. J. (1996). Invariant chain protects class II histocompatibility antigens from binding intact polypeptides in the endoplasmic reticulum. EMBO J. 15, 418–428. Cabrera, T., Maleno, I., Lopez-Nevot, M. A., Redondo, M., Fernandez, M. A., Collado, A., and Garrido, F. (2003). High frequency of HLA-B44 allelic losses in human solid tumors. Hum. Immunol. 64, 941–950. Cella, M., Engering, A., Pinet, V., Pieters, J., and Lanzavecchia, A. (1997). Inflammatory stimuli induce accumulation of MHC class II complexes on dendritic cells. Nature 388, 782–787. Ceman, S., Petersen, J. W., Pinet, V., and DeMars, R. (1994). Gene required for normal MHC class II expression and function is localized to approximately 45 kb of DNA in the class II region of the MHC. J. Immunol. 152, 2865–2873. Champelovier, P., Simon, A., Garrel, C., Levacher, G., Praloran, V., and Seigneurin, D. (2003). Is interferon gamma one key of metastatic potential increase in human bladder carcinoma? Clin. Cancer. Res. 9, 4562–4569. Cho, S. G., Attaya, M., and Monaco, J. J. (1991). New class II-like genes in the murine MHC. Nature 353, 573–576. Chow, A., Toomre, D., Garrett, W., and Mellman, I. (2002). Dendritic cell maturation triggers retrograde MHC class II transport from lysosomes to the plasma membrane. Nature 418, 988–994. Cohen, J. H., Vischer, T. L., Carquin, J., Blanchard, F., von Fliedner, V. E., and Jeannet, M. (1984). HLA-DR antigens and the antibody response against Epstein-Barr virus. Tissue Antigens 23, 156–162. Concha, A., Esteban, F., Cabrera, T., Ruiz-Cabello, F., and Garrido, F. (1991). Tumor aggressiveness and MHC class I and II antigens in laryngeal and breast cancer. Semin. Cancer Biol. 2, 47–54. Copier, J., Kleijmeer, M. J., Ponnambalam, S., Oorschot, V., Potter, P., Trowsdale, J., and Kelly, A. (1996). Targeting signal and subcellular compartments involved in the intracellular trafficking of HLA-DMB. J. Immunol. 157, 1017–1027. Cotner, T., and Pious, D. (1995). HLA-DR beta chains enter into an aggregated complex containing GRP-78/BiP prior to their degradation by the pre-Golgi degradative pathway. J. Biol. Chem. 270, 2379–2386. Cresswell, P. (1994). Assembly, transport, and function of MHC class II molecules. Annu. Rev. Immunol. 12, 259–293. Cresswell, P. (1996). Invariant chain structure and MHC class II function. Cell 84, 505–507.

MHC Class II Molecules in Cancer

149

Cresswell, P. (2000). Intracellular surveillance: Controlling the assembly of MHC class I-peptide complexes. Traffic 1, 301–305. Cresswell, P., Blum, J. S., Davis, J. E., and Marks, M. S. (1990). Transport and expression of HLA class-II glycoproteins. Immunol. Res. 9, 190–199. Croce, M., De Ambrosis, A., Corrias, M. V., Pistoia, V., Occhino, M., Meazza, R., GironMichel, J., Azzarone, B., Accolla, R. S., and Ferrini, S. (2003). Different levels of control prevent interferon-gamma-inducible HLA-class II expression in human neuroblastoma cells. Oncogene 22, 7848–7857. Davidson, H. W. (1999). Direct transport of newly synthesized HLA-DR from the trans-Golgi network to major histocompatibility complex class II containing compartments (MIICS) demonstrated using a novel tyrosine-sulfated chimera. J. Biol. Chem. 274, 27315–27322. den Haan, J. M., Meadows, L. M., Wang, W., Pool, J., Blokland, E., Bishop, T. L., Reinhardus, C., Shabanowitz, J., Offringa, R., Hunt, D. F., Engelhard, V. H., and Goulmy, E. (1998). The minor histocompatibility antigen HA-1: A diallelic gene with a single amino acid polymorphism. Science 279, 1054–1057. Denzin, L. K., and Cresswell, P. (1995). HLA-DM induces CLIP dissociation from MHC class II alpha beta dimers and facilitates peptide loading. Cell 82, 155–165. Denzin, L. K., Hammond, C., and Cresswell, P. (1996). HLA-DM interactions with intermediates in HLA-DR maturation and a role for HLA-DM in stabilizing empty HLA-DR molecules. J. Exp. Med. 184, 2153–2165. Denzin, L. K., Sant’Angelo, D. B., Hammond, C., Surman, M. J., and Cresswell, P. (1997). Negative regulation by HLA-DO of MHC class II-restricted antigen processing. Science 278, 106–109. Deussing, J., Roth, W., Saftig, P., Peters, C., Ploegh, H. L., and Villadangos, J. A. (1998). Cathepsins B and D are dispensable for major histocompatibility complex class II-mediated antigen presentation. Proc. Natl. Acad. Sci. USA 95, 4516–4521. Doebele, R. C., Pashine, A., Liu, W., Zaller, D. M., Belmares, M., Busch, R., and Mellins, E. D. (2003). Point mutations in or near the antigen-binding groove of HLA-DR3 implicate class II-associated invariant chain peptide affinity as a constraint on MHC class II polymorphism. J. Immunol. 170, 4683–4692. Douek, D. C., and Altmann, D. M. (1997). HLA-DO is an intracellular class II molecule with distinctive thymic expression. Int. Immunol. 9, 355–364. Dudley, M. E., Wunderlich, J. R., Robbins, P. F., Yang, J. C., Hwu, P., Schwartzentruber, D. J., Topalian, S. L., Sherry, R., Restifo, N. P., Hubicki, A. M., Robinson, M. R., Raffeld, M., Duray, P., Seipp, C. A., Rogers-Freezer, L., Morton, K. E., Mavroukakis, S. A., White, D. E., and Rosenberg, S. A. (2002). Cancer regression and autoimmunity in patients after clonal repopulation with antitumor lymphocytes. Science 298, 850–854. Dummer, W., Bastian, B. C., Ernst, N., Schanzle, C., Schwaaf, A., and Brocker, E. B. (1996). Interleukin-10 production in malignant melanoma: Preferential detection of IL-10-secreting tumor cells in metastatic lesions. Int. J. Cancer 66, 607–610. Dunbar, P. R., Chen, J. L., Chao, D., Rust, N., Teisserenc, H., Ogg, G. S., Romero, P., Weynants, P., and Cerundolo, V. (1999). Cutting edge: Rapid cloning of tumor-specific CTL suitable for adoptive immunotherapy of melanoma. J. Immunol. 162, 6959–6962. Engering, A. J., Cella, M., Fluitsma, D., Brockhaus, M., Hoefsmit, E. C., Lanzavecchia, A., and Pieters, J. (1997). The mannose receptor functions as a high capacity and broad specificity antigen receptor in human dendritic cells. Eur. J. Immunol. 27, 2417–2425. Falkenburg, J. H., Marijt, W. A., Heemskerk, M. H., and Willemze, R. (2002). Minor histocompatibility antigens as targets of graft-versus-leukemia reactions. Curr. Opin. Hematol. 9, 497–502. Fanger, N. A., Voigtlaender, D., Liu, C., Swink, S., Wardwell, K., Fisher, J., Graziano, R. F., Pfefferkorn, L. C., and Guyre, P. M. (1997). Characterization of expression, cytokine

150

Marije Marsman et al.

regulation, and effector function of the high affinity IgG receptor Fc gamma RI (CD64) expressed on human blood dendritic cells. J. Immunol. 158, 3090–3098. Fernandez-Borja, M., Wubbolts, R., Calafat, J., Janssen, H., Divecha, N., Dusseljee, S., and Neefjes, J. (1999). Multivesicular body morphogenesis requires phosphatidyl-inositol 3-kinase activity. Curr. Biol. 9, 55–58. Fineschi, B., Arneson, L. S., Naujokas, M. F., and Miller, J. (1995). Proteolysis of major histocompatibility complex class II-associated invariant chain is regulated by the alternatively spliced gene product, p41. Proc. Natl. Acad. Sci. USA 92, 10257–10261. Fling, S. P., Arp, B., and Pious, D. (1994). HLA-DMA and -DMB genes are both required for MHC class II/peptide complex formation in antigen-presenting cells. Nature 368, 554–558. Fremont, D. H., Crawford, F., Marrack, P., Hendrickson, W. A., and Kappler, J. (1998). Crystal structure of mouse H2-M. Immunity 9, 385–393. Fung-Leung, W. P., Surh, C. D., Liljedahl, M., Pang, J., Leturcq, D., Peterson, P. A., Webb, S. R., and Karlsson, L. (1996). Antigen presentation and T cell development in H2-Mdeficient mice. Science 271, 1278–1281. Garcia-Lora, A., Algarra, I., and Garrido, F. (2003). MHC class I antigens, immune surveillance, and tumor immune escape. J. Cell Physiol. 195, 346–355. Germain, R. N., and Rinker, A. G., Jr. (1993). Peptide binding inhibits protein aggregation of invariant-chain free class II dimers and promotes surface expression of occupied molecules. Nature 363, 725–728. Gromet, M. A., Epstein, W. L., and Blois, M. S. (1978). The regressing thin malignant melanoma: A distinctive lesion with metastatic potential. Cancer 42, 2282–2292. Guncar, G., Pungercic, G., Klemencic, I., Turk, V., and Turk, D. (1999). Crystal structure of MHC class II-associated p41 li fragment bound to cathepsin L reveals the structural basis for differentiation between cathepsins L and S. EMBO J. 18, 793–803. Guy, K., Krajewski, A. S., and Dewar, A. E. (1986). Expression of MHC class II antigens in human B-cell leukemia and non-Hodgkin’s lymphoma. Br. J. Cancer 53, 161–173. Guyre, P. M., Graziano, R. F., Goldstein, J., Wallace, P. K., Morganelli, P. M., Wardwell, K., and Howell, A. L. (1997). Increased potency of Fc-receptor-targeted antigens. Cancer Immunol. Immunother. 45, 146–148. Hartmann, S., Kyewski, B., Sonnenburg, B., and Lucius, R. (1997). A filarial cysteine protease inhibitor down-regulates T cell proliferation and enhances interleukin-10 production. Eur. J. Immunol. 27, 2253–2260. Hernandez, L. D., Hueffer, K., Wenk, M. R., and Galan, J. E. (2004). Salmonella modulates vesicular traffic by altering phosphoinositide metabolism. Science 304, 1805–1807. Holden, D. W. (2002). Trafficking of the Salmonella vacuole in macrophages. Traffic 3, 161–169. Honey, K., Benlagha, K., Beers, C., Forbush, K., Teyton, L., Kleijmeer, M. J., Rudensky, A. Y., and Bendelac, A. (2002). Thymocyte expression of cathepsin L is essential for NKT cell development. Nat. Immunol. 3, 1069–1074. Honey, K., Duff, M., Beers, C., Brissette, W. H., Elliott, E. A., Peters, C., Maric, M., Cresswell, P., and Rudensky, A. (2001). Cathepsin S regulates the expression of cathepsin L and the turnover of gamma-interferon-inducible lysosomal thiol reductase in B lymphocytes. J. Biol. Chem. 276, 22573–22578. Hynes, N. E., and Stern, D. F. (1994). The biology of erbB-2/neu/HER-2 and its role in cancer. Biochim. Biophys. Acta 1198, 165–184. Jager, E., Jager, D., Karbach, J., Chen, Y. T., Ritter, G., Nagata, Y., Gnjatic, S., Stockert, E., Arand, M., Old, L. J., and Knuth, A. (2000). Identification of NY-ESO-1 epitopes presented by human histocompatibility antigen (HLA)-DRB4*0101-0103 and recognized by CD4(þ) T lymphocytes of patients with NY-ESO-1-expressing melanoma. J. Exp. Med. 191, 625–630.

MHC Class II Molecules in Cancer

151

Jordens, I., Fernandez-Borja, M., Marsman, M., Dusseljee, S., Janssen, L., Calafat, J., Janssen, H., Wubbolts, R., and Neefjes, J. (2001). The Rab7 effector protein RILP controls lysosomal transport by inducing the recruitment of dynein-dynactin motors. Curr. Biol. 11, 1680–1685. Karlsson, L., Surh, C. D., Sprent, J., and Peterson, P. A. (1991). A novel class II MHC molecule with unusual tissue distribution. Nature 351, 485–488. Kelly, A. P., Monaco, J. J., Cho, S. G., and Trowsdale, J. (1991). A new human HLA class IIrelated locus, DM. Nature 353, 571–573. Kenty, G., and Bikoff, E. K. (1999). BALB/c invariant chain mutant mice display relatively efficient maturation of CD4þ T cells in the periphery and secondary proliferative responses elicited upon peptide challenge. J. Immunol. 163, 232–241. Kinlen, L. (1992). Immunosuppressive therapy and acquired immunological disorders. Cancer Res. 52, 5474s–5476s. Kinlen, L. J. (2000). Non-Hodgkin’s lymphoma after immunosuppressive therapy. Gut 47, 462–463. Kirkin, A. F., thor Straten, P., Hansen, M. R., Barfoed, A., Dzhandzhugazyan, K. N., and Zeuthen, J. (1999). Establishment of gp100 and MART-1/Melan-A-specific cytotoxic T lymphocyte clones using in vitro immunization against preselected highly immunogenic melanoma cell clones. Cancer Immunol. Immunother. 48, 239–246. Kleijmeer, M., Ramm, G., Schuurhuis, D., Griffith, J., Rescigno, M., Ricciardi-Castagnoli, P., Rudensky, A. Y., Ossendorp, F., Melief, C. J., Stoorvogel, W., and Geuze, H. J. (2001). Reorganization of multivesicular bodies regulates MHC class II antigen presentation by dendritic cells. J. Cell Biol. 155, 53–63. Koppelman, B., Neefjes, J. J., de Vries, J. E., and de Waal Malefyt, R. (1997). Interleukin-10 down-regulates MHC class II alphabeta peptide complexes at the plasma membrane of monocytes by affecting arrival and recycling. Immunity 7, 861–871. Kropshofer, H., Arndt, S. O., Moldenhauer, G., Hammerling, G. J., and Vogt, A. B. (1997). HLA-DM acts as a molecular chaperone and rescues empty HLA-DR molecules at lysosomal pH. Immunity 6, 293–302. Kropshofer, H., Vogt, A. B., Moldenhauer, G., Hammer, J., Blum, J. S., and Hammerling, G. J. (1996). Editing of the HLA-DR-peptide repertoire by HLA-DM. EMBO J. 15, 6144–6154. Lamb, C. A., Yewdell, J. W., Bennink, J. R., and Cresswell, P. (1991). Invariant chain targets HLA class II molecules to acidic endosomes containing internalized influenza virus. Proc. Natl. Acad. Sci. USA 88, 5998–6002. Lanzavecchia, A. (1985). Antigen-specific interaction between T and B cells. Nature 314, 537–539. Lanzavecchia, A. (1996). Mechanisms of antigen uptake for presentation. Curr. Opin. Immunol. 8, 348–354. Lennon-Dumenil, A. M., Bakker, A. H., Wolf-Bryant, P., Ploegh, H. L., and LagaudriereGesbert, C. (2002). A closer look at proteolysis and MHC-class-II-restricted antigen presentation. Curr. Opin. Immunol. 14, 15–21. Lennon-Dumenil, A. M., Roberts, R. A., Valentijn, K., Driessen, C., Overkleeft, H. S., Erickson, A., Peters, P. J., Bikoff, E., Ploegh, H. L., and Wolf Bryant, P. (2001). The p41 isoform of invariant chain is a chaperone for cathepsin L. EMBO J. 20, 4055–4064. Liljedahl, M., Kuwana, T., Fung-Leung, W. P., Jackson, M. R., Peterson, P. A., and Karlsson, L. (1996). HLA-DO is a lysosomal resident which requires association with HLA-DM for efficient intracellular transport. EMBO J. 15, 4817–4824. Liljedahl, M., Winqvist, O., Surh, C. D., Wong, P., Ngo, K., Teyton, L., Peterson, P. A., Brunmark, A., Rudensky, A. Y., Fung-Leung, W. P., and Karlsson, L. (1998). Altered antigen presentation in mice lacking H2-O. Immunity 8, 233–243.

152

Marije Marsman et al.

Lotteau, V., Teyton, L., Peleraux, A., Nilsson, T., Karlsson, L., Schmid, S. L., Quaranta, V., and Peterson, P. A. (1990). Intracellular transport of class II MHC molecules directed by invariant chain. Nature 348, 600–605. Lovitch, S. B., Petzold, S. J., and Unanue, E. R. (2003). Cutting edge: H-2DM is responsible for the large differences in presentation among peptides selected by I-Ak during antigen processing. J. Immunol. 171, 2183–2186. Mach, B., Steimle, V., and Reith, W. (1994). MHC class II-deficient combined immunodeficiency: A disease of gene regulation. Immunol. Rev. 138, 207–221. Manoury, B., Gregory, W. F., Maizels, R. M., and Watts, C. (2001). Bm-CPI-2, a cystatin homolog secreted by the filarial parasite Brugia malayi inhibits class II MHC-restricted antigen processing. Curr. Biol. 11, 447–451. Marks, M. S., Blum, J. S., and Cresswell, P. (1990). Invariant chain trimers are sequestered in the rough endoplasmic reticulum in the absence of association with HLA class II antigens. J. Cell Biol. 111, 839–855. Marks, M. S., Germain, R. N., and Bonifacino, J. S. (1995a). Transient aggregation of major histocompatibility complex class II chains during assembly in normal spleen cells. J. Biol. Chem. 270, 10475–10481. Marks, M. S., Roche, P. A., van Donselaar, E., Woodruff, L., Peters, P. J., and Bonifacino, J. S. (1995b). A lysosomal targeting signal in the cytoplasmic tail of the beta chain directs HLADM to MHC class II compartments. J. Cell Biol. 131, 351–369. Martin, W. D., Hicks, G. G., Mendiratta, S. K., Leva, H. I., Ruley, H. E., and Van Kaer, L. (1996). H2-M mutant mice are defective in the peptide loading of class II molecules, antigen presentation, and T cell repertoire selection. Cell 84, 543–550. Masucci, M. G., Stam, N. J., Torsteinsdottir, S., Neefjes, J. J., Klein, G., and Ploegh, H. L. (1989). Allele-specific down-regulation of MHC class I antigens in Burkitt lymphoma lines. Cell Immunol. 120, 396–400. Matza, D., Kerem, A., Medvedovsky, H., Lantner, F., and Shachar, I. (2002a). Invariant chaininduced B cell differentiation requires intramembrane proteolytic release of the cytosolic domain. Immunity 17, 549–560. Matza, D., Kerem, A., and Shachar, I. (2003). Invariant chain, a chain of command. Trends Immunol. 24, 264–268. Matza, D., Lantner, F., Bogoch, Y., Flaishon, L., Hershkoviz, R., and Shachar, I. (2002b). Invariant chain induces B cell maturation in a process that is independent of its chaperonic activity. Proc. Natl. Acad. Sci. USA 99, 3018–3023. Meidenbauer, N., Marienhagen, J., Laumer, M., Vogl, S., Heymann, J., Andreesen, R., and Mackensen, A. (2003). Survival and tumor localization of adoptively transferred Melan-Aspecific T cells in melanoma patients. J. Immunol. 170, 2161–2169. Melief, C. J. (1992). Tumor eradication by adoptive transfer of cytotoxic T lymphocytes. Adv. Cancer Res. 58, 143–175. Mellins, E., Kempin, S., Smith, L., Monji, T., and Pious, D. (1991). A gene required for class IIrestricted antigen presentation maps to the major histocompatibility complex. J. Exp. Med. 174, 1607–1615. Milenic, D. E., Brady, E. D., and Brechbiel, M. W. (2004). Antibody-targeted radiation cancer therapy. Nat. Rev. Drug Discov. 3, 488–499. Miyazaki, T., Wolf, P., Tourne, S., Waltzinger, C., Dierich, A., Barois, N., Ploegh, H., Benoist, C., and Mathis, D. (1996). Mice lacking H2-M complexes, enigmatic elements of the MHC class II peptide-loading pathway. Cell 84, 531–541. Moore, K. W., Rousset, F., and Banchereau, J. (1991). Evolving principles in immunopathology: Interleukin 10 and its relationship to Epstein-Barr virus protein BCRF1. Springer Semin. Immunopathol. 13, 157–166.

MHC Class II Molecules in Cancer

153

Morris, P., Shaman, J., Attaya, M., Amaya, M., Goodman, S., Bergman, C., Monaco, J. J., and Mellins, E. (1994). An essential role for HLA-DM in antigen presentation by class II major histocompatibility molecules. Nature 368, 551–554. Mosyak, L., Zaller, D. M., and Wiley, D. C. (1998). The structure of HLA-DM, the peptide exchange catalyst that loads antigen onto class II MHC molecules during antigen presentation. Immunity 9, 377–383. Munz, C., Bickham, K. L., Subklewe, M., Tsang, M. L., Chahroudi, A., Kurilla, M. G., Zhang, D., O’Donnell, M., and Steinman, R. M. (2000). Human CD4(þ) T lymphocytes consistently respond to the latent Epstein-Barr virus nuclear antigen EBNA1. J. Exp. Med. 191, 1649–1660. Mutis, T., and Goulmy, E. (2002). Hematopoietic system-specific antigens as targets for cellular immunotherapy of hematological malignancies. Semin. Hematol. 39, 23–31. Nakagawa, T. Y., and Rudensky, A. Y. (1999). The role of lysosomal proteinases in MHC class II-mediated antigen processing and presentation. Immunol. Rev. 172, 121–129. Neefjes, J. J., and Ploegh, H. L. (1992). Inhibition of endosomal proteolytic activity by leupeptin blocks surface expression of MHC class II molecules and their conversion to SDS resistance alpha beta heterodimers in endosomes. EMBO J. 11, 411–416. Neefjes, J. J., Stollorz, V., Peters, P. J., Geuze, H. J., and Ploegh, H. L. (1990). The biosynthetic pathway of MHC class II but not class I molecules intersects the endocytic route. Cell 61, 171–183. Nijenhuis, M., Calafat, J., Kuijpers, K. C., Janssen, H., de Haas, M., Nordeng, T. W., Bakke, O., and Neefjes, J. J. (1994). Targeting major histocompatibility complex class II molecules to the cell surface by invariant chain allows antigen presentation upon recycling. Eur. J. Immunol. 24, 873–883. Nijenhuis, M., and Neefjes, J. (1994). Early events in the assembly of major histocompatibility complex class II heterotrimers from their free subunits. Eur. J. Immunol. 24, 247–256. Odorizzi, C. G., Trowbridge, I. S., Xue, L., Hopkins, C. R., Davis, C. D., and Collawn, J. F. (1994). Sorting signals in the MHC class II invariant chain cytoplasmic tail and transmembrane region determine trafficking to an endocytic processing compartment. J. Cell Biol. 126, 317–330. Ogrinc, T., Dolenc, I., Ritonja, A., and Turk, V. (1993). Purification of the complex of cathepsin L and the MHC class II-associated invariant chain fragment from human kidney. FEBS Lett. 336, 555–559. Ossendorp, F., Mengede, E., Camps, M., Filius, R., and Melief, C. J. (1998). Specific T helper cell requirement for optimal induction of cytotoxic T lymphocytes against major histocompatibility complex class II negative tumors. J. Exp. Med. 187, 693–702. Ostmeier, H., Fuchs, B., Otto, F., Mawick, R., Lippold, A., Krieg, V., and Suter, L. (1999). Can immunohistochemical markers and mitotic rate improve prognostic precision in patients with primary melanoma? Cancer 85, 2391–2399. O’Sullivan, D. M., Noonan, D., and Quaranta, V. (1987). Four la invariant chain forms derive from a single gene by alternate splicing and alternate initiation of transcription/translation. J. Exp. Med. 166, 444–460. Parkhurst, M. R., DePan, C., Riley, J. P., Rosenberg, S. A., and Shu, S. (2003). Hybrids of dendritic cells and tumor cells generated by electrofusion simultaneously present immunodominant epitopes from multiple human tumor-associated antigens in the context of MHC class I and class II molecules. J. Immunol. 170, 5317–5325. Pashine, A., Busch, R., Belmares, M. P., Munning, J. N., Doebele, R. C., Buckingham, M., Nolan, G. P., and Mellins, E. D. (2003). Interaction of HLA-DR with an acidic face of HLA-DM disrupts sequence-dependent interactions with peptides. Immunity 19, 183–192. Passlick, B., Izbicki, J. R., Simmel, S., Kubuschok, B., Karg, O., Habekost, M., Thetter, O., Schweiberer, L., and Pantel, K. (1994). Expression of major histocompatibility class I and

154

Marije Marsman et al.

class II antigens and intercellular adhesion molecule-1 on operable non-small cell lung carcinomas: Frequency and prognostic significance. Eur. J. Cancer 30A, 376–381. Perraudeau, M., Taylor, P. R., Stauss, H. J., Lindstedt, R., Bygrave, A. E., Pappin, D. J., Ellmerich, S., Whitten, A., Rahman, D., Canas, B., Walport, M. J., Botto, M., and Altmann, D. M. (2000). Altered major histocompatibility complex class II peptide loading in H2-Odeficient mice. Eur. J. Immunol. 30, 2871–2880. Peters, P. J., Neefjes, J. J., Oorschot, V., Ploegh, H. L., and Geuze, H. J. (1991). Segregation of MHC class II molecules from MHC class I molecules in the Golgi complex for transport to lysosomal compartments. Nature 349, 669–676. Peterson, M., and Miller, J. (1992). Antigen presentation enhanced by the alternatively spliced invariant chain gene product p41. Nature 357, 596–598. Pierre, P., and Mellman, I. (1998). Developmental regulation of invariant chain proteolysis controls MHC class II trafficking in mouse dendritic cells. Cell 93, 1135–1145. Pierre, P., Turley, S. J., Gatti, E., Hull, M., Meltzer, J., Mirza, A., Inaba, K., Steinman, R. M., and Mellman, I. (1997). Developmental regulation of MHC class II transport in mouse dendritic cells. Nature 388, 787–792. Pieters, J., Bakke, O., and Dobberstein, B. (1993). The MHC class II-associated invariant chain contains two endosomal targeting signals within its cytoplasmic tail. J. Cell Sci. 106(Pt. 3), 831–846. Pinet, V., Vergelli, M., Martin, R., Bakke, O., and Long, E. O. (1995). Antigen presentation mediated by recycling of surface HLA-DR molecules. Nature 375, 603–606. Pond, L., Kuhn, L. A., Teyton, L., Schutze, M. P., Tainer, J. A., Jackson, M. R., and Peterson, P. A. (1995). A role for acidic residues in di-leucine motif-based targeting to the endocytic pathway. J. Biol. Chem. 270, 19989–19997. Pond, L., and Watts, C. (1999). Functional early endosomes are required for maturation of major histocompatibility complex class II molecules in human B lymphoblastoid cells. J. Biol. Chem. 274, 18049–18054. Raposo, G., Nijman, H. W., Stoorvogel, W., Liejendekker, R., Harding, C. V., Melief, C. J., and Geuze, H. J. (1996). B lymphocytes secrete antigen-presenting vesicles. J. Exp. Med. 183, 1161–1172. Reid, P. A., and Watts, C. (1990). Cycling of cell-surface MHC glycoproteins through primaquine-sensitive intracellular compartments. Nature 346, 655–657. Renkvist, N., Castelli, C., Robbins, P. F., and Parmiani, G. (2001). A listing of human tumor antigens recognized by T cells. Cancer Immunol. Immunother. 50, 3–15. Riberdy, J. M., Newcomb, J. R., Surman, M. J., Barbosa, J. A., and Cresswell, P. (1992). HLADR molecules from an antigen-processing mutant cell line are associated with invariant chain peptides. Nature 360, 474–477. Riese, R. J., and Chapman, H. A. (2000). Cathepsins and compartmentalization in antigen presentation. Curr. Opin. Immunol. 12, 107–113. Riese, R. J., Mitchell, R. N., Villadangos, J. A., Shi, G. P., Palmer, J. T., Karp, E. R., De Sanctis, G. T., Ploegh, H. L., and Chapman, H. A. (1998). Cathepsin S activity regulates antigen presentation and immunity. J. Clin. Invest. 101, 2351–2363. Riese, R. J., Wolf, P. R., Bromme, D., Natkin, L. R., Villadangos, J. A., Ploegh, H. L., and Chapman, H. A. (1996). Essential role for cathepsin S in MHC class II-associated invariant chain processing and peptide loading. Immunity 4, 357–366. Rimsza, L. M., Roberts, R. A., Miller, T. P., Unger, J. M., LeBlanc, M., Braziel, R. M., Weisenberger, D. D., Chan, W. C., Muller-Hermelink, H. K., Jaffe, E. S., Gascoyne, R. D., Campo, E., Fuchs, D. A., Spier, C. M., Fisher, R. I., Delabie, J., Rosenwald, A., Staudt, L. M., and Grogan, T. M. (2004). Loss of MHC class II gene and protein expression in diffuse large B-cell lymphoma is related to decreased tumor immunosurveillance and poor patient survival

MHC Class II Molecules in Cancer

155

regardless of other prognostic factors: A follow-up study from the Leukemia and Lymphoma Molecular Profiling Project. Blood 103, 4251–4258. Robinson, J., Waller, M. J., Parham, P., Bodmer, J. G., and Marsh, S. G. (2001). IMGT/HLA Database—A sequence database for the human major histocompatibility complex. Nucleic Acids Res. 29, 210–213. Roche, P. A., Marks, M. S., and Cresswell, P. (1991). Formation of a nine-subunit complex by HLA class II glycoproteins and the invariant chain. Nature 354, 392–394. Roche, P. A., Teletski, C. L., Stang, E., Bakke, O., and Long, E. O. (1993). Cell surface HLADR-invariant chain complexes are targeted to endosomes by rapid internalization. Proc. Natl. Acad. Sci. USA 90, 8581–8585. Rock, K. L., Benacerraf, B., and Abbas, A. K. (1984). Antigen presentation by hapten-specific B lymphocytes. I. Role of surface immunoglobulin receptors. J. Exp. Med. 160, 1102–1113. Roucard, C., Thomas, C., Pasquier, M. A., Trowsdale, J., Sotto, J. J., Neefjes, J., and van Ham, M. (2001). In vivo and in vitro modulation of HLA-DM and HLA-DO is induced by B lymphocyte activation. J. Immunol. 167, 6849–6858. Sahara, H., and Shastri, N. (2003). Second class minors: Molecular identification of the autosomal H46 histocompatibility locus as a peptide presented by major histocompatibility complex class II molecules. J. Exp. Med. 197, 375–385. Sairenji, T., Ohnishi, E., Inouye, S., and Kurata, T. (1998). Induction of interleukin-10 on activation of Epstein-Barr virus in EBV-infected B-cell lines. Viral Immunol. 11, 221–231. Saleh, F. H., Crotty, K. A., Hersey, P., and Menzies, S. W. (2001). Primary melanoma tumor regression associated with an immune response to the tumor-associated antigen melan-A/ MART-1. Int. J. Cancer 94, 551–557. Saleh, F. H., Crotty, K. A., Hersey, P., Menzies, S. W., and Rahman, W. (2003). Autonomous histopathological regression of primary tumors associated with specific immune responses to cancer antigens. J. Pathol. 200, 383–395. Salomon, D. S., Brandt, R., Ciardiello, F., and Normanno, N. (1995). Epidermal growth factorrelated peptides and their receptors in human malignancies. Crit. Rev. Oncol. Hematol. 19, 183–232. Sanderson, F., Kleijmeer, M. J., Kelly, A., Verwoerd, D., Tulp, A., Neefjes, J. J., Geuze, H. J., and Trowsdale, J. (1994). Accumulation of HLA-DM, a regulator of antigen presentation, in MHC class II compartments. Science 266, 1566–1569. Sanderson, F., Thomas, C., Neefjes, J., and Trowsdale, J. (1996). Association between HLADM and HLA-DR in vivo. Immunity 4, 87–96. Santin, A. D., Hermonat, P., Ravaggi, A., Chiriva-Internati, M., Hiserodt, J. C., Tian, E., Carter, C. A., Pecorelli, S., and Parham, G. P. (1998). Effects of retinoic acid combined with interferon-gamma on the expression of major-histocompatibility-complex molecules and intercellular adhesion molecule-1 in human cervical cancer. Int. J. Cancer 75, 254–258. Schaiff, W. T., Hruska, K. A., Jr., McCourt, D. W., Green, M., and Schwartz, B. D. (1992). HLA-DR associates with specific stress proteins and is retained in the endoplasmic reticulum in invariant chain negative cells. J. Exp. Med. 176, 657–666. Schonemeyer, A., Lucius, R., Sonnenburg, B., Brattig, N., Sabat, R., Schilling, K., Bradley, J., and Hartmann, S. (2001). Modulation of human T cell responses and macrophage functions by onchocystatin, a secreted protein of the filarial nematode Onchocerca volvulus. J. Immunol. 167, 3207–3215. Schultz, H., Schinke, S., Weiss, J., Cerundolo, V., Gross, W. L., and Gadola, S. (2003). BPIANCA in transporter associated with antigen presentation (TAP) deficiency: Possible role in susceptibility to Gram-negative bacterial infections. Clin. Exp. Immunol. 133, 252–259. Schumacher, T. N., Heemels, M. T., Neefjes, J. J., Kast, W. M., Melief, C. J., and Ploegh, H. L. (1990). Direct binding of peptide to empty MHC class I molecules on intact cells and in vitro. Cell 62, 563–567.

156

Marije Marsman et al.

Shachar, I., and Flavell, R. A. (1996). Requirement for invariant chain in B cell maturation and function. Science 274, 106–108. Sherman, M. A., Weber, D. A., and Jensen, P. E. (1995). DM enhances peptide binding to class II MHC by release of invariant chain-derived peptide. Immunity 3, 197–205. Siegrist, C. A., Martinez-Soria, E., Kern, I., and Mach, B. (1995). A novel antigen-processingdefective phenotype in major histocompatibility complex class II-positive CIITA transfectants is corrected by interferon-gamma. J. Exp. Med. 182, 1793–1799. Siemasko, K., and Clark, M. R. (2001). The control and facilitation of MHC class II antigen processing by the BCR. Curr. Opin. Immunol. 13, 32–36. Slager, E. H., Borghi, M., van der Minne, C. E., Aarnoudse, C. A., Havenga, M. J., Schrier, P. I., Osanto, S., and Griffioen, M. (2003). CD4þ Th2 cell recognition of HLA-DR-restricted epitopes derived from CAMEL: A tumor antigen translated in an alternative open reading frame. J. Immunol. 170, 1490–1497. Sloan, V. S., Cameron, P., Porter, G., Gammon, M., Amaya, M., Mellins, E., and Zaller, D. M. (1995). Mediation by HLA-DM of dissociation of peptides from HLA-DR. Nature 375, 802–806. Song, W., Wagle, N. M., Banh, T., Whiteford, C. C., Ulug, E., and Pierce, S. K. (1997). Wortmannin, a phosphatidylinositol 3-kinase inhibitor, blocks the assembly of peptideMHC class II complexes. Int. Immunol. 9, 1709–1722. Spierings, E., Wieles, B., and Goulmy, E. (2004). Minor histocompatibility antigens—Big in tumor therapy. Trends Immunol. 25, 56–60. Steimle, V., Otten, L. A., Zufferey, M., and Mach, B. (1993). Complementation cloning of an MHC class II transactivator mutated in hereditary MHC class II deficiency (or bare lymphocyte syndrome). Cell 75, 135–146. Stern, L. J., Brown, J. H., Jardetzky, T. S., Gorga, J. C., Urban, R. G., Strominger, J. L., and Wiley, D. C. (1994). Crystal structure of the human class II MHC protein HLA-DR1 complexed with an influenza virus peptide. Nature 368, 215–221. Stratikos, E., Mosyak, L., Zaller, D. M., and Wiley, D. C. (2002). Identification of the lateral interaction surfaces of human histocompatibility leukocyte antigen (HLA)-DM with HLADR1 by formation of tethered complexes that present enhanced HLA-DM catalysis. J. Exp. Med. 196, 173–183. Strubin, M., Berte, C., and Mach, B. (1986). Alternative splicing and alternative initiation of translation explain the four forms of the la antigen-associated invariant chain. EMBO J. 5, 3483–3488. Sutmuller, R. P., Schurmans, L. R., van Duivenvoorde, L. M., Tine, J. A., van Der Voort, E. I., Toes, R. E., Melief, C. J., Jager, M. J., and Offringa, R. (2000). Adoptive T cell immunotherapy of human uveal melanoma targeting gp 100. J. Immunol. 165, 7308–7315. Takamura, Y., Ikeda, H., Kanaseki, T., Toyota, M., Tokino, T., Imai, K., Houkin, K., and Sato, N. (2004). Regulation of MHC class II expression in glioma cells by class II transactivator (CIITA). Glia 45, 392–405. Tan, M. C., Mommaas, A. M., Drijfhout, J. W., Jordens, R., Onderwater, J. J., Verwoerd, D., Mulder, A. A., van der Heiden, A. N., Scheidegger, D., Oomen, L. C., Ottenhoff, T. H., Tulp, A., Neefjes, J. J., and Koning, F. (1997). Mannose receptor-mediated uptake of antigens strongly enhances HLA class II-restricted antigen presentation by cultured dendritic cells. Eur. J. Immunol. 27, 2426–2435. Tatsumi, T., Kierstead, L. S., Ranieri, E., Gesualdo, L., Schena, F. P., Finke, J. H., Bukowski, R. M., Mueller-Berghaus, J., Kirkwood, J. M., Kwok, W. W., and Storkus, W. J. (2002). Disease-associated bias in T helper type 1 (Th1)/Th2 CD4(þ) T cell responses against MAGE-6 in HLA-DRB10401(þ) patients with renal cell carcinoma or melanoma. J. Exp. Med. 196, 619–628.

MHC Class II Molecules in Cancer

157

Thurner, B., Haendle, I., Roder, C., Dieckmann, D., Keikavoussi, P., Jonuleit, H., Bender, A., Maczek, C., Schreiner, D., von den Driesch, P., Brocker, E. B., Steinman, R. M., Enk, A., Kampgen, E., and Schuler, G. (1999). Vaccination with mage-3A1 peptide-pulsed mature, monocyte-derived dendritic cells expands specific cytotoxic T cells and induces regression of some metastases in advanced stage IV melanoma. J. Exp. Med. 190, 1669–1678. Tonnelle, C., DeMars, R., and Long, E. O. (1985). DO beta: A new beta chain gene in HLA-D with a distinct regulation of expression. EMBO J. 4, 2839–2847. Topalian, S. L., Gonzales, M. I., Parkhurst, M., Li, Y. F., Southwood, S., Sette, A., Rosenberg, S. A., and Robbins, P. F. (1996). Melanoma-specific CD4þ T cells recognize nonmutated HLA-DR-restricted tyrosinase epitopes. J. Exp. Med. 183, 1965–1971. Topalian, S. L., Rivoltini, L., Mancini, M., Markus, N. R., Robbins, P. F., Kawakami, Y., and Rosenberg, S. A. (1994). Human CD4þ T cells specifically recognize a shared melanomaassociated antigen encoded by the tyrosinase gene. Proc. Natl. Acad. Sci. USA 91, 9461–9465. Townsend, A., Ohlen, C., Bastin, J., Ljunggren, H. G., Foster, L., and Karre, K. (1989). Association of class I major histocompatibility heavy and light chains induced by viral peptides. Nature 340, 443–448. Trombetta, E. S., Ebersold, M., Garrett, W., Pypaert, M., and Mellman, I. (2003). Activation of lysosomal function during dendritic cell maturation. Science 299, 1400–1403. Ullrich, H. J., Doring, K., Gruneberg, U., Jahnig, F., Trowsdale, J., and van Ham, S. M. (1997). Interaction between HLA-DM and HLA-DR involves regions that undergo conformational changes at lysosomal pH. Proc. Natl. Acad. Sci. USA 94, 13163–13168. van Bergen, J., Ossendorp, F., Jordens, R., Mommaas, A. M., Drijfhout, J. W., and Koning, F. (1999). Get into the groove! Targeting antigens to MHC class II. Immunol. Rev. 172, 87–96. van der Bruggen, P., Traversari, C., Chomez, P., Lurquin, C., De Plaen, E., Van den Eynde, B., Knuth, A., and Boon, T. (1991). A gene encoding an antigen recognized by cytolytic T lymphocytes on a human melanoma. Science 254, 1643–1647. van Ham, M., van Lith, M., Lillemeier, B., Tjin, E., Gruneberg, U., Rahman, D., Pastoors, L., van Meijgaarden, K., Roucard, C., Trowsdale, J., Ottenhoff, T., Pappin, D., and Neefjes, J. (2000). Modulation of the major histocompatibility complex class II-associated peptide repertoire by human histocompatibility leukocyte antigen (HLA)-DO. J. Exp. Med. 191, 1127–1136. van Ham, S. M., Gruneberg, U., Malcherek, G., Broker, I., Melms, A., and Trowsdale, J. (1996). Human histocompatibility leukocyte antigen (HLA)-DM edits peptides presented by HLA-DR according to their ligand binding motifs. J. Exp. Med. 184, 2019–2024. van Ham, S. M., Tjin, E. P., Lillemeier, B. F., Gruneberg, U., van Meijgaarden, K. E., Pastoors, L., Verwoerd, D., Tulp, A., Canas, B., Rahman, D., Ottenhoff, T. H., Pappin, D. J., Trowsdale, J., and Neefjes, J. (1997). HLA-DO is a negative modulator of HLA-DM-mediated MHC class II peptide loading. Curr. Biol. 7, 950–957. van Lith, M., van Ham, M., Griekspoor, A., Tjin, E., Verwoerd, D., Calafat, J., Janssen, H., Reits, E., Pastoors, L., and Neefjes, J. (2001). Regulation of MHC class II antigen presentation by sorting of recycling HLA-DM/DO and class II within the multivesicular body. J. Immunol. 167, 884–892. van Lith, M., van Ham, M., and Neefjes, J. (2002). Novel polymorphisms in HLA-DOA and HLA-DOB in B-cell malignancies. Immunogenetics 54, 591–595. Vieira, P., de Waal-Malefyt, R., Dang, M. N., Johnson, K. E., Kastelein, R., Fiorentino, D. F., deVries, J. E., Roncarolo, M. G., Mosmann, T. R., and Moore, K. W. (1991). Isolation and expression of human cytokine synthesis inhibitory factor cDNA clones: Homology to Epstein-Barr virus open reading frame BCRFI. Proc. Natl. Acad. Sci. USA 88, 1172–1176.

158

Marije Marsman et al.

Villadangos, J. A., Riese, R. J., Peters, C., Chapman, H. A., and Ploegh, H. L. (1997). Degradation of mouse invariant chain: Roles of cathepsins S and D and the influence of major histocompatibility complex polymorphism. J. Exp. Med. 186, 549–560. Viville, S., Neefjes, J., Lotteau, V., Dierich, A., Lemeur, M., Ploegh, H., Benoist, C., and Mathis, D. (1993). Mice lacking the MHC class II-associated invariant chain. Cell 72, 635–648. Vogt, A. B., Moldenhauer, G., Hammerling, G. J., and Kropshofer, H. (1997). HLA-DM stabilizes empty HLA-DR molecules in a chaperone-like fashion. Immunol. Lett. 57, 209–211. Voo, K. S., Fu, T., Heslop, H. E., Brenner, M. K., Rooney, C. M., and Wang, R. F. (2002). Identification of HLA-DP3-restricted peptides from EBNA1 recognized by CD4(þ) T cells. Cancer Res. 62, 7195–7199. Wang, E., Phan, G. Q., and Marincola, F. M. (2001). T-cell-directed cancer vaccines: The melanoma model. Expert Opin. Biol. Ther. 1, 277–290. Weber, D. A., Evavold, B. D., and Jensen, P. E. (1996). Enhanced dissociation of HLA-DRbound peptides in the presence of HLA-DM. Science 274, 618–620. Wubbolts, R., Fernandez-Borja, M., Jordens, I., Reits, E., Dusseljee, S., Echeverri, C., Vallee, R. B., and Neefjes, J. (1999). Opposing motor activities of dynein and kinesin determine retention and transport of MHC class II-containing compartments. J. Cell. Sci. 112(Pt. 6), 785–795. Wubbolts, R., Fernandez-Borja, M., Oomen, L., Verwoerd, D., Janssen, H., Calafat, J., Tulp, A., Dusseljee, S., and Neefjes, J. (1996). Direct vesicular transport of MHC class II molecules from lysosomal structures to the cell surface. J. Cell. Biol. 135, 611–622. Wubbolts, R., and Neefjes, J. (1999). Intracellular transport and peptide loading of MHC class II molecules: Regulation by chaperones and motors. Immunol. Rev. 172, 189–208. You, Z., Huang, X., Hester, J., Toh, H. C., and Chen, S. Y. (2001). Targeting dendritic cells to enhance DNA vaccine potency. Cancer Res. 61, 3704–3711. Young, J. A., and Trowsdale, J. (1990). The HLA-DNA (DZA) gene is correctly expressed as a 1.1 kb mature mRNA transcript. Immunogenetics 31, 386–388. Zeng, G., Touloukian, C. E., Wang, X., Restifo, N. P., Rosenberg, S. A., and Wang, R. F. (2000). Identification of CD4þ T cell epitopes from NY-ESO-1 presented by HLA-DR molecules. J. Immunol. 165, 1153–1159. Zeng, G., Wang, X., Robbins, P. F., Rosenberg, S. A., and Wang, R. F. (2001). CD4(þ) T cell recognition of MHC class II-restricted epitopes from NY-ESO-1 presented by a prevalent HLA DP4 allele: Association with NY-ESO-1 antibody production. Proc. Natl. Acad. Sci. USA 98, 3964–3969. Zitvogel, L., Regnault, A., Lozier, A., Wolfers, J., Flament, C., Tenza, D., Ricciardi-Castagnoli, P., Raposo, G., and Amigorena, S. (1998). Eradication of established murine tumors using a novel cell-free vaccine: Dendritic cell-derived exosomes. Nat. Med. 4, 594–600. Zwart, W. in preparation.

Soluble Mediators of Inflammation During Tumor Development Stephen C. Robinson* and Lisa M. Coussens*,{,{ *Cancer Research Institute, {

Department of Pathology, and Comprehensive Cancer Center, University of California San Francisco, San Francisco, California 94143 {

I. Introduction II. Leukocytes and Inflammation A. Adherence and Immobilization of Leukocytes at Sites of Tissue Damage B. Cytokines, Chemokines, and Leukocyte Recruitment III. Inflammation and Tumor Progression A. TNF- and Cancer Development B. Chemokines Promote Neoplastic Progression C. Altered Regulation of Chemokines in Cancer D. Chemokines and Angiogenesis E. Chemokines and Metastasis IV. Conclusions References

Tissues maintain homeostasis by monitoring and responding to varied physical interactions between cells and their microenvironment. In situations where acute tissue damage occurs, such as wounding, pathogenic assault, or toxic exposure, regulatory circuits that monitor tissue homeostasis are rapidly engaged to initiate tissue repair by regulating cell polarity, proliferation and death, matrix metabolism, inflammation, and vascular and lymphatic function. The critical feature of regulating these acute responses is the innate ability to discriminate between homeostatic versus damaged tissue states and engage or disengage regulatory machinery as appropriate; thus, a major distinction between acute versus chronic disease is the altered ability to appropriately activate and/ or inactivate reparative regulatory programs. Since cancer is a chronic disease characterized by altered cell polarity, enhanced cell survival, inflammation, increased matrix metabolism, and enhanced vascular and lymphatic function, considerable attention is now focused on understanding the cellular and molecular mechanisms regulating these responsive pathways. Since chemoattractant cytokines are important mediators of leukocyte recruitment following acute tissue stress, and demonstrate altered characteristics of expression and activation in chronically inflamed tissue, they have been implicated as key regulators of inflammation and angiogenesis during cancer development. This chapter focuses on the clinical and experimental data implicating proinflammatory cytokines and chemokines as important potentiators of carcinogenesis. ß 2005 Elsevier Inc.

Advances in CANCER RESEARCH 0065-230X/05 $35.00

159

Copyright 2005, Elsevier Inc. All rights reserved

160

Stephen C. Robinson and Lisa M. Coussens

I. INTRODUCTION Tissue and organ homeostasis is maintained by tightly regulated interactions between cells and the microenvironment in which they live (Bissell and Radisky, 2001; Radisky et al., 2001). In response to tissue injury, for example, a skin wound, a multifactorial network of chemical signals, released by injured cells or produced by and in response to infectious agents, such as lipopolysaccharide (LPS), endotoxin, or activation of complement, initiates and maintains a host response designed to ‘‘heal’’ afflicted tissues that involves activation and directed migration of leukocytes from the venous system to sites of damage. A multistep mechanism is believed to coordinate recruitment of inflammatory cells to sites of tissue injury and to the provisional extracellular matrix (ECM) that forms scaffolding upon which epithelial cells, fibroblast, and vascular cells proliferate and migrate (Fig. 1), and together provides a nidus for reconstitution of the normal tissue microenvironment (Chettibi and Ferguson, 1999).

Fig. 1 Neoplastic progression is associated with inflammation and angiogenesis. Hematoxylin and eosin (H&E; panels A and F) staining of paraffin-embedded murine skin sections reveals histopathologic differences between normal ear skin (A) and a carcinoma removed from a K14-HPV16 transgenic mouse (F). Immunolocalization of CD45þ leukocytes (panels B and D) reveals a marked increase in premalignant tissue (D) compared to quiescent stroma beneath normal epidermis (B; arrows). The morphology and architecture of capillaries is also altered during malignant progression; immunolocalization of CD31þ endothelial cells (panels C and E; red staining; arrows) demonstrates an increase in the number and diameter of capillaries in premalignant tissue (E) compared with non-neoplastic normal skin (C).

Soluble Mediators of Inflammation in Tumors

161

A family of soluble chemotactic cytokines that possess a relatively high degree of specificity for chemoattraction of specific leukocyte populations (Balkwill and Mantovani, 2001; Homey et al., 2002; Proudfoot, 2002; Rossi and Zlotnik, 2000) recruit downstream effector cells and dictate the natural evolution of inflammatory responses. The profile of cytokines and chemokines persisting at any inflammatory site plays an important role in either resolution of tissue damage or development of chronic disease. The key concept is that normal wound healing is usually self-limiting where production of anti-inflammatory cytokines follows activation of pro-inflammatory cytokines closely, thus enabling resolution of damage. Chronic inflammation, by contrast, appears to result from persistence of the initiating factors or in failure of mechanisms required for resolving inflammatory responses. Deregulation of any of the converging factors can lead to a myriad of abnormalities and, ultimately, disease pathogenesis. Chronic tissue inflammation is now regarded as a risk factor for cancer (Balkwill and Mantovani, 2001; Coussens and Werb, 2002; Shacter and Weitzman, 2002; Thun et al., 2004); thus, understanding how leukocytes interact with soluble and insoluble tissue components may help to elucidate rate-limiting regulatory mechanisms that may be used as therapeutic targets for anticancer modalities. This chapter focuses on important cell–cell and cell–ECM interactions that regulate leukocyte recruitment and activation downstream of cytokine and chemokine signaling, and examines key differences between acute versus chronic inflammation associated with angiogenesis and cancer development.

II. LEUKOCYTES AND INFLAMMATION Bone marrow-derived stem cells divide to produce common lymphoid and myeloid precursor stem cells from which circulating leukocytes are derived (Janeway et al., 2001). T and B lymphocytes (lymphoid cells) compose the adaptive immune response by virtue of their memory functions and heightened response capabilities following exposure to antigens (Janeway et al., 2001). In contrast, leukocytes of the innate immune system, for example, polymorphonuclear granulocytes (neutrophils, eosinophils, and basophils), monocytes/macrophages, mast cells (MCs), and dendritic cells, do not possess memory capabilities, but instead respond to antigens de novo following each encounter. Resident tissue macrophages and MCs act as sentinel cells within tissues and are the first lines of defense against bacterial and allergic stimuli, respectively (Auger and Roos, 1992; Galli and Kitamura, 1987). MCs play important roles in acute inflammation due to their release of stored and newly synthesized inflammatory mediators, such as histamine,

162

Stephen C. Robinson and Lisa M. Coussens

cytokines, and proteases complexed to highly sulfated proteoglycans, as well as lipid mediators and cytokines (Wedemeyer et al., 2000). MCs and/ or neutrophils (and sometimes eosinophils) are the first recruited effectors of the acute inflammatory response. Monocytes, which differentiate into macrophages in tissues (Auger and Roos, 1992), migrate to sites of tissue injury guided by chemotactic factors. Monocytes, in the presence of granulocyte macrophage-colony stimulating factor (GM-CSF) and interleukin-4 (IL-4), differentiate into immature dendritic cells (DC) (Sallusto and Lanzavecchia, 1994; Talmor et al., 1998). DCs migrate into inflamed peripheral tissue where they capture antigens and, following maturation, migrate to lymph nodes to stimulate T lymphocyte activation. Once activated, macrophages are a major source of growth factors and cytokines, and profoundly affect endothelial, epithelial, and mesenchymal cells in the local microenvironment. These diverse leukocytes, each endowed with unique antimicrobial, cell stimulatory, and ECM modulation properties, are exquisitely designed to aid in regenerative processes. Their continuous circulation within the vascular system enables rapid responses to tissue damage, including acute responses to pathogenic assault or enhanced cellular proliferation resulting from initiating mutational events that precede neoplastic progression.

A. Adherence and Immobilization of Leukocytes at Sites of Tissue Damage Leukocytes are typically freely circulating within vascular networks (Kubes, 2002). Following an inflammatory stimulus, however, a multistep pathway directing leukocyte infiltration of tissue is enabled that involves leukocyte adherence to vascular walls via engagement with endothelial cell adhesion molecules P-, L-, and E-selectin (Gotsch et al., 1994; Ley, 2003; Rosen, 2004; Ulbrich et al., 2003; Venturi et al., 2003; Weller et al., 1992). Weak adhesive interactions between selectins and selectin ligands permit leukocyte rolling and the interaction of leukocyte chemokine receptors with chemokines bound to glycosaminoglycans (GAGs) on the cell surface of endothelial cells. Signaling of chemokines through leukocyte chemokine receptors promotes leukocyte avidity for the endothelium and, finally, initiates leukocyte immobilization via 4 1 and 4 7 integrins and subsequent binding to endothelial vascular cell adhesion molecule-1 (VCAM-1) and MAdCAM-1, respectively (Fig. 2). Following firm adhesion, leukocytes undergo transendothelial cell migration (diapedesis) that can progress under conditions of shear flow (Cinamon et al., 2001) and, most likely, involves alterations in endothelial cell junctions (reviewed by Muller, 2001). This multistep process of leukocyte adherence, rolling, and immobilization on vascular endothelium is tightly regulated and responsive to

Soluble Mediators of Inflammation in Tumors

163

Fig. 2 Leukocyte rolling and transendothelial cell migration. Leukocytes roll along inflamed endothelium mediated by expression of selections on both endothelial cells (E- and P-selectin) and leukocytes (L-selectin) that interact transiently with P selectin glycoprotein ligand 1 (PSGL-1), ESGL-1 expressed on leukocytes, and mucosal addressin cell adhesion molecule 1 (MAdCAM-1) expressed on endothelial cells. Leukocyte rolling permits chemokines attached to glycosaminoglycans on the lumen of endothelial cells to stimulate chemokine receptors expressed on leukocytes. Chemokine signaling through chemokine receptors activates leukocyte integrins, permitting firm arrest. 4 integrins very late antigen (VLA-4/ 4 1) and 4 7, together with 2 integrins lymphocyte function antigen (LFA-1/L 2) and Mac-1 (M 2), interact with vascular cellular adhesion molecule 1 (VCAM-1), intracellular adhesion molecule 1/2 (ICAM-1/2), and MAdCAM-1. Finally, leukocytes undergo transendothelial cell migration via endothelial cell junctions in response to chemotactic gradients.

spatial and temporal perturbations in tissue organization and architecture that variably alter bioavailability of soluble mediators of the process, namely, chemokines, cytokines, and selectins. A common feature of vasculature in acutely inflamed tissue and angiogenic vasculature associated with tumor development is high selectin expression on vascular endothelial cells (Meager, 1999; Rosen, 2004) and enhanced expression and/or bioavailability of chemokines in neoplastic microenvironments (Conti et al., 2004; Richmond et al., 2004; Spadaro and Forni, 2004). These common features are thought to enhance leukocyte recruitment into neoplastic tissue as well as providing a potential mechanism for metastatic spread of tumor cells inappropriately expressing leukocyte adhesion and/or recruitment machinery (Sasisekharan et al., 2002).

164

Stephen C. Robinson and Lisa M. Coussens

B. Cytokines, Chemokines, and Leukocyte Recruitment Leukocyte extravasation out of the vasculature is stimulated by chemotactic signals arising from ‘‘damaged’’ tissues. These signals typically involve enhanced local production and/or activation of proinflammatory cytokines such as tumor necrosis factor-alpha (TNF-), increased bioavailability of chemokines, activation of the complement system, or deposition of immunoglobulin (Ig) (Balkwill and Mantovani, 2001; Choy and Panayi, 2001; Hanada and Yoshimura, 2002; Walport, 2001). TNF- is a multifunctional cytokine that enhances leukocyte recruitment via binding to high-affinity transmembrane TNF- receptors (TNF-R1 or 2) (Kollias and Kontoyiannis, 2002; Locksley et al., 2001) and initiates intracellular signaling cascades, leading to increased transcription of other pro-inflammatory cytokines, e.g., IL-1, IL-6, IL-8, granulocyte–monocyte colony-stimulating factor (GM-CSF) (Haworth et al., 1991; Nawroth et al., 1986), chemokines (Biedermann et al., 2000; Wright et al., 2004), and proteolytic enzymes associated with tissue remodeling (Arnott et al., 2002; Badolato and Oppenheim, 1996; Baram et al., 2001; Robinson et al., 2002; Zhou et al., 2003). Under homeostatic conditions, expression of TNF- mRNA is tightly regulated and maintained at low levels, while bioavailability of latent membrane-associated TNF- protein is regulated by the actions of various metalloproteases that cleave, or shed, it from the membrane (Amour et al., 1998; Black et al., 1997; Moss et al., 1997); thus, tissues maintain a reservoir of TNF- that can be rapidly mobilized to engage reparative machinery. Chemokines, by contrast, represent a group of 40 proteins that bind to and stimulate G-protein-coupled seven transmembrane-domain chemokine receptors expressed on leukocytes and are key for directed leukocyte migration, activation, and survival (Baggiolini and Loetscher, 2000). Chemokines are divided into four groups, depending upon their amino acid arrangement of N-terminal cysteine residues: C, CC, CXC, and CX3C chemokines (Zlotnik and Yoshie, 2000). While possessing varied target receptor specificities, all chemokines acquire a ‘‘chemokine fold,’’ adopting similar superimposable three-dimensional structures (Fernandez and Lolis, 2002). The CXC subgroup of chemokines is further classified by their unique amino acid arrangement (glutamic acid–leucine–arginine, or ELR motif) in their N-termini (Strieter et al., 1995). Those with an ELRþ motif display proangiogenic properties whereas those with ELR motifs tend to be angiostatic (Belperio et al., 2000). Regulation of chemokine function is complex and involves transcriptional and posttranslational control mechanisms. Expression of chemokine mRNA is tightly controlled in a cell-type dependent manner (Thelen, 2001). Following tissue damage, bacterial products

Soluble Mediators of Inflammation in Tumors

165

such as LPS, activation of complement components, or enhanced intracellular signaling downstream of TNF-, IL-1, or Ig induce proinflammatory chemokine mRNA expression in stromal fibroblasts, resident leukocytes, epithelial cells, and endothelial cells (Gosset et al., 1999; Luster, 2002; Proudfoot, 2002; Vicari and Caux, 2002; Wilson and Balkwill, 2002). Once produced, chemokine function is tightly regulated posttranslationally by varied interactions with soluble and insoluble ECM components and by the expression and availability of cognate chemokine receptors expressed on target cells. The majority of chemokines are basic proteins that engage ECM through electrostatic interactions of their carboxy-termini with linear sulfated glycosaminoglycans (GAGs) such as heperan-sulfate and heparin (Kuschert et al., 1999; Middleton et al., 2002; Proudfoot et al., 2003). GAGs are negatively charged complex polysaccharides with vast structural diversity that, when attached to core proteins, form large proteoglycan aggregates (Shriver et al., 2002; Trowbridge and Gallo, 2002). An advantage to maintaining high local levels of latent proinflammatory molecules is that they can be rapidly released from sequestration as well as facilitating rapid formation of haptotactic gradients directing migration of leukocytes expressing cognate chemokine receptors (Middleton et al., 2002) (Fig. 2). Chemokine–GAG interactions are an essential feature for sequestration and higher-order oligomer formation for a number of chemokines including CCL2/MCP-1, CCL3/MIP-1, and CCL5/RANTES as well as being essential for appropriate chemokine recruitment of leukocytes in vivo (Proudfoot et al., 2003). Chemokines demonstrate wide-ranging affinities for GAGs and vice versa (Kuschert et al., 1999; Witt and Lander, 1994). The predominant GAG expressed by endothelial cells is heperan sulfate (Ihrcke et al., 1993); however, chondroitin and dermatin sulfates also sequester chemokines and are commonly expressed by vascular endothelial cells (Nelimarkka et al., 1997). Variations in endothelial cell heperan sulfate and GAG structure is organ-dependent and is thought to provide critical regulatory information for appropriate organ-specific leukocyte recruitment in response to tissue damage (Marquezini et al., 1995; Mertens et al., 1992; Murch et al., 1993; Netelenbos et al., 2001; Oohira et al., 1983). Thus, it is likely that in addition to regulating appropriate leukocyte recruitment to facilitate tissue repair following acute trauma, developing neoplasms enlist alternative GAG structures to effectively recruit and maintain pro-tumorigenic leukocytes that potentiate malignant progression (Sasisekharan et al., 2002). Sequestered chemokines are released and rendered bioavailable by actions of diverse extracellular proteases, such as serine, cysteine, and metalloproteinases (MPs), commonly associated with tissue remodeling during acute as well as chronic inflammation and in developing neoplasms (Curran and

166 Table I

Chemokines Are Posttranslationally Modified by Proteasesa

Chemokine

Cleavage by protease

CCL2 (MCP-1) CCL3 (MIP-1) CCL3L1 (LD78 )

MMP-1, MMP-3 Cathepsin D DPPIV (CD26)

CCL4 (MIP-1 ) CCL5 (RANTES)

Cathepsin D DPPIV (CD26)

CCL7 (MCP-3) CCL8 (MCP-2)

MMP-1, MMP-2, MMP-3, MMP-14 MMP-1, MMP-3

CCL13 (MCP-4)

MMP-1, MMP-3

CCL21 (SLC) CXCL1 (GR0-) CXCL4 (PF-4) CXCL5 (ENA-78)

Cathepsin D MMP-9 MMP-9 MMP-8, MMP-9 Cathepsin G

CXCL6 (GCP-2)

MMP-8, MMP-9 ?

Effect Aminoterminal cleavage and loss of activity Cleavage and inactivation (rapid) Aminoterminal cleavage retention of CCR5 agonist activity 30-fold increased CCR1 agonist activity Cleavage and inactivation (rapid) Aminoterminal cleavage and generation of CCR1/3 antagonist Aminoterminal cleavage, formation of competitive antagonist with anti-inflammatory properties Aminoterminal cleavage and generation of an antagonist Aminoterminal cleavage and generation of an antagonist Cleavage and inactivation (slow) Aminoterminal cleavage (slow) and loss of activity Aminoterminal cleavage (slow) and loss of activity Carboxyterminal cleavage, unknown effect on function Aminoterminal cleavage and increased neutrophil chemotactic activity Carboxyterminal cleavage, unknown effect on function Aminoterminal cleavage and increased neutrophil chemotactic activity (mouse)

Reference McQuibban et al., 2002 Wolf et al., 2003 Proost et al., 2000

Wolf et al., 2003 Proost et al., 1998 McQuibban et al., 2000, 2002 McQuibban et al., 2002 McQuibban et al., 2002 Wolf et al., 2003 Van den Steen et al., 2000 Van den Steen et al., 2000 Van den Steen et al., 2003 Nufer et al., 1999 Van den Steen et al., 2003 Wuyts et al., 1999

CXCL7 (NAP-2)

Cathepsin G ?

CXCL8 (IL-8)

MMP-9

CXCL9 (Mig)

MMP-8, Cathepsin G (slow), Proteinase 3, Thrombin MMP-8, MMP-9 DPPIV (CD26)

CXCL10 (IP-10)

MMP-8, MMP-9 DPPIV (CD26)

CXCL11 (I-TAC)

DPPIV (CD26)

CXCL12 (SDF-1)

MMP-1, MMP-2, MMP-3, MMP-9, MMP-13, MMP-14 MMP-8

Cleavage of CTAP-III generating CXCL7 Carboxyterminal cleavage and increased neutrophil chemotactic activity Aminoterminal cleavage (fast), and over 10-fold increased activity on CXCR1 Aminoterminal cleavage and increased activity

Brandt et al., 1991 Brandt et al., 1993; Ehlert et al., 1998 Van den Steen et al., 2000 Hebert et al., 1990; Padrines et al., 1994

Carboxyterminal cleavage, unknown effect on function Aminoterminal cleavage, loss of chemotactic activity through CXCR3, and retention of angiostatic activity Carboxyterminal cleavage, unknown effect on function Aminoterminal cleavage, generation of CXCR3 antagonist, and retention of angiostatic activity Aminoterminal cleavage and generation of CXCR3 antagonist Aminoterminal cleavage and loss of activity

Van den Steen, et al., 2003

McQuibban et al., 2001

Cleavage and loss of activity

Petit et al., 2002

Proost et al., 2001 Van den Steen et al., 2003 Proost et al., 2001 Proost et al., 2001

aChemokine activity can be activated and inactivated by various proteases the consequences of which can modify inflammatory processes. Listed are chemokines and respective proteolytic enzymes demonstrated to cleave them. In addition, the consequences of chemokine modification are described.

167

168

Stephen C. Robinson and Lisa M. Coussens

Murray, 1999; Egeblad and Werb, 2002; Overall and Lopez-Otin, 2002; Sternlicht and Bergers, 2000) (Table I). Proteolytic modification of chemokines through cleavage of amino- and carboxy-termini alters chemokine structure and variably affects chemokine function. For example, matrix metalloproteinase-9 (MMP-9) cleaves six amino acids at the amino terminus of CXCL8/IL-8, generating a CXCL8/IL-8 derivative possessing greater than 10-fold higher affinity for the receptor CXCR1 and thus enabling enhanced neutrophil recruitment and further MMP-9 secretion (Van den Steen et al., 2000). Alternatively, MMP-2 inactivates CCL7/MCP-3 chemokine signaling by amino-terminal modification generating functional antagonists to respective chemokine receptors (McQuibban et al., 2000). Extracellular proteinases also regulate chemokine activity by cleaving GAG-bound chemokines, i.e., MMP-7 releases syndecan-1/mouse keratinocyte chemokine (KC/CXCL1) complexes resulting in directed neutrophil influx to sites of injury (Li et al., 2002b). Thus, MMP modification of soluble or GAG-bound chemokines can facilitate either initiation or resolution of inflammation in a microenvironment-dependent manner. Since numerous chemokines have been identified as targets for serine and metalloproteinases (Overall et al., 2002) (Table I) with disparate effects on chemokine function, the balance between secretion of specific chemokines and extracellular proteases may define the overall outcome of specific inflammatory processes.

III. INFLAMMATION AND TUMOR PROGRESSION While tumors are composed of neoplastic cells, they also contain a diverse array of activated stromal cells, including endothelial and vascular smooth muscle cells forming the blood vasculature and lymphatics, fibroblasts, and immune cells, all of which coexist in a dynamic ECM that together foster cancer development. Leukocytes compose a large percentage of the total cellular repertoire in many tumor types (Balkwill and Mantovani, 2001; Coussens and Werb, 2001; Funada et al., 2003; Hamada et al., 2002; Ishigami et al., 2003; Li et al., 2002a; Noguchi et al., 2003); thus, the inflammatory component of a developing neoplasm, much like a wound, may include a diverse leukocytic population, e.g., neutrophils, DCs, macrophages, eosinophils, MCs, and lymphocytes, all of which are capable of producing an assorted array of cytokines, cytotoxic mediators such as reactive oxygen species (ROS) (Hussain et al., 2003), serine and cysteine proteases, MMPs, (Egeblad and Werb, 2002), membrane-perforating agents, and soluble mediators of cell killing such as TNF-, interleukins, and interferons (IFNs) (Dranoff, 2004; Kuper et al., 2000; Wahl and Kleinman, 1998; Wilson and Balkwill, 2002).

Soluble Mediators of Inflammation in Tumors

169

In 1863, Rudolf Virchow hypothesized that leukocyte infiltration reflected the origin of cancer at sites of chronic inflammation (Virchow, 1863). Many experimental and epidemiological studies now support this theory (Hussain et al., 2003; Koki and Masferrer, 2002; Kuper et al., 2000; Ness and Cauley, 2004; Shacter and Weitzman, 2002; Velicer et al., 2004) and accumulating clinical data suggest that leukocytes and sustained states of inflammation influence tumor development by both pro- and antitumor mechanisms (Bingle et al., 2002; Mantovani, 1994; Mantovani et al., 1992; Ohno et al., 2002). Moreover, long-term use of anti-inflammatory drugs significantly decreases the risk of cancer-related inflammatory conditions, such as colon cancer development in patients with familial adenomatous polyposis (Thun et al., 2004), among others. The persistent presence of leukocytes can promote neoplastic progression by providing soluble mediators that influence DNA stability and/or initiate reparative programs, leading to ECM remodeling and cellular proliferation. For example, activated leukocytes provide a continued source of reactive oxygen that may initiate oxidative DNA damage in proliferating cells (Hussain et al., 2003). Furthermore, leukocytes provide mitogenic factors which promote cellular proliferation and proteases that initiate tissue remodeling and angiogenic responses in neoplastic stroma. Tumor-associated macrophages (TAMs) derived from circulating monocytes are a major component of the host leukocyte infiltrate in neoplastic tissues that are recruited primarily by monocyte chemotactic protein (MCP) chemokines (Mantovani et al., 2002). TAMs are differentially activated by various cytokines and therefore play distinctly separate roles in neoplastic environments. In experimental mouse models of carcinogenesis, activation of TAMs by IL-2, IFN, and IL-12 promotes immune responses, resulting in destruction of neoplastic cells (Satoh et al., 2003; Tsung et al., 2002). In contrast, TAMs also produce various potent proangiogenic and lymphangiogenic growth factors, cytokines, and proteases that positively induce neoplastic progression (Schoppmann et al., 2002). During development of melanoma, activated macrophages produce transforming growth factorbeta (TGF- ), TNF-, IL-1, arachidonate metabolites, and extracellular proteases (Torisu et al., 2000). In response, melanocytes express CXCL8/IL8 and vascular endothelial growth factor A (VEGF-A), thereby inducing vascular angiogenesis under paracrine control (Ono et al., 1999). Indeed, macrophage infiltration is closely associated with the depth of invasion of primary melanoma due, in part, to macrophage-regulated tumor-associated angiogenesis (Jonjic et al., 1992). Furthermore, TAM infiltration and expression of VEGF in breast cancer is associated with increased angiogenesis and poor prognosis (Leek et al., 2000). In addition to altering the local balance of proangiogenic factors in tumor development, during human cervical carcinogenesis, TAMs express lymphangiogenic factors VEGF-C

170

Stephen C. Robinson and Lisa M. Coussens

and VEGF-D as well as the VEGF receptor-3 (VEGFR-3), resulting in peritumoral lymphangiogenesis linked to lymphatic metastases (Tsung et al., 2002). TAMs also induce VCAM-1 expression on mesothelial cells, a step also believed to be key for tumor cell dissemination into the peritoneum (Jonjic et al., 1992); thus, by placing TAMs at the center of the recruitment and response to angiogenic and lymphangiogenic stimuli, they may foster spread of tumors. Macrophage recruitment to and function within sites of neoplastic growth have been experimentally assessed by crossing transgenic mice expressing Polyoma middle T (PyMT) driven by the MMTV LTR, which are prone to development of mammary cancer, with mice containing a null mutation in the CSF-1 gene (Csf1op) (Lin et al., 2001). Absence of CSF-1 expression is without apparent consequence for early neoplastic development whereas development of late-stage invasive carcinoma and pulmonary metastases is significantly attenuated. The key difference between PyMT mice and PyMT/Csf1op/Csf1op mice is not in the apparent proliferative capacity of neoplastic epithelial cells, but failure to recruit mature macrophages into neoplastic tissue in the absence of CSF-1. Restoration of CSF-1 expression, specifically to mammary epithelium in CSF-1-null/PyMT mice, restores macrophage recruitment, primary tumor development, and metastatic potential (Lin et al., 2001). Furthermore, the actions of CSF-1 may be organ specific, where a similar study showed that subcutaneous growth of Lewis lung cancer is impaired in Csf1op/Csf1op mice (Nowicki et al., 1996). In the latter example, however, tumors displayed a decreased proliferative capacity and marked necrosis, apparently resulting from diminished angiogenesis and impaired tumor-stroma formation. These defects were corrected by treatment of tumor-bearing mice with recombinant CSF-1 (Nowicki et al., 1996). Together, these data provide evidence for CSF-1-dependent recruitment of macrophages and the malignant potential of initiated epithelial cells. Chemokine regulation of macrophage recruitment was also demonstrated in a mouse model of breast cancer, where the chemokine receptor antagonist Met-RANTES was employed to block the constitutive activity of CCL5/RANTES by 410.4 mammary carcinoma cells (Robinson et al., 2003). Treatment with the inhibitor significantly reduced intratumoral macrophage recruitment, resulting in decreased tumor growth. Taken together, these data provide a link between macrophage recruitment to tumor tissue and malignant outcome and provide potential targets for cancer therapeutics. Neoplastic processes can be initiated by activated leukocytes other than macrophages. Functional experiments and genetic manipulation of mice indicate that neutrophils, MCs, eosinophils, and activated T-lymphocytes make significant contributions to malignancies by also releasing extracellular

Soluble Mediators of Inflammation in Tumors

171

proteases, proangiogenic factors, and chemokines (Bergers et al., 2000; Coussens et al., 1999, 2000; Tazawa et al., 2003). MC infiltration is associated with poor clinical outcome in cancers of the lung, where they are associated with enhanced angiogenesis (Imada et al., 2000). MCs can directly stimulate angiogenesis through autocrine and paracrine production of VEGF, basic fibroblast growth factor (bFGF), CXCL8/IL-8, TGF- , and extracellular protease production (Coussens et al., 1999, 2000; Norrby, 2002). The importance of MC infiltration into early neoplastic tissue has been demonstrated experimentally in a mouse model of skin carcinogenesis, for example, K14-HPV16 (Coussens et al., 1999). MCs infiltrate premalignant hyperplastic, dysplastic, and invasive fronts of carcinomas, where they degranulate in close apposition to capillaries and epithelial basement membranes, releasing mast cell-specific serine proteases MCP-4 (chymase) and MCP-6 (tryptase) and MMP-9. Notably, premalignant angiogenesis was abated in MCdeficient HPV16 transgenic mice, suggesting that neoplastic progression in this model involves exploitation of an inflammatory response to tissue abnormality. Taken together, these data demonstrate promoting effects of leukocytes during cancer development. The mechanism(s) through which leukocytes are recruited and activated are complex; however, chemokine regulation of leukocyte recruitment is an attractive target for regulating this process.

A. TNF-a and Cancer Development The production of cytokines in neoplastic microenvironments can significantly enhance inflammatory responses and the cytokine TNF- demonstrates tumor-promoting effects (Balkwill, 2002). Using the classical model of twostage experimental skin carcinogenesis, TNF- has been identified as a key participant in neoplastic progression of 9,10-dimethyl-1,2-benzanthracene (DMBA) initiated and 12-O-tetradecanoylphorbol-13-acetate (TPA) promoted keratinocytes (Moore et al., 1999; Scott et al., 2003). TNF--deficient mice, or mice treated with neutralizing antibodies to TNF-, when exposed to DMBA and TPA, are resistant to skin carcinogenesis (Moore et al., 1999; Scott et al., 2003). TNF- mediates the actions of these two-stage carcinogens through induction of proinflammatory cytokines such as GM-CSF in epithelial keratinocytes as well as other AP-1 responsive genes such as MMP9 and MMP-3 (Arnott et al., 2002) in stromal cells, both downstream events requiring expression of TNF receptor-1 and -2 (Arnott et al., 2004). In addition to autocrine actions of TNF- on keratinocytes, inflammation and particularly neutrophil infiltration are significantly diminished in TNF-/ mice, suggesting that TNF- mediated inflammation may play a crucial role in later stages of neoplastic progression as well (Moore et al., 1999).

172

Stephen C. Robinson and Lisa M. Coussens

In several human cancers, such as mammary carcinomas, bone sarcomas, myelomas, and lymphomas, production of TNF- is associated with poor prognosis (Leek et al., 1998; Rutkowski et al., 2003; Tsimberidou et al., 2003; Warzocha et al., 1997), suggesting that local production of TNF- into the local neoplastic environment may promote tumor development through proliferative effects on epithelial and stromal cells and enhance inflammation through regulated chemokine production and subsequent leukocyte recruitment. Conversely, when given at high doses as an anticancer therapeutic, TNF- is an effective anticancer agent for soft tissue sarcomas where TNF- is delivered by isolated limb perfusion at 50 times the tolerated dose and results in apoptosis of tumor endothelial cells (Eggermont et al., 1996; Lejeune et al., 1998; van der Veen et al., 2000). Taken together, these data imply that while high levels of TNF- production in the microenvironment may aid tumor destruction, the physiological concentrations detected in neoplastic tissues may be significantly lower, allowing tumor-promoting effects of TNF- to prevail.

B. Chemokines Promote Neoplastic Progression The chemokine gene family, while initially defined as soluble factors regulating directional leukocyte migration during states of acute inflammation (Walz et al., 1987; Yoshimura et al., 1987), are now regarded as potent regulators of cancer development (Murphy, 2001; Rossi and Zlotnik, 2000; Vicari and Caux, 2002). Initial interest for chemokine involvement during malignancy came from reports that immune-deficient animals, when challenged with tumor cell lines, effectively recruited innate leukocytes, suggesting that neoplastic cells either produce chemotactic factors or induce their expression in activated cells (Richmond and Thomas, 1986). It is now appreciated that the chemokine-receptor system can be dramatically modified in neoplastic tissue, particularly at the invasive edges of tumors (Richmond et al., 2004; Vicari and Caux, 2002; Wilson and Balkwill, 2002), where maintaining leukocyte presence favors malignant development by modulating stromal and neoplastic cell behavior. For example, chemokine receptors can be expressed on neoplastic cells such as CXCR2 expression by melanoma cells (Norgauer et al., 1996); significantly blocking CXCL1/GRO or its receptor CXCR2 inhibits melanoma cell growth in vitro (Norgauer et al., 1996). The chemokines CXCL1/GRO, CXCL2/ GRO , CXCL3/GRO , CXCL8/IL-8, and CXCL12/SDF-1 are also all directly involved in tumor growth (Owen et al., 1997; Scotton et al., 2002). CXCL1/GRO, originally purified from human malignant melanoma-conditioned medium and characterized as an autocrine growth factor (Bordoni et al., 1990; Richmond and Thomas, 1986), when overexpressed,

Soluble Mediators of Inflammation in Tumors

173

increases the ability of melanoma cells, expressing the appropriate chemokine receptors, to form colonies on soft agar and their tumorigenicity in nude mice (Balentien et al., 1991). Similar observations have been reported for CXCL2/GRO and CXCL3/GRO (Owen et al., 1997). Viruses implicated in cancer development also exploit the power of chemokine function by expressing chemokine receptors (Yang et al., 2000). An example of this is Kaposi’s sarcoma (KS) herpesvirus G-protein coupled receptor (KSHV-GPCR), a product of HHV8, which has a direct role in neoplastic transformation of infected cells. Overexpression of KSHV-GPCR in hematopoietic cells alone results in KS-type lesions (Yang et al., 2000). KSHV-GPCR is unique among chemokine receptors in being constitutively active (Bais et al., 1998). Chemokines such as CXCL1/GRO and CXCL8/ IL-8 increase the constitutive activity of KSHV-GPCR (Bais et al., 1998; Gershengorn et al., 1998). Mutated forms of CXCR2 receptor that are constitutively active are also capable of inducing cellular transformation similar to KSHV-GPCR (Burger et al., 1999). Together, these data imply that in chemokine-rich environments, such as is found in neoplastic tissues, initiated cells co-opt mechanisms utilized by leukocytes during inflammation to promote their growth and survival.

C. Altered Regulation of Chemokines in Cancer In premalignant tissues, ‘‘normal’’ communication between resident cells is altered and favors molecular and cellular pathways maintaining leukocyte presence (Conti et al., 2004; Coussens et al., 1999; Lin et al., 2001; Negus et al., 1997; Ordemann et al., 2002). Constitutive CC chemokine expression by many neoplastic cells, e.g., CCL5/RANTES expression in breast cancer (Azenshtein et al., 2002) and CCL1/5 (I-309/RANTES) in ovarian cancers, favors sustained macrophage and lymphocyte recruitment and is often associated with poor prognosis (Azenshtein et al., 2002; Luboshits et al., 1999; Mantovani, 1994; Negus et al., 1997). Macrophages possess phenotypic plasticity that is dependent upon activation by distinct sets of cytokines, chemokines, and prostaglandins at sites of inflammation. For example, IFN- -stimulated alternatively activated (M1) macrophages are associated with IL-12 and INF- production in addition to the chemokines CXCL9/MIG and CXCL10/IP-10 that stimulate cytotoxic (Th1) cellmediated immunity (Dixon et al., 2000). In contrast, M2 macrophages stimulated by IL-4 and IL-13 produce IL-10 (Kambayashi et al., 1995; Sica et al., 2000), TGF- (Maeda et al., 1995), and Th2 recruiting chemokines, such as CCL22/MDC (Bonecchi et al., 1998), that act to diminish lymphocyte-mediated immunity. In addition to innate leukocyte recruitment, chemokine expression in solid tumors often results in immune subversion

174

Stephen C. Robinson and Lisa M. Coussens

(reviewed by Balkwill and Mantovani, 2001; Coussens and Werb, 2002); thus, chemokines exert divergent effects on recruitment and regulation of different leukocyte populations, suggesting that understanding chemokine regulation may allow development of chemokine antagonists that promote antitumoral responses.

D. Chemokines and Angiogenesis It is well established that chemokines directly regulate angiogenesis through pro-angiogenic ELRþ chemokines that stimulate endothelial cell chemotaxis (Belperio et al., 2000), while ELR CXC chemokines, e.g., CXCL4/PF-4, CXCL9/ MIG, and CXCL10/IP-10, possess angiostatic activities (Strieter et al., 1995). ELRþ CXC ligands stimulate CXCR2 expressed on endothelial cells and, to a lesser degree, to CXCR1 (Keane et al., 2004). CXCL8/IL-8 expression and CXCL5/ENA-78 are elevated in bronchogenic and small cell lung cancers, respectively, and both are associated with increased vascular density (Arenberg et al., 1998; Smith et al., 1994). In contrast, ELR CXC ligands bind to CXCR3, inhibiting endothelial cell migration and angiogenesis (Strieter et al., 2004). Tumors expressing high levels of CXCL9/Mig or CXCL10/IP-10 are both associated with tumor regression (Sgadari et al., 1996, 1997) and overexpression in experimental tumors results in increased necrosis and vascular damage (Teruya-Feldstein et al., 1997). The role of CXCL12/SDF-1 and its receptor CXCR4 in angiogenesis is unclear. CXCL12/SDF-1 is an ELR CXC chemokine and has been demonstrated to attenuate the angiogenic activity of ELRþ CXC chemokines, bFGF, and VEGF (Arenberg et al., 1997). Conversely, CXCL12/SDF-1 induces the migration of human umbilical vein endothelial cells, and disruption of CXCR4 demonstrates its role for vascularization of the gastrointestinal tract (Tachibana et al., 1998) and cardiac and cerebellar development (Shalinsky et al., 1998). Chemokines can also influence angiogenesis indirectly through leukocyte recruitment that provides pro-angiogenic molecules during inflammation (Cursiefen et al., 2004; Sunderkotter et al., 1994) and neoplastic progression (Leek et al., 2000). Moreover, leukocytes provide extracellular proteases such as MMP-9 (Coussens et al., 2000; Ribatti et al., 2001) that release growth factors from the matrix to promote angiogenesis (Bergers et al., 2000; Coussens and Werb, 2002) or remodel ECM proteins and generate pro- and anti-angiogenic fragments (reviewed by Sottile, 2004). While it is not always clear whether the angiostatic and angiogenic effects of chemokines are direct or indirect, it is clear that the balance between the two regulates vascular cell physiology and overall neoplastic development.

Soluble Mediators of Inflammation in Tumors

175

E. Chemokines and Metastasis Tumor metastases most likely develop because of selective interactions between neoplastic cells in favorable microenvironments. Malignant cells that possess metastatic capacity have properties endowing them with the ability to invade and survive in ectopic tissue, venous and/or lymphatic environments, as well as ability to reside and proliferate at a distal site. A number of hypotheses have been proposed to explain the organ-specific metastasis of malignant cells and include the idea that metastatic cells either settle in organ microenvironments that support their growth or, alternatively, are specifically recruited to target organs through chemotactic factors (Fidler, 2003; Hanahan and Weinberg, 2000). CXCL12/SDF-1, for example, is an essential chemokine for leukocyte homing to multiple organs and is therefore unique in that it is the product of resting cells in multiple tissue types (Rossi and Zlotnik, 2000); however, it is highly expressed in target organs where breast cancer metastases are common, e.g., lymph nodes and bone marrow (Muller et al., 2001). In addition to breast cancer, CXCL12/ SDF-1 directs chemotaxis of a number of malignant cells in vitro (Moore, 2001; Scotton et al., 2001), and the induction of chemotaxis by extracts of organs targeted by breast cancer cells (bone marrow, liver, lung, lymph nodes) can be neutralized by anti-CXCR4 antibodies (Muller et al., 2001) or the synthetic CXCR4 inhibitor AMD3100 (Scotton et al., 2002). These data suggest that the distinct patterns of breast cancer metastasis are, in part, governed by exclusive interactions between CXCR4 expressed on metastatic cells and its ligand CXCL12/SDF-1 expressed in recipient organs (Muller et al., 2001). How do neoplastic cells upregulate CXCR4 expression? Evidence suggests that local microenvironmental factors such as hypoxia increase CXCR4 expression by leukocytes, endothelial, and neoplastic cells, providing evidence for at least one mechanism permitting CXCR4 expression and cancer cell metastasis (Schioppa et al., 2003; Staller et al., 2003). The mechanism of neoplastic cell metastasis through CXCR4dependent migration cannot be easily explained when malignant cells express high levels of CXCL12/SDF-1, such as are observed in human ovarian cancer (Scotton et al., 2002). One explanation for the ability of metastatic cells to escape high local chemokine concentrations may be explained by the mechanism of hematopoietic stem cells’ release from bone marrow and their expression of CXCR4 (Fruehauf and Seggewiss, 2003), where MMP cleavage of CXCL12/SDF-1 effectively permits stem cell release into circulation (Petit et al., 2002). A situation similar to the release of hematopoietic stem cells could be conceived in tumors where expression of extracellular proteases deactivates stromal- or neoplastic-cell derived CXCL12/SDF-1 (Wolf et al., 2003), enabling metastatic cells to respond

176

Stephen C. Robinson and Lisa M. Coussens

to chemokine gradients produced at sites of metastasis. The broader implications of this data are that chemokines may direct the spectrum of metastases in diverse cancer types and that posttranslational modification of those chemokines may add an additional level of control.

IV. CONCLUSIONS Experimental advances now suggest that by understanding key regulatory mechanisms utilized for acute activation and deactivation of inflammatory cascades, as compared to persistent inflammation, in chronic disease and/or in neoplastic tissues, novel targets will be identified that will aid in development of effective anticancer therapeutics. Advances in the development of chemokine antagonists, primarily for targeting human immunodeficiency virus (HIV) entry into leukocytes via chemokine co-receptors and inhibiting leukocyte infiltration into sites of inflammation (Proudfoot, 2002), have also proved effective in reducing both the migration of activated leukocytes into neoplastic tissues (Robinson et al., 2003) and tumor cell survival through CXCR4 stimulation (Rubin et al., 2003). Interactions of leukocytes with ECM and their effective removal, leading to regression of inflammation, suggests that targeting leukocyte-ECM interactions may prove useful for inhibiting leukocytes’ infiltration into neoplastic tissues. Furthermore, strategies adopted by neoplastic cells to utilize inflamed endothelium for hematogenous metastasis has promoted interest in using antagonists originally designed as anti-inflammatory agents to prevent leukocyte rolling (Zacharski, 2003). Apart from their functional role in regulating leukocyte migration into inflamed and neoplastic tissue, chemokines have direct and indirect roles in regulating angiogenesis. The angiogenic component of a tumor is an attractive target for cancer therapies, and emerging data from clinical trials of drugs that target angiogenic cytokines prove that this is an effective form of therapy for some cancers (Hurwitz, 2004). Leukocytes provide a plethora of cytokines, chemokines, and extracellular proteinases that modify soluble and insoluble stromal components of tumors regulating angiogenesis. Understanding these complex interactions is necessary for manipulating the tumor microenvironment, altering the balance in favor anti-angiogenic ECM molecule and chemokine production, while preventing the release of pro-angiogenic molecules from the ECM and promoting inactivation of chemokines responsible for antitumor immunity. Thus, targeting angiogenic chemokines or extracellular proteases that regulate angiogenic processes may provide additional strategies for preventing neoplastic proliferation and tumor growth.

Soluble Mediators of Inflammation in Tumors

177

REFERENCES Amour, A., Slocombe, P. M., Webster, A., Butler, M., Knight, C. G., Smith, B. J., Stephens, P. E., Shelley, C., Hutton, M., Knauper, V., Docherty, A. J., and Murphy, G. (1998). TNF-alpha converting enzyme (TACE) is inhibited by TIMP-3. FEBS Lett. 435, 39–44. Arenberg, D. A., Keane, M. P., DiGiovine, B., Kunkel, S. L., Morris, S. B., Xue, Y. Y., Burdick, M. D., Glass, M. C., Iannettoni, M. D., and Strieter, R. M. (1998). Epithelial-neutrophil activating peptide (ENA-78) is an important angiogenic factor in non-small cell lung cancer. J. Clin. Invest. 102, 465–472. Arenberg, D. A., Polverini, P. J., Kunkel, S. L., Shanafelt, A., and Strieter, R. M. (1997). In vitro and in vivo systems to assess role of C-X-C chemokines in regulation of angiogenesis. Methods Enzymol. 288, 190–220. Arnott, C. H., Scott, K. A., Moore, R. J., Hewer, A., Phillips, D. H., Parker, P., Balkwill, F. R., and Owens, D. M. (2002). Tumour necrosis factor-alpha mediates tumour promotion via a PKC alpha- and AP-1-dependent pathway. Oncogene 21, 4728–4738. Arnott, C. H., Scott, K. A., Moore, R. J., Robinson, S. C., Thompson, R. G., and Balkwill, F. R. (2004). Expression of both TNF-alpha receptor subtypes is essential for optimal skin tumour development. Oncogene 23, 1902–1910. Auger, M. J., and Roos, J. A. (1992). In ‘‘The Macrophage’’ (C. E. Lewis and J. O. McGee, Eds.), pp. 3–74. University Press, Oxford, UK. Azenshtein, E., Luboshits, G., Shina, S., Neumark, E., Shahbazian, D., Weil, M., Wigler, N., Keydar, I., and Ben-Baruch, A. (2002). The CC chemokine RANTES in breast carcinoma progression: Regulation of expression and potential mechanisms of promalignant activity. Cancer Res. 62, 1093–1102. Badolato, R., and Oppenheim, J. J. (1996). Role of cytokines, acute-phase proteins, and chemokines in the progression of rheumatoid arthritis. Semin. Arthritis Rheum. 26, 526–538. Baggiolini, M., and Loetscher, P. (2000). Chemokines in inflammation and immunity. Immunol. Today 21, 418–420. Bais, C., Santomasso, B., Coso, O., Arvanitakis, L., Raaka, E. G., Gutkind, J. S., Asch, A. S., Cesarman, E., Gershengorn, M. C., Mesri, E. A., and Gerhengorn, M. C. (1998). G-proteincoupled receptor of Kaposi’s sarcoma-associated herpesvirus is a viral oncogene and angiogenesis activator. Nature 391, 86–89. Balentien, E., Mufson, B. E., Shattuck, R. L., Derynck, R., and Richmond, A. (1991). Effects of MGSA/GRO alpha on melanocyte transformation. Oncogene 6, 1115–1124. Balkwill, F. (2002). Tumor necrosis factor or tumor promoting factor? Cytokine Growth Factor Rev. 13, 135–141. Balkwill, F., and Mantovani, A. (2001). Inflammation and cancer: Back to Virchow? Lancet 357, 539–545. Baram, D., Vaday, G. G., Salamon, P., Drucker, I., Hershkoviz, R., and Mekori, Y. A. (2001). Human mast cells release metalloproteinase-9 on contact with activated t cells: Juxtacrine regulation by tnf-alpha. J. Immunol. 167, 4008–4016. Belperio, J. A., Keane, M. P., Arenberg, D. A., Addison, C. L., Ehlert, J. E., Burdick, M. D., and Strieter, R. M. (2000). CXC chemokines in angiogenesis. J. Leukoc. Biol. 68, 1–8. Bergers, G., Brekken, R., McMahon, G., Vu, T. H., Itoh, T., Tamaki, K., Tanzawa, K., Thorpe, P., Itohara, S., Werb, Z., and Hanahan, D. (2000). Matrix metalloproteinase-9 triggers the angiogenic switch during carcinogenesis. Nat. Cell Biol. 2, 737–744. Biedermann, T., Kneilling, M., Mailhammer, R., Maier, K., Sander, C. A., Kollias, G., Kunkel, S. L., Hultner, L., and Rocken, M. (2000). Mast cells control neutrophil recruitment during

178

Stephen C. Robinson and Lisa M. Coussens

T cell-mediated delayed-type hypersensitivity reactions through tumor necrosis factor and macrophage inflammatory protein 2. J. Exp. Med. 192, 1441–1452. Bingle, L., Brown, N. J., and Lewis, C. E. (2002). The role of tumour-associated macrophages in tumour progression: Implications for new anticancer therapies. J. Pathol. 196, 254–265. Bissell, M. J., and Radisky, D. (2001). Putting tumours in context. Nat. Rev. Cancer 1, 46–54. Black, R. A., Rauch, C. T., Kozlosky, C. J., Peschon, J. J., Slack, J. L., Wolfson, M. F., Castner, B. J., Stocking, K. L., Reddy, P., Srinivasan, S., Nelson, N., Boiani, N., Schooley, K. A., Gerhart, M., Davis, R., Fitzner, J. N., Johnson, R. S., Paxton, R. J., March, C. J., and Cerretti, D. P. (1997). A metalloproteinase disintegrin that releases tumour-necrosis factoralpha from cells. Nature 385, 729–733. Bonecchi, R., Sozzani, S., Stine, J. T., Luini, W., D’Amico, G., Allavena, P., Chantry, D., and Mantovani, A. (1998). Divergent effects of interleukin-4 and interferon-gamma on macrophage-derived chemokine production: An amplification circuit of polarized T helper 2 responses. Blood 92, 2668–2671. Bordoni, R., Fine, R., Murray, D., and Richmond, A. (1990). Characterization of the role of melanoma growth stimulatory activity (MGSA) in the growth of normal melanocytes, nevocytes, and malignant melanocytes. J. Cell Biochem. 44, 207–219. Brandt, E., Petersen, F., and Flad, H. D. (1993). A novel molecular variant of the neutrophilactivating peptide NAP-2 with enhanced biological activity is truncated at the C-terminus: Identification by antibodies with defined epitope specificity. Mol. Immunol. 30, 979–991. Brandt, E., Van Damme, J., and Flad, H. D. (1991). Neutrophils can generate their activator neutrophil-activating peptide 2 by proteolytic cleavage of platelet-derived connective tissueactivating peptide III. Cytokine 3, 311–321. Burger, M., Burger, J. A., Hoch, R. C., Oades, Z., Takamori, H., and Schraufstatter, I. U. (1999). Point mutation causing constitutive signaling of CXCR2 leads to transforming activity similar to Kaposi’s sarcoma herpesvirus-G protein-coupled receptor. J. Immunol. 163, 2017–2022. Chettibi, S., and Ferguson, M. W. J. (1999). Wound repair: An overview. In ‘‘Inflammation: Basic Principles and Clinical Correlates’’ (J. I. Gallin and R. Snyderman, Eds.), pp. 865–881. Lipincott, Williams and Wilkinson, Philadelphia. Choy, E. H., and Panayi, G. S. (2001). Cytokine pathways and joint inflammation in rheumatoid arthritis. N. Engl. J. Med. 344, 907–916. Cinamon, G., Shinder, V., and Alon, R. (2001). Shear forces promote lymphocyte migration across vascular endothelium bearing apical chemokines. Nat. Immunol. 2, 515–522. Conti, I., Dube, C., and Rollins, B. J. (2004). Chemokine-based pathogenetic mechanisms in cancer. Novartis Found. Symp. 256, 29–41; discussion 41–52, 266–269. Coussens, L. M., Raymond, W. W., Bergers, G., Laig-Webster, M., Behrendtsen, O., Werb, Z., Caughey, G. H., and Hanahan, D. (1999). Inflammatory mast cells up-regulate angiogenesis during squamous epithelial carcinogenesis. Genes Dev. 13, 1382–1397. Coussens, L. M., Tinkle, C. L., Hanahan, D., and Werb, Z. (2000). MMP-9 supplied by bone marrow-derived cells contributes to skin carcinogenesis. Cell 103, 481–490. Coussens, L. M., and Werb, Z. (2001). Inflammatory cells and cancer: Think different! J. Exp. Med. 193, F23–F26. Coussens, L. M., and Werb, Z. (2002). Inflammation and cancer. Nature 420, 860–867. Curran, S., and Murray, G. I. (1999). Matrix metalloproteinases in tumour invasion and metastasis. J. Pathol. 189, 300–308. Cursiefen, C., Chen, L., Borges, L. P., Jackson, D., Cao, J., Radziejewski, C., D’Amore, P. A., Dana, M. R., Wiegand, S. J., and Streilein, J. W. (2004). VEGF-A stimulates lymphangiogenesis and hemangiogenesis in inflammatory neovascularization via macrophage recruitment. J. Clin. Invest. 113, 1040–1050.

Soluble Mediators of Inflammation in Tumors

179

Dixon, A. E., Mandac, J. B., Madtes, D. K., Martin, P. J., and Clark, J. G. (2000). Chemokine expression in Th1 cell-induced lung injury: Prominence of IFN-gamma-inducible chemokines. Am. J. Physiol. Lung Cell Mol. Physiol. 279, L592–L599. Dranoff, G. (2004). Cytokines in cancer pathogenesis and cancer therapy. Nat. Rev. Cancer 4, 11–22. Egeblad, M., and Werb, Z. (2002). New functions for the matrix metalloproteinases in cancer progression. Nat. Rev. Cancer 2, 161–174. Eggermont, A. M., Schraffordt Koops, H., Lienard, D., Kroon, B. B., van Geel, A. N., Hoekstra, H. J., and Lejeune, F. J. (1996). Isolated limb perfusion with high-dose tumor necrosis factor-alpha in combination with interferon-gamma and melphalan for nonresectable extremity soft tissue sarcomas: A multicenter trial. J. Clin. Oncol. 14, 2653–2665. Ehlert, J. E., Gerdes, J., Flad, H. D., and Brandt, E. (1998). Novel C-terminally truncated isoforms of the CXC chemokine beta-thromboglobulin and their impact on neutrophil functions. J. Immunol. 161, 4975–4982. Fernandez, E. J., and Lolis, E. (2002). Structure, function, and inhibition of chemokines. Annu. Rev. Pharmacol. Toxicol. 42, 469–499. Fidler, I. J. (2003). Timeline: The pathogenesis of cancer metastasis: The ‘seed and soil’ hypothesis Revisited. Nat. Rev. Cancer 3, 453–458. Fruehauf, S., and Seggewiss, R. (2003). It’s moving day: Factors affecting peripheral blood stem cell mobilization and strategies for improvement. Br. J. Haematol. 122, 360–375. Funada, Y., Noguchi, T., Kikuchi, R., Takeno, S., Uchida, Y., and Gabbert, H. E. (2003). Prognostic significance of CD8þ T cell and macrophage peritumoral infiltration in colorectal cancer. Oncol. Rep. 10, 309–313. Galli, S. J., and Kitamura, Y. (1987). Genetically mast-cell-deficient W/Wv and Sl/Sld mice. Their value for the analysis of the roles of mast cells in biologic responses in vivo. Am. J. Pathol. 127, 191–198. Gershengorn, M. C., Geras-Raaka, E., Varma, A., and Clark-Lewis, I. (1998). Chemokines activate Kaposi’s sarcoma-associated herpesvirus G protein-coupled receptor in mammalian cells in culture. J. Clin. Invest. 102, 1469–1472. Gosset, P., Tillie-Leblond, I., Oudin, S., Parmentier, O., Wallaert, B., Joseph, M., and Tonnel, A. B. (1999). Production of chemokines and proinflammatory and antiinflammatory cytokines by human alveolar macrophages activated by IgE receptors. J. Allergy Clin. Immunol. 103, 289–297. Gotsch, U., Jager, U., Dominis, M., and Vestweber, D. (1994). Expression of P-selectin on endothelial cells is upregulated by LPS and TNF-alpha in vivo. Cell Adhes. Commun. 2, 7–14. Hamada, I., Kato, M., Yamasaki, T., Iwabuchi, K., Watanabe, T., Yamada, T., Itoyama, S., Ito, H., and Okada, K. (2002). Clinical effects of tumor-associated macrophages and dendritic cells on renal cell carcinoma. Anticancer Res. 22, 4281–4284. Hanada, T., and Yoshimura, A. (2002). Regulation of cytokine signaling and inflammation. Cytokine Growth Factor Rev. 13, 413–421. Hanahan, D., and Weinberg, R. A. (2000). The hallmarks of cancer. Cell 100, 57–70. Haworth, C., Brennan, F. M., Chantry, D., Turner, M., Maini, R. N., and Feldmann, M. (1991). Expression of granulocyte-macrophage colony-stimulating factor in rheumatoid arthritis: Regulation by tumor necrosis factor-alpha. Eur. J. Immunol. 21, 2575–2579. Hebert, C. A., Luscinskas, F. W., Kiely, J. M., Luis, E. A., Darbonne, W. C., Bennett, G. L., Liu, C. C., Obin, M. S., Gimbrone, M. A., Jr., and Baker, J. B. (1990). Endothelial and leukocyte forms of IL-8. Conversion by thrombin and interactions with neutrophils. J. Immunol. 145, 3033–3040. Homey, B., Muller, A., and Zlotnik, A. (2002). Chemokines: Agents for the immunotherapy of cancer? Nat. Rev. Immunol. 2, 175–184.

180

Stephen C. Robinson and Lisa M. Coussens

Hurwitz, H. I. (2004). Introduction: Targeting angiogenesis in cancer therapy. Oncologist 9 (Suppl. 1), 1. Hussain, S. P., Hofseth, L. J., and Harris, C. C. (2003). Radical causes of cancer. Nat. Rev. Cancer 3, 275–286. Ihrcke, N. S., Wrenshall, L. E., Lindman, B. J., and Platt, J. L. (1993). Role of heparan sulfate in immune system-blood vessel interactions. Immunol. Today 14, 500–505. Imada, A., Shijubo, N., Kojima, H., and Abe, S. (2000). Mast cells correlate with angiogenesis and poor outcome in stage I lung adenocarcinoma. Eur. Respir. J. 15, 1087–1093. Ishigami, S., Natsugoe, S., Tokuda, K., Nakajo, A., Okumura, H., Matsumoto, M., Miyazono, F., Hokita, S., and Aikou, T. (2003). Tumor-associated macrophage (TAM) infiltration in gastric cancer. Anticancer Res. 23, 4079–4083. Janeway, C. A., Travers, P., Walport, M., and Shlomchik, M. (2001). Immunobiology. Garland Publishing, New York and London. Jonjic, N., Peri, G., Bernasconi, S., Sciacca, F. L., Colotta, F., Pelicci, P., Lanfrancone, L., and Mantovani, A. (1992). Expression of adhesion molecules and chemotactic cytokines in cultured human mesothelial cells. J. Exp. Med. 176, 1165–1174. Kambayashi, T., Alexander, H. R., Fong, M., and Strassmann, G. (1995). Potential involvement of IL-10 in suppressing tumor-associated macrophages. Colon-26-derived prostaglandin E2 inhibits TNF-alpha release via a mechanism involving IL-10. J. Immunol. 154, 3383–3390. Keane, M. P., Belperio, J. A., Xue, Y. Y., Burdick, M. D., and Strieter, R. M. (2004). Depletion of CXCR2 inhibits tumor growth and angiogenesis in a murine model of lung cancer. J. Immunol. 172, 2853–2860. Koki, A. T., and Masferrer, J. L. (2002). Celecoxib: A Specific COX-2 Inhibitor With Anticancer Properties. Cancer Control 9, 28–35. Kollias, G., and Kontoyiannis, D. (2002). Role of TNF/TNFR in autoimmunity: Specific TNF receptor blockade may be advantageous to anti-TNF treatments. Cytokine Growth Factor Rev. 13, 315–321. Kubes, P. (2002). The complexities of leukocyte recruitment. Semin. Immunol. 14, 65–72. Kuper, H., Adami, H. O., and Trichopoulos, D. (2000). Infections as a major preventable cause of human cancer. J. Intern. Med. 248, 171–183. Kuschert, G. S., Coulin, F., Power, C. A., Proudfoot, A. E., Hubbard, R. E., Hoogewerf, A. J., and Wells, T. N. (1999). Glycosaminoglycans interact selectively with chemokines and modulate receptor binding and cellular responses. Biochemistry 38, 12959–12968. Leek, R. D., Hunt, N. C., Landers, R. J., Lewis, C. E., Royds, J. A., and Harris, A. L. (2000). Macrophage infiltration is associated with VEGF and EGFR expression in breast cancer. J. Pathol. 190, 430–436. Leek, R. D., Landers, R., Fox, S. B., Ng, F., Harris, A. L., and Lewis, C. E. (1998). Association of tumour necrosis factor alpha and its receptors with thymidine phosphorylase expression in invasive breast carcinoma. Br. J. Cancer 77, 2246–2251. Lejeune, F. J., Ruegg, C., and Lienard, D. (1998). Clinical applications of TNF-alpha in cancer. Curr. Opin. Immunol. 10, 573–580. Ley, K. (2003). The role of selectins in inflammation and disease. Trends Mol. Med. 9, 263–268. Li, C., Shintani, S., Terakado, N., Nakashiro, K., and Hamakawa, H. (2002a). Infiltration of tumor-associated macrophages in human oral squamous cell carcinoma. Oncol. Rep. 9, 1219–1223. Li, Q., Park, P. W., Wilson, C. L., and Parks, W. C. (2002b). Matrilysin shedding of syndecan-1 regulates chemokine mobilization and transepithelial efflux of neutrophils in acute lung injury. Cell 111, 635–646. Lin, E. Y., Nguyen, A. V., Russell, R. G., and Pollard, J. W. (2001). Colony-stimulating factor 1 promotes progression of mammary tumors to malignancy. J. Exp. Med. 193, 727–740.

Soluble Mediators of Inflammation in Tumors

181

Locksley, R. M., Killeen, N., and Lenardo, M. J. (2001). The TNF and TNF receptor superfamilies: Integrating mammalian biology. Cell 104, 487–501. Luboshits, G., Shina, S., Kaplan, O., Engelberg, S., Nass, D., Lifshitz-Mercer, B., Chaitchik, S., Keydar, I., and Ben-Baruch, A. (1999). Elevated expression of the CC chemokine regulated on activation, normal T cell expressed and secreted (RANTES) in advanced breast carcinoma. Cancer Res. 59, 4681–4687. Luster, A. D. (2002). The role of chemokines in linking innate and adaptive immunity. Curr. Opin. Immunol. 14, 129–135. Maeda, H., Kuwahara, H., Ichimura, Y., Ohtsuki, M., Kurakata, S., and Shiraishi, A. (1995). TGF-beta enhances macrophage ability to produce IL-10 in normal and tumor-bearing mice. J. Immunol. 155, 4926–4932. Mantovani, A. (1994). Tumor-associated macrophages in neoplastic progression: A paradigm for the in vivo function of chemokines. Lab. Invest. 71, 5–16. Mantovani, A., Bottazzi, B., Colotta, F., Sozzani, S., and Ruco, L. (1992). The origin and function of tumor-associated macrophages. Immunol. Today 13, 265–270. Mantovani, A., Sozzani, S., Locati, M., Allavena, P., and Sica, A. (2002). Macrophage polarization: Tumor-associated macrophages as a paradigm for polarized M2 mononuclear phagocytes. Trends Immunol. 23, 549–555. Marquezini, M. V., Strunz, C. M., Dallan, L. A., and Toledo, O. M. (1995). Glycosaminoglycan distribution in atherosclerotic saphenous vein grafts. Cardiology 86, 143–146. McQuibban, G. A., Butler, G. S., Gong, J. H., Bendall, L., Power, C., Clark-Lewis, I., and Overall, C. M. (2001). Matrix metalloproteinase activity inactivates the CXC chemokine stromal cell-derived factor-1. J. Biol. Chem. 276, 43503–43508. McQuibban, G. A., Gong, J. H., Tam, E. M., McCulloch, C. A., Clark-Lewis, I., and Overall, C. M. (2000). Inflammation dampened by gelatinase A cleavage of monocyte chemoattractant protein-3. Science 289, 1202–1206. McQuibban, G. A., Gong, J. H., Wong, J. P., Wallace, J. L., Clark-Lewis, I., and Overall, C. M. (2002). Matrix metalloproteinase processing of monocyte chemoattractant proteins generates CC chemokine receptor antagonists with anti-inflammatory properties in vivo. Blood 100, 1160–1167. Meager, A. (1999). Cytokine regulation of cellular adhesion molecule expression in inflammation. Cytokine Growth Factor Rev. 10, 27–39. Mertens, G., Cassiman, J. J., Van den Berghe, H., Vermylen, J., and David, G. (1992). Cell surface heparan sulfate proteoglycans from human vascular endothelial cells. Core protein characterization and antithrombin III binding properties. J. Biol. Chem. 267, 20435–20443. Middleton, J., Patterson, A. M., Gardner, L., Schmutz, C., and Ashton, B. A. (2002). Leukocyte extravasation: Chemokine transport and presentation by the endothelium. Blood 100, 3853–3860. Moore, M. A. (2001). The role of chemoattraction in cancer metastases. Bioessays 23, 674–676. Moore, R. J., Owens, D. M., Stamp, G., Arnott, C., Burke, F., East, N., Holdsworth, H., Turner, L., Rollins, B., Pasparakis, M., Kollias, G., and Balkwill, F. (1999). Mice deficient in tumor necrosis factor-alpha are resistant to skin carcinogenesis. Nat. Med. 5, 828–831. Moss, M. L., Jin, S. L., Milla, M. E., Bickett, D. M., Burkhart, W., Carter, H. L., Chen, W. J., Clay, W. C., Didsbury, J. R., Hassler, D., Hoffman, C. R., Kost, T. A., Lambert, M. H., Leesnitzer, M. A., McCauley, P., McGeehan, G., Mitchell, J., Moyer, M., Pahel, G., Rocque, W., Overton, L. K., Schoenen, F., Seaton, T., Su, J. L., Warner, J., Willard, D., and Becherer, J. D. (1997). Cloning of a disintegrin metalloproteinase that processes precursor tumournecrosis factor-alpha. Nature 385, 733–736. Muller, A., Homey, B., Soto, H., Ge, N., Catron, D., Buchanan, M. E., McClanahan, T., Murphy, E., Yuan, W., Wagner, S. N., Barrera, J. L., Mohar, A., Verastegui, E., and Zlotnik,

182

Stephen C. Robinson and Lisa M. Coussens

A. (2001). Involvement of chemokine receptors in breast cancer metastasis. Nature 410, 50–56. Muller, W. A. (2001). Migration of leukocytes across endothelial junctions: Some concepts and controversies. Microcirculation 8, 181–193. Murch, S. H., MacDonald, T. T., Walker-Smith, J. A., Levin, M., Lionetti, P., and Klein, N. J. (1993). Disruption of sulphated glycosaminogly cans in intestinal inflammation. Lancet 341, 711–714. Murphy, P. M. (2001). Chemokines and the molecular basis of cancer metastasis. N. Engl. J. Med. 345, 833–835. Nawroth, P. P., Bank, I., Handley, D., Cassimeris, J., Chess, L., and Stern, D. (1986). Tumor necrosis factor/cachectin interacts with endothelial cell receptors to induce release of interleukin 1. J. Exp. Med. 163, 1363–1375. Negus, R. P., Stamp, G. W., Hadley, J., and Balkwill, F. R. (1997). Quantitative assessment of the leukocyte infiltrate in ovarian cancer and its relationship to the expression of C-C chemokines. Am. J. Pathol. 150, 1723–1734. Nelimarkka, L., Kainulainen, V., Schonherr, E., Moisander, S., Jortikka, M., Lammi, M., Elenius, K., Jalkanen, M., and Jarvelainen, H. (1997). Expression of small extracellular chondroitin/dermatan sulfate proteoglycans is differentially regulated in human endothelial cells. J. Biol. Chem. 272, 12730–12737. Ness, R. B., and Cauley, J. A. (2004). Antibiotics and breast cancer—what’s the meaning of this? JAMA 291, 880–881. Netelenbos, T., Drager, A. M., van het Hof, B., Kessler, F. L., Delouis, C., Huijgens, P. C., van den Born, J., and van Dijk, W. (2001). Differences in sulfation patterns of heparan sulfate derived from human bone marrow and umbilical vein endothelial cells. Exp. Hematol. 29, 884–893. Noguchi, T., Wada, S., Takeno, S., Moriyama, H., Kimura, Y., and Uchida, Y. (2003). Lymph node metastasis could be predicted by evaluation of macrophage infiltration and hsp70 expression in superficial carcinoma of the esophagus. Oncol. Rep. 10, 1161–1164. Norgauer, J., Metzner, B., and Schraufstatter, I. (1996). Expression and growth-promoting function of the IL-8 receptor beta in human melanoma cells. J. Immunol. 156, 1132–1137. Norrby, K. (2002). Mast cells and angiogenesis. APMIS 110, 355–371. Nowicki, A., Szenajch, J., Ostrowska, G., Wojtowicz, A., Wojtowicz, K., Kruszewski, A. A., Maruszynski, M., Aukerman, S. L., and Wiktor-Jedrzejczak, W. (1996). Impaired tumor growth in colony-stimulating factor 1 (CSF-1)-deficient, macrophage-deficient op/op mouse: Evidence for a role of CSF-1- dependent macrophages in formation of tumor stroma. Int. J. Cancer 65, 112–119. Nufer, O., Corbett, M., and Walz, A. (1999). Amino-terminal processing of chemokine ENA78 regulates biological activity. Biochemistry 38, 636–642. Ohno, S., Inagawa, H., Soma, G., and Nagasue, N. (2002). Role of tumor-associated macrophage in malignant tumors: Should the location of the infiltrated macrophages be taken into account during evaluation? Anticancer Res. 22, 4269–4275. Ono, M., Torisu, H., Fukushi, J., Nishie, A., and Kuwano, M. (1999). Biological implications of macrophage infiltration in human tumor angiogenesis. Cancer Chemother. Pharmacol. 43, S69–S71. Oohira, A., Wight, T. N., and Bornstein, P. (1983). Sulfated proteoglycans synthesized by vascular endothelial cells in culture. J. Biol. Chem. 258, 2014–2021. Ordemann, J., Jacobi, C. A., Braumann, C., Schwenk, W., Volk, H. D., and Muller, J. M. (2002). Immunomodulatory changes in patients with colorectal cancer. Int. J. Colorectal Dis. 17, 37–41. Overall, C. M., and Lopez-Otin, C. (2002). Strategies for MMP inhibition in cancer: Innovations for the post-trial era. Nat. Rev. Cancer 2, 657–672.

Soluble Mediators of Inflammation in Tumors

183

Overall, C. M., McQuibban, G. A., and Clark-Lewis, I. (2002). Discovery of chemokine substrates for matrix metalloproteinases by exosite scanning: A new tool for degradomics. Biol. Chem. 383, 1059–1066. Owen, J. D., Strieter, R., Burdick, M., Haghnegahdar, H., Nanney, L., Shattuck-Brandt, R., and Richmond, A. (1997). Enhanced tumor-forming capacity for immortalized melanocytes expressing melanoma growth stimulatory activity/growth-regulated cytokine beta and gamma proteins. Int. J. Cancer 73, 94–103. Padrines, M., Wolf, M., Walz, A., and Baggiolini, M. (1994). Interleukin-8 processing by neutrophil elastase, cathepsin G and proteinase-3. FEBS Lett. 352, 231–235. Petit, I., Szyper-Kravitz, M., Nagler, A., Lahav, M., Peled, A., Habler, L., Ponomaryov, T., Taichman, R. S., Arenzana-Seisdedos, F., Fujii, N., Sandbank, J., Zipori, D., and Lapidot, T. (2002). G-CSF induces stem cell mobilization by decreasing bone marrow SDF-1 and upregulating CXCR4. Nat. Immunol. 3, 687–694. Proost, P., De Meester, I., Schols, D., Struyf, S., Lambeir, A. M., Wuyts, A., Opdenakker, G., De Clercq, E., Scharpe, S., and Van Damme, J. (1998). Amino-terminal truncation of chemokines by CD26/dipeptidyl-peptidase IV. Conversion of RANTES into a potent inhibitor of monocyte chemotaxis and HIV-1-infection. J. Biol. Chem. 273, 7222–7227. Proost, P., Menten, P., Struyf, S., Schutyser, E., De Meester, I., and Van Damme, J. (2000). Cleavage by CD26/dipeptidyl peptidase IV converts the chemokine LD78beta into a most efficient monocyte attractant and CCR1 agonist. Blood 96, 1674–1680. Proost, P., Schutyser, E., Menten, P., Struyf, S., Wuyts, A., Opdenakker, G., Detheux, M., Parmentier, M., Durinx, C., Lambeir, A. M., Neyts, J., Liekens, S., Maudgal, P. C., Billiau, A., and Van Damme, J. (2001). Amino-terminal truncation of CXCR3 agonists impairs receptor signaling and lymphocyte chemotaxis, while preserving antiangiogenic properties. Blood 98, 3554–3561. Proudfoot, A. E. (2002). Chemokine receptors: Multifaceted therapeutic targets. Nat. Rev. Immunol. 2, 106–115. Proudfoot, A. E., Handel, T. M., Johnson, Z., Lau, E. K., LiWang, P., Clark-Lewis, I., Borlat, F., Wells, T. N., and Kosco-Vilbois, M. H. (2003). Glycosaminoglycan binding and oligomerization are essential for the in vivo activity of certain chemokines. Proc. Natl. Acad. Sci. USA 100, 1885–1890. Radisky, D., Hagios, C., and Bissell, M. J. (2001). Tumors are unique organs defined by abnormal signaling and context. Semin. Cancer Biol. 11, 87–95. Rhee, J. S., and Coussens, L. M. (2002). RECKing MMP function: Implications for cancer development. Trends Cell Biol. 12, 209–211. Ribatti, D., Vacca, A., Nico, B., Crivellato, E., Roncali, L., and Dammacco, F. (2001). The role of mast cells in tumour angiogenesis. Br. J. Haematol. 115, 514–521. Richmond, A., Fan, G. H., Dhawan, P., and Yang, J. (2004). How do chemokine/chemokine receptor activations affect tumorigenesis? Novartis Found. Symp. 256, 74–89. Richmond, A., and Thomas, H. (1986). Purification of melanoma growth stimulatory activity. J. Cell. Physiol. 129, 375–384. Robinson, S. C., Scott, K. A., and Balkwill, F. R. (2002). Chemokine stimulation of monocyte matrix metalloproteinase-9 requires endogenous TNF-alpha. Eur. J. Immunol. 32, 404–412. Robinson, S. C., Scott, K. A., Wilson, J. L., Thompson, R. G., Proudfoot, A. E., and Balkwill, F. R. (2003). A chemokine receptor antagonist inhibits experimental breast tumor growth. Cancer Res. 63, 8360–8365. Rosen, S. D. (2004). Ligands for L-selectin: Homing, inflammation, and beyond. Annu. Rev. Immunol. 22, 129–156. Rossi, D., and Zlotnik, A. (2000). The biology of chemokines and their receptors. Annu. Rev. Immunol. 18, 217–242.

184

Stephen C. Robinson and Lisa M. Coussens

Rubin, J. B., Kung, A. L., Klein, R. S., Chan, J. A., Sun, Y., Schmidt, K., Kieran, M. W., Luster, A. D., and Segal, R. A. (2003). A small-molecule antagonist of CXCR4 inhibits intracranial growth of primary brain tumors. Proc. Natl. Acad. Sci. USA 100, 13513–13518. Rutkowski, P., Kaminska, J., Kowalska, M., Ruka, W., and Steffen, J. (2003). Cytokine and cytokine receptor serum levels in adult bone sarcoma patients: Correlations with local tumor extent and prognosis. J. Surg. Oncol. 84, 151–159. Sallusto, F., and Lanzavecchia, A. (1994). Efficient presentation of soluble antigen by cultured human dendritic cells is maintained by granulocyte/macrophage colony-stimulating factor plus interleukin 4 and downregulated by tumor necrosis factor alpha. J. Exp. Med. 179, 1109–1118. Sasisekharan, R., Shriver, Z., Venkataraman, G., and Narayanasami, U. (2002). Roles of heparan-sulphate glycosaminoglycans in cancer. Nat. Rev. Cancer 2, 521–528. Satoh, T., Saika, T., Ebara, S., Kusaka, N., Timme, T. L., Yang, G., Wang, J., Mouraviev, V., Cao, G., Fattah el, M. A., and Thompson, T. C. (2003). Macrophages transduced with an adenoviral vector expressing interleukin 12 suppress tumor growth and metastasis in a preclinical metastastatic prostate cancer model. Cancer Res. 63, 7853–7860. Schioppa, T., Uranchimeg, B., Saccani, A., Biswas, S. K., Doni, A., Rapisarda, A., Bernasconi, S., Saccani, S., Nebuloni, M., Vago, L., Mantovani, A., Melillo, G., and Sica, A. (2003). Regulation of the chemokine receptor CXCR4 by hypoxia. J. Exp. Med. 198, 1391–1402. Schoppmann, S., Birner, P., Stockl, J., Kalt, R., Ullrich, R., Caucig, C., Kriehuber, E., Nagy, K., Alitalo, K., and Kerjaschki, D. (2002). Tumor-associated macrophages express lymphatic endothelial growth factors and are related to peritumoral lymphangiogenis. Am. J. Pathol. 161, 947–956. Scott, K. A., Moore, R. J., Arnott, C. H., East, N., Thompson, R. G., Scallon, B. J., Shealy, D. J., and Balkwill, F. R. (2003). An anti-tumor necrosis factor-alpha antibody inhibits the development of experimental skin tumors. Mol. Cancer Ther. 2, 445–451. Scotton, C. J., Wilson, J. L., Milliken, D., Stamp, G., and Balkwill, F. R. (2001). Epithelial cancer cell migration: A role for chemokine receptors? Cancer Res. 61, 4961–4965. Scotton, C. J., Wilson, J. L., Scott, K., Stamp, G., Wilbanks, G. D., Fricker, S., Bridger, G., and Balkwill, F. R. (2002). Multiple actions of the chemokine CXCL12 on epithelial tumor cells in human ovarian cancer. Cancer Res. 62, 5930–5938. Sgadari, C., Angiolillo, A. L., Cherney, B. W., Pike, S. E., Farber, J. M., Koniaris, L. G., Vanguri, P., Burd, P. R., Sheikh, N., Gupta, G., Teruya-Feldstein, J., and Tosato, G. (1996). Interferoninducible protein-10 identified as a mediator of tumor necrosis in vivo. Proc. Natl. Acad. Sci. USA 93, 13791–13796. Sgadari, C., Farber, J. M., Angiolillo, A. L., Liao, F., Teruya-Feldstein, J., Burd, P. R., Yao, L., Gupta, G., Kanegane, C., and Tosato, G. (1997). Mig, the monokine induced by interferongamma, promotes tumor necrosis in vivo. Blood 89, 2635–2643. Shacter, E., and Weitzman, S. A. (2002). Chronic inflammation and cancer. Oncology 16, 217–226. Shalinsky, D. R., Brekken, J., Zou, H., Kolis, S., Wood, A., Webber, S., and Appelt, K. (1998). Antitumor efficacy of AG3340 associated with maintenance of minimum effective plasma concentrations and not total daily dose, exposure or peak plasma concentrations. Invest. New Drugs 16, 303–313. Shriver, Z., Liu, D., and Sasisekharan, R. (2002). Emerging views of heparan sulfate glycosaminoglycan structure/activity relationships modulating dynamic biological functions. Trends Cardiovasc. Med. 12, 71–77. Sica, A., Saccani, A., Bottazzi, B., Polentarutti, N., Vecchi, A., van Damme, J., and Mantovani, A. (2000). Autocrine production of IL-10 mediates defective IL-12 production and NF-kappa B activation in tumor-associated macrophages. J. Immunol. 164, 762–767.

Soluble Mediators of Inflammation in Tumors

185

Smith, D. R., Polverini, P. J., Kunkel, S. L., Orringer, M. B., Whyte, R. I., Burdick, M. D., Wilke, C. A., and Strieter, R. M. (1994). Inhibition of interleukin 8 attenuates angiogenesis in bronchogenic carcinoma. J. Exp. Med. 179, 1409–1415. Sottile, J. (2004). Regulation of angiogenesis by extracellular matrix. Biochim. Biophys. Acta 1654, 13–22. Spadaro, M., and Forni, G. (2004). Proinflammatory cytokines, immune response and tumour progression. Novartis Found. Symp. 256, 92–99. Staller, P., Sulitkova, J., LIsztwan, J., Moch, H., Oakeley, E., and Krek, W. (2003). Chemokine receptor CXC4 downregulated by von-Hippel-Lindau tumour suppressor pVHL. Nature 425, 307–311. Sternlicht, M. D., and Bergers, G. (2000). Matrix metalloproteinases as emerging targets in anticancer therapy: Status and prospects. Emerg. Theurpeut. Targets 4, 609–633. Strieter, R. M., Belperio, J. A., Phillips, R. J., and Keane, M. P. (2004). Chemokines: Angiogenesis and metastases in lung cancer. Novartis Found. Symp. 256, 173–184. Strieter, R. M., Polverini, P. J., Kunkel, S. L., Arenberg, D. A., Burdick, M. D., Kasper, J., Dzuiba, J., Van Damme, J., Walz, A., Marriott, D., Chan, S. Y., Roczniak, S., and Shanafelt, A. B. (1995). The functional role of the ELR motif in CXC chemokine-mediated angiogenesis. J. Biol. Chem. 270, 27348–27357. Sunderkotter, C., Steinbrink, K., Goebeler, M., Bhardwaj, R., and Sorg, C. (1994). Macrophages and angiogenesis. J. Leukoc. Biol. 55, 410–422. Tachibana, K., Hirota, S., Iizasa, H., Yoshida, H., Kawabata, K., Kataoka, Y., Kitamura, Y., Matsushima, K., Yoshida, N., Nishikawa, S., Kishimoto, T., and Nagasawa, T. (1998). The chemokine receptor CXCR4 is essential for vascularization of the gastrointestinal tract. Nature 393, 591–594. Talmor, M., Mirza, A., Turley, S., Mellman, I., Hoffman, L. A., and Steinman, R. M. (1998). Generation of large numbers of immature and mature dendritic cells from rat bone marrow cultures. Eur. J. Immunol. 28, 811–817. Tazawa, H., Okada, F., Kobayashi, T., Tada, M., Mori, Y., Une, Y., Sendo, F., Kobayashi, M., and Hosokawa, M. (2003). Infiltration of neutrophils is required for acquisition of metastatic phenotype of benign murine fibrosarcoma cells: Implication of inflammation-associated carcinogenesis and tumor progression. Am. J. Pathol. 163, 2221–2232. Teruya-Feldstein, J., Jaffe, E. S., Burd, P. R., Kanegane, H., Kingma, D. W., Wilson, W. H., Longo, D. L., and Tosato, G. (1997). The role of Mig, the monokine induced by interferongamma, and IP-10, the interferon-gamma-inducible protein-10, in tissue necrosis and vascular damage associated with Epstein-Barr virus-positive lymphoproliferative disease. Blood 90, 4099–4105. Thelen, M. (2001). Dancing to the tune of chemokines. Nat. Immunol. 2, 129–134. Thun, M. J., Henley, S. J., and Gansler, T. (2004). Inflammation and cancer: An epidemiological perspective. Novartis Found. Symp. 256, 6–21. Torisu, H., Ono, M., Kiryu, H., Furue, M., Ohmoto, Y., Nakayama, J., Nishioka, Y., Sone, S., and Kuwano, M. (2000). Macrophage infiltration correlates with tumor stage and angiogenesis in human malignant melanoma: Possible involvement of TNFalpha and IL- 1alpha. Int. J. Cancer 85, 182–188. Trowbridge, J. M., and Gallo, R. L. (2002). Dermatan sulfate: New functions from an old glycosaminoglycan. Glycobiology 12, 117R–125R. Tsimberidou, A. M., Waddelow, T., Kantarjian, H. M., Albitar, M., and Giles, F. J. (2003). Pilot study of recombinant human soluble tumor necrosis factor (TNF) receptor (p75) fusion protein (TNFR:Fc; Enbrel) in patients with refractory multiple myeloma: increase in plasma TNF alpha levels during treatment. Leuk. Res. 27, 375–380. Tsung, K., Dolan, J. P., Tsung, Y. L., and Norton, J. A. (2002). Macrophages as effector cells in interleukin 12-induced T cell-dependent tumor rejection. Cancer Res. 62, 5069–5075.

186

Stephen C. Robinson and Lisa M. Coussens

Ulbrich, H., Eriksson, E. E., and Lindbom, L. (2003). Leukocyte and endothelial cell adhesion molecules as targets for therapeutic interventions in inflammatory disease. Trends Pharmacol. Sci. 24, 640–647. Van den Steen, P. E., Husson, S. J., Proost, P., Van Damme, J., and Opdenakker, G. (2003). Carboxyterminal cleavage of the chemokines MIG and IP-10 by gelatinase B and neutrophil collagenase. Biochem. Biophys. Res. Commun. 310, 889–896. Van den Steen, P. E., Proost, P., Wuyts, A., Van Damme, J., and Opdenakker, G. (2000). Neutrophil gelatinase B potentiates interleukin-8 tenfold by aminoterminal processing, whereas it degrades CTAP-III, PF-4, and GRO-alpha and leaves RANTES and MCP-2 intact. Blood 96, 2673–2681. van der Veen, A. H., de Wilt, J. H., Eggermont, A. M., van Tiel, S. T., Seynhaeve, A. L., and ten Hagen, T. L. (2000). TNF-alpha augments intratumoural concentrations of doxorubicin in TNF-alpha-based isolated limb perfusion in rat sarcoma models and enhances anti-tumour effects. Br. J. Cancer 82, 973–980. Velicer, C. M., Heckbert, S. R., Lampe, J. W., Potter, J. D., Robertson, C. A., and Taplin, S. H. (2004). Antibiotic use in relation to the risk of breast cancer. JAMA 291, 827–835. Venturi, G. M., Tu, L., Kadono, T., Khan, A. I., Fujimoto, Y., Oshel, P., Bock, C. B., Miller, A. S., Albrecht, R. M., Kubes, P., Steeber, D. A., and Tedder, T. F. (2003). Leukocyte migration is regulated by L-selectin endoproteolytic release. Immunity 19, 713–724. Vicari, A. P., and Caux, C. (2002). Chemokines in cancer. Cytokine Growth Factor Rev. 13, 143–154. Virchow, R. (1863). Aetologie der neoplastichen Geschwulste/Pathogenie der neoplastischen Geschwulste. Verlag von August Hirschwald, Berlin, Germany. Wahl, L. M., and Kleinman, H. K. (1998). Tumor-associated macrophages as targets for cancer therapy. J. Natl. Cancer. Inst. 90, 1583–1584. Walport, M. J. (2001). Complement. Second of two parts. N. Engl. J. Med. 344, 1140–1144. Walz, A., Peveri, P., Aschauer, H., and Baggiolini, M. (1987). Purification and amino acid sequencing of NAF, a novel neutrophil-activating factor produced by monocytes. Biochem. Biophys. Res. Commun. 149, 755–761. Warzocha, K., Salles, G., Bienvenu, J., Bastion, Y., Dumontet, C., Renard, N., NeidhardtBerard, E. M., and Coiffier, B. (1997). Tumor necrosis factor ligand-receptor system can predict treatment outcome in lymphoma patients. J. Clin. Oncol. 15, 499–508. Wedemeyer, J., Tsai, M., and Galli, S. J. (2000). Roles of mast cells and basophils in innate and acquired immunity. Curr. Opin. Immunol. 12, 624–631. Weller, A., Isenmann, S., and Vestweber, D. (1992). Cloning of the mouse endothelial selectins. Expression of both E- and P-selectin is inducible by tumor necrosis factor alpha. J. Biol. Chem. 267, 15176–15183. Wilson, J., and Balkwill, F. (2002). The role of cytokines in the epithelial cancer microenvironment. Semin. Cancer Biol. 12, 113–120. Witt, D. P., and Lander, A. D. (1994). Differential binding of chemokines to glycosaminoglycan subpopulations. Curr. Biol. 4, 394–400. Wolf, M., Clark-Lewis, I., Buri, C., Langen, H., Lis, M., and Mazzucchelli, L. (2003). Cathepsin D specifically cleaves the chemokines macrophage inflammatory protein-1 alpha, macrophage inflammatory protein-1 beta, and SLC that are expressed in human breast cancer. Am. J. Pathol. 162, 1183–1190. Wright, T. W., Pryhuber, G. S., Chess, P. R., Wang, Z., Notter, R. H., and Gigliotti, F. (2004). TNF receptor signaling contributes to chemokine secretion, inflammation, and respiratory deficits during Pneumocystis pneumonia. J. Immunol. 172, 2511–2521. Wuyts, A., D’Haese, A., Cremers, V., Menten, P., Lenaerts, J. P., De Loof, A., Heremans, H., Proost, P., and Van Damme, J. (1999). NH2- and COOH-terminal truncations of murine

Soluble Mediators of Inflammation in Tumors

187

granulocyte chemotactic protein-2 augment the in vitro and in vivo neutrophil chemotactic potency. J. Immunol. 163, 6155–6163. Yang, T. Y., Chen, S. C., Leach, M. W., Manfra, D., Homey, B., Wiekowski, M., Sullivan, L., Jenh, C. H., Narula, S. K., Chensue, S. W., and Lira, S. A. (2000). Transgenic expression of the chemokine receptor encoded by human herpesvirus 8 induces an angioproliferative disease resembling Kaposi’s sarcoma. J. Exp. Med. 191, 445–454. Yoshimura, T., Matsushima, K., Tanaka, S., Robinson, E. A., Appella, E., Oppenheim, J. J., and Leonard, E. J. (1987). Purification of a human monocyte-derived neutrophil chemotactic factor that has peptide sequence similarity to other host defense cytokines. Proc. Natl. Acad. Sci. USA 84, 9233–9237. Zacharski, L. R. (2003). Heparin treatment of malignancy: The case for clinical trials in colon cancer. Thromb. Res. 110, 213–214. Zhou, M., Zhang, Y., Ardans, J. A., and Wahl, L. M. (2003). Interferon-gamma differentially regulates monocyte matrix metalloproteinase-1 and -9 through tumor necrosis factor-alpha and caspase 8. J. Biol. Chem. 278, 45406–45413. Zlotnik, A., and Yoshie, O. (2000). Chemokines: A new classification system and their role in immunity. Immunity 12, 121–127.

Classical and Nonclassical HLA Class I Antigen and NK Cell–Activating Ligand Changes in Malignant Cells: Current Challenges and Future Directions Chien-Chung Chang,* Michael Campoli,* and Soldano Ferrone Department of Immunology, Roswell Park Cancer Institute, Buffalo, New York 14263

I. Introduction II. Detection of HLA Antigen and NK Cell–Activating Ligand Expression in Malignant Lesions III. Frequency and Molecular Mechanisms Underlying Abnormal HLA Class I Antigen Phenotypes in Malignant Lesions A. Total HLA Class I Antigen Loss and Marked Downregulation B. Selective HLA Class I Antigen Heavy Chain Loss and/or Downregulation IV. Nonclassical HLA Class I Antigen, MICA/B, and ULBP Expression by Malignant Cells A. HLA-E, -F, and -G Antigen Expression B. MICA/B and ULBP Ligand Expression V. Role of Immune Selective Pressure in the Generation of Lesions with Defects in Classical and Nonclassical HLA Class I Antigens and in NK Cell–Activating Ligand Expression VI. HLA Class I Antigen and HLA Class I-TA–Derived Peptide Complex Expression by Malignant Cells VII. Conclusions References

Changes in classical and nonclassical HLA class I antigen and NK cell–activating ligand expression have been identified in malignant lesions. These changes, which are described in this chapter, are believed to play a major role in the clinical course of the disease since both HLA class I antigens and NK cell–activating ligands are critical to the interaction between tumor cells and components of both innate and adaptive immune systems. Nevertheless, there is still debate in the literature about the biologic and functional significance of HLA class I antigen and NK cell–activating ligand abnormalities in malignant lesions. The reasons for this debate are reviewed. They include (i) the incomplete association between classical HLA class I antigen changes and the clinical course of the disease; (ii) the relatively limited number of malignant lesions that have been analyzed for nonclassical HLA class I antigen and NK cell–activating ligand expression; and (iii) the conflicting data regarding the role of immunoselection in the generation of malignant cells with HLA antigen and NK cell–activating ligand *Equal contributions to this work Advances in CANCER RESEARCH 0065-230X/05 $35.00

189

Copyright 2005, Elsevier Inc. All rights reserved

190

Chien-Chung Chang et al.

abnormalities. The technical limitations associated with the assessment of HLA antigen and NK cell–activating ligand expression in malignant lesions as well as the immunological and nonimmunological variables that may confound the impact of HLA antigen and NK cell–activating ligand changes on the clinical course of the disease are also discussed. Future studies aimed at overcoming these limitations and characterizing these variables are expected to provide a solution to the current debate regarding the significance of HLA class I antigen and NK cell–activating ligand abnormalities in malignant lesions. ß 2005 Elsevier Inc.

I. INTRODUCTION In humans, as in other animal species, malignant transformation of cells is often associated with changes in gene expression and in their antigenic profile. The latter include changes in human leukocyte antigen (HLA) expression, which have been convincingly documented in a number of malignant tumors (Marincola et al., 2000). For some time, characterization of these changes has not been a major topic of research in tumor immunology in spite of the identification in early studies of associations between changes in HLA antigen expression and clinical course of the disease in patients with malignancies (Marincola et al., 2000). The lack of interest in the analysis of HLA antigen changes in malignant lesions was largely due to skepticism regarding the cancer immune surveillance theory (Stutman, 1974, 1979), the artifactual nature of some of the early published data (Bortin and Truitt, 1980, 1981), difficulties in analyzing HLA antigen expression in tissues because of limitations of the available methodology as well as limited specificity of the available antisera. However, during the past three decades, the enthusiastic interest in applying active specific T cell–based immunotherapy for the treatment of malignant diseases (Parmiani et al., 2002), the realization of the crucial role of HLA antigens in the interactions of tumor cells with the host immune system (Crowley et al., 1991; Rosenberg et al., 1988; Wolfel et al., 1989), the revival of the cancer immune surveillance theory (Dunn et al., 2002), the availability of HLA antigen–specific monoclonal antibodies (mAb), as well as improvements in immunohistochemical (IHC) techniques (Marincola et al., 2000) have led to renewed interest in the molecular characterization, functional significance, and clinical relevance of HLA antigen abnormalities in malignant cells. Moreover, the appreciation of the role of immune escape of tumor cells from immune surveillance in the poor clinical response rates observed in the majority of the T cell–based immunotherapy clinical trials conducted to date (Ferrone, 2002; Khong and Restifo, 2002; Parmiani et al., 2002) has highlighted the need to characterize the molecular mechanisms by which tumor cells evade immune recognition and destruction. For many years, investigators have focused their studies on the characterization of the expression and functional properties of classical HLA class I

HLA and NKG2D Ligand Changes in Malignant Cells

191

antigens in malignant lesions since these antigens play a critical role in the presentation of tumor antigen (TA)-derived peptides to TA-specific cytotoxic T lymphocytes (CTL) (Crowley et al., 1991; Rosenberg et al., 1988; Wolfel et al., 1989), and modulate the interactions of natural killer (NK) cells and T cell subpopulations with target cells (Ferrone, 2002; Ikeda et al., 1997; Malmberg et al., 2002). Since 2000, these studies have been expanded to include the analysis of the expression and functional characteristics of nonclassical HLA class I antigens such as HLA-E, -F, and -G, the phylogenetically distant MHC class I-related chain (MIC) as well as the UL16-binding proteins (ULBP) in malignant lesions, given their ability to act as immunosuppressive molecules and NK cell–activating ligands, respectively (Bahram, 2000; Carosella et al., 2003; O’Callaghan and Bell, 1998; Seliger et al., 2003a; Vivier et al., 2002). IHC staining with mAb has demonstrated that changes in both classical (Marincola et al., 2000) and nonclassical (Carosella et al., 2003) HLA class I antigen expression occur in malignant lesions and may be associated with histopathological markers of poor prognosis of the disease and with reduced disease free interval and survival (Marincola et al., 2000). Nonetheless, skepticism remains in the field of tumor immunology concerning the true biological and functional significance of HLA antigen changes in malignant lesions (Pardoll, 2003). This uncertainty is due in part to (i) the lack of double blind studies providing conclusive evidence about the clinical significance of HLA class I antigen defects in malignant lesions; (ii) the conflicting information available regarding the association between changes in HLA class I antigen expression and clinical course of the disease in patients with malignant disease (Marincola et al., 2000); and (iii) the inconsistent results obtained by a number of investigators regarding the frequency of nonclassical HLA class I antigen expression, such as HLA-G, in several types of malignant tumors (Chang and Ferrone, 2003). In this chapter, we first review the available information about the frequency and molecular mechanisms underlying abnormalities in classical and nonclassical HLA class I antigen as well as MIC and ULBP antigen expression in malignant lesions. Second, we address the role of immune selective pressure, imposed by the host immune system, in the generation of malignant lesions with HLA class I antigen defects. Third, we discuss the reasons underlying the lack of complete association between defects in HLA class I antigen expression in malignant lesions and clinical course of the disease. Lastly, we discuss the lines of research which are likely to improve our understanding of the role of HLA class I antigen and NK cell–activating ligand changes in malignant lesions. In organizing this chapter, we have not attempted to review each subject in its entirety, but rather have focused on what we believe are the critical issues in this field. We refer the interested reader to published reviews on this topic (Carosella et al., 2003; Chang and

192

Chien-Chung Chang et al.

Ferrone, 2003; Garrido and Algarra, 2001; Marincola et al., 2000; Seliger et al., 2003a) for additional information.

II. DETECTION OF HLA ANTIGEN AND NK CELL–ACTIVATING LIGAND EXPRESSION IN MALIGNANT LESIONS Because of the need to assess HLA antigen expression at the protein level and to define its cellular distribution, analysis of classical HLA class I antigen expression in malignant lesions has primarily been and is being performed through IHC staining of surgically removed tumor sections with mAb which, in most cases, recognize determinants shared by classical and nonclassical HLA class I antigens (Carosella et al., 2003; Marincola et al., 2000). mAb have been utilized because the limited specificity and high background staining obtained with conventional allo- and xeno-antisera have hindered their application in IHC techniques. Only in recent years have mAb recognizing a large number of HLA class I allospecificities become available due to the major effort to develop mAb for HLA typing. Unfortunately, not many of the available HLA allospecificity–specific mAb work in IHC reactions. As a result, HLA class I allospecificity expression in malignant lesions has only been analyzed to a limited extent. This limitation is not likely to be overcome in the near future, since the declining interest in the use of antibody-based methods for HLA typing has had a negative impact on the development of mAb recognizing classical HLA class I allospecificities. Furthermore, the methodology to detect HLA class I allospecificity expression, which does not require HLA class I allospecificity–specific antibodies (Anastassiou et al., 2003), is not suitable to test large numbers of tissue samples. It is noteworthy that analysis of nonclassical HLA class I antigen and NK cell–activating ligand expression in malignant lesions is a relatively new field of research and progress in this area is hindered by the lack and/or limited availability of reagents suitable for staining normal tissues and malignant lesions. As a result, the majority of the studies published have utilized reverse transcription-polymerase chain reaction (RT-PCR) analysis of surgically removed tumor sections. The results of these studies are conclusive when mRNA is not detected. However, they must be interpreted with caution when mRNA is expressed, because of the frequent discordance between transcription and translation of a gene in malignant cells, as demonstrated for the nonclassical HLA class I antigen HLA-G (Chang and Ferrone, 2003). Frozen tissue sections have been for many years the only substrate that could be used in IHC reactions since the determinants recognized by the available anti-HLA antigen and, more recently, anti-NK cell–activating

HLA and NKG2D Ligand Changes in Malignant Cells

193

ligand mAb are not expressed in formalin-fixed, paraffin-embedded tissue sections. Although analysis of frozen tissue sections has provided useful information regarding the tissue distribution of both HLA antigens and NK cell ligands, they do not represent optimal substrates in IHC staining for several reasons. First, the use of frozen tissue sections as substrates in IHC staining reactions may result in loss of structural and cellular detail, making morphologic evaluation and IHC analysis difficult (Chan, 2000; Dabbs, 2002; Taylor, 1986). Second, the use of frozen tissue hampers retrospective studies, since most archived clinical samples are fixed with formalin and embedded with paraffin (Chan, 2000; Dabbs, 2002; Taylor, 1986). Lastly, preparation of frozen tissue sections requires significantly more time than formalin fixation of tissues and is not a routine procedure. These limitations have provided a stimulus to look for HLA antigen– and NK cell–activating ligand–specific mAb which stain formalin-fixed, paraffin-embedded tissue sections. In recent years, mAb that detect monomorphic (Perosa et al., 2003; Sernee et al., 1998; Stam et al., 1986) and locus-specific determinants (Cho et al., manuscript in preparation) of classical HLA class I antigens, beta-2 microglobulin ( 2m) (Lampson et al., 1983), monomorphic determinants of the nonclassical HLA class I antigen HLA-G (Paul et al., 2000), and monomorphic determinants of HLA class II antigens (Temponi et al., 1993) in formalin-fixed, paraffin-embedded tissue sections have been developed. These reagents, along with improvements in methods for antigen retrieval (Chan, 2000; Dabbs, 2002; Taylor, 1986), have allowed the use of formalin-fixed, paraffin-embedded tissues to analyze HLA antigen expression in surgically removed tumors, thereby facilitating the use of archived clinical samples in retrospective studies. These methodological improvements are likely to facilitate the analysis of malignant lesions for HLA antigen expression in departments of pathology. Until recently, analysis of HLA antigen and NK cell–activating ligand expression in tissues has been restricted to HLA laboratories because of pathologists’ reluctance to use frozen tissue sections in IHC reactions and skepticism about the clinical significance of data about HLA class I antigen expression in malignant lesions. However, it should be stressed that frozen tissue sections must still be used as substrates in IHC staining in order to characterize HLA allospecificity expression, since the corresponding polymorphic determinants are lost during the fixation of tissues with formalin and their embedding with paraffin. Moreover, frozen tissue sections must also be used as substrates in IHC staining to characterize nonclassical HLA class I antigen, HLA-E and HLA-F, as well as NK cell ligand expression, since none of the currently available HLA-E, HLA-F, and NK cell–activating ligand-specific mAb stains formalin-fixed, paraffin-embedded tissue sections in IHC reactions. HLA antigen expression is assessed by microscopic reading of stained tissue sections. For many years, there have been marked differences in the

194

Chien-Chung Chang et al.

methodology utilized by investigators to score HLA antigen expression. These differences—the lack of a standardized IHC staining technique and the use of mAb with different characteristics in terms of fine specificity and association constants—have hindered the comparison of the results published in the literature about HLA antigen expression in malignant lesions. To overcome these limitations, at least in part, the HLA and Cancer component of the 12th International Histocompatibility Workshop established a classification system for the assessment of HLA antigen expression in malignant lesions (Garrido et al., 1997). According to this classification system, lesions are scored as positive, heterogeneous, and negative, when the percentage of stained tumor cells in the entire lesion is more than 75%, between 75 and 25% inclusive, and less than 25%, respectively. Furthermore, staining intensity is scored as
(absent),  (dull), and þ (bright) and staining intensity of adjacent normal structures (i.e., lymphoid and endothelial cells) are used as an internal control to evaluate staining intensity of malignant cells. Despite this attempt at standardization, it should be stressed that analysis of IHC staining is largely subjective, since it relies on the assessment of tissue staining as estimated by independent observers. Therefore, caution should be exercised in comparing results obtained in different laboratories, although variations in the percentage of stained cells enumerated by two experienced investigators within the same laboratory is less than 10% (Kageshita et al., 1999) and a good concordance has been reported in the results obtained by four laboratories in a workshop organized by the HLA and Cancer component of the 13th International Histocompatibility Workshop (Campoli et al., 2004). It is hoped that the subjective nature of IHC techniques will be overcome in the near future by the development of equipment for computer-based IHC staining and evaluation of IHC staining results. These improvements should facilitate standardization of the analysis of HLA antigen and NK cell ligand expression in malignant lesions and should provide more quantitative data about HLA antigen and NK cell ligand expression in both normal and pathological tissues. However, even when these technical difficulties are solved, the following additional limitations must still be dealt with. First, IHC staining assesses the presence or absence of a moiety, but provides no information about its functional properties. For example, detection of the antigen processing machinery (APM) components in a malignant lesion does not prove that peptides are generated and loaded on 2m-HLA class I heavy chain complexes. Second, the correlation between results of IHC staining of cells by anti-HLA antigen or anti-NK cell–activating ligand mAb with target cell susceptibility to TA-specific CTL or NK cell–mediated lysis is not known, since no study has compared the sensitivity of antibody-based assays with that of CTL-based and NK cell–based assays to detect HLA antigens and NK cell–activating ligands on the cell membrane.

HLA and NKG2D Ligand Changes in Malignant Cells

195

III. FREQUENCY AND MOLECULAR MECHANISMS UNDERLYING ABNORMAL HLA CLASS I ANTIGEN PHENOTYPES IN MALIGNANT LESIONS The development of HLA antigen–specific mAb has had a major impact on the analysis of the expression of these molecules in tissues, since these reagents have overcome the restrictions imposed by conventional antisera because of their low titer, limited specificity, and interference of contaminating antibodies. As a result, beginning in the 1980s and continuing today, a large number of frozen and formalin-fixed malignant lesions have been tested with HLA class I antigen–specific mAb in IHC reactions (Cabrera et al., 2003a; Garrido and Algarra, 2001; Hicklin et al., 1999; Marincola et al., 2000). The results of these studies have convincingly shown that malignant transformation of cells is associated with classical HLA class I antigen loss or downregulation in most malignancies and with their upregulation in a few malignancies. Abnormalities in classical HLA class I antigen expression in malignant cells range from total HLA class I antigen loss or downregulation to selective loss of one HLA class I allospecificity, as schematically shown in Fig. 1. In addition, a combination of these phenotypes, sometimes in association with nonclassical HLA class I antigen expression and/or with abnormalities in NK cell–activating ligand expression, has also been frequently found in malignant cells. The significance of the latter phenotypes will be discussed later. Since 2000, a significant amount of information has been accumulated regarding the multiple molecular mechanisms underlying abnormal classical HLA class I antigen phenotypes. This information has mostly been derived from the analysis of cell lines with defective classical HLA class I antigen expression utilizing immunochemical assays and recombinant DNA technology (Wang et al., 1993). These mechanisms, which will be reviewed in the next few sections, have been documented, in most cases, also in the malignant lesions from which the cell lines had been derived (Hicklin et al., 1998; Wang et al., 1993). Therefore, the abnormalities identified in cell lines do not represent artifacts of in vitro cell culture, but reflect in vivo defects.

A. Total HLA Class I Antigen Loss and Marked Downregulation The subjective nature of the evaluation of IHC staining of tissue sections has led to a marked variation in the reported frequency of HLA antigen expression in malignant lesions and has made it difficult to assess the differentiation between loss and marked downregulation of the molecule

196

Chien-Chung Chang et al.

Fig. 1 Molecular mechanisms underlying abnormal HLA class I antigen phenotypes identified in malignant cells. (A) Total HLA class I antigen (HLA-A, -B, and -C) loss is caused by loss of 2m expression and/or function. An example is represented by the large deletion of the 2m gene detected by PCR in the lung adenocarcinoma cell line RPCI 9530 as compared to the HLA class I antigen–positive peripheral blood lymphocytes (i) (Chang et al., unpublished observations). Selective HLA class I allospecificity loss is caused by loss of the gene(s) which encode(s) the lost HLA class I allele(s) or by mutations which inhibit their transcription or translation. An example of selective HLA-A3 allospecificity loss is presented (ii). Total loss of one HLA class I haplotype is caused by total or partial loss of one copy of chromosome 6, which encodes the genes for HLA class I heavy chains. An example of HLA-A2, -B7, -Cw1 haplotype loss is presented (iii). (B) Total HLA class I antigen downregulation is caused by loss or downregulation of antigen processing machinery components. An example of TAP1 loss is presented (iv). Selective HLA class I locus downregulation may be caused by locus-specific defects in HLA class I gene transcription. An example of HLA-B, -C antigen downregulation is presented (v).

HLA and NKG2D Ligand Changes in Malignant Cells

197

Fig. 2 Frequency of HLA class I antigen, TAP1, and tapasin downregulation in malignant lesions of different embryological origin. The most common types of solid tumors for which surgically removed lesions have been analyzed for HLA class I antigen ( ), HLA class I allospecificity ( ), TAP1 ( ), and tapasin ( ) expression are shown. Figures indicate the total number of lesions analyzed. Total and selective HLA class I antigen downregulation are combined. ND: not determined. (Data have been adapted from Amiot et al., 1998; Atkins et al., 2004a,b; Dissemond et al., 2003b,c; Drenou et al., 2002; Marincola et al., 2000; Ogino et al., 2003a; Ritz et al., 2001; Romero et al., 2005; Seliger et al., 2000, 2003b; Vitale et al., 1998, 2005; Weinman et al., 2003; Wetzler et al., 2001).

being investigated. Although we realize that the functional implications and the molecular mechanisms underlying total HLA class I antigen loss and pronounced downregulation are markedly different, for the limitations of the evaluation of the IHC staining methodology we have discussed before, we have grouped these two types of defects. The most common types of solid tumors, for which more than 100 surgically removed primary lesions have been analyzed, include head and neck squamous cell carcinoma (HNSCC), breast carcinoma, lung carcinoma, hepatoma, colon carcinoma, renal cell carcinoma (RCC), cervical carcinoma, prostate carcinoma, melanoma, leukemia (Wetzler et al., 2001), and lymphoma (Amiot et al., 1998; Drenou et al., 2002; Marincola et al., 2000) (Fig. 2). For other tumors, including stomach (Lopez-Nevot et al., 1989), pancreas (Scupoli et al., 1996; Torres et al., 1996), bladder (Cabrera et al., 2003b; CordonCardo et al., 1991), ovarian (Le et al., 2002; Vitale et al., 2004), germ cell (Klein et al., 1990), and basal cell (Kageshita et al., 1992; Natali et al., 1983; Ruiz-Cabello et al., 1989) carcinomas, the number of lesions analyzed is fewer than 100; therefore, the information available regarding HLA class I antigen expression is not sufficient to draw definitive conclusions. With the exception of liver carcinoma (Bi et al., 2002; Fukusato et al., 1986; Kurokohchi et al., 1996), leukemia (Wetzler et al., 2001), and lymphoma (Amiot et al., 1998; Drenou et al., 2002), the frequency of HLA class I antigen loss and/or downregulation has been found to

198

Chien-Chung Chang et al.

range from 16 to 80% of the various types of tumors stained with mAb recognizing monomorphic determinants. Breast carcinoma and prostate carcinoma demonstrate the highest frequency and RCC and melanoma demonstrate the lowest frequency (Marincola et al., 2000). As will be discussed, these findings are noteworthy in view of the suggested role of immune selective pressure in the generation of malignant lesions with HLA class I antigen loss or downregulation and of the potential role of immunological events in the pathogenesis and clinical course of the disease in RCC (Huland and Heinzer, 2003) and melanoma (Guerry, 1998). It is likely that multiple reasons contribute to the differences in the frequency of classical HLA class I defects in various types of tumors. Some of them are technical in nature and, as discussed before, include the sensitivity of the IHC reaction used, the characteristics of the mAb used in the IHC reactions, and the subjective evaluation of IHC staining; all of these factors may affect the results that are obtained. Furthermore, HLA antigen and NK cell–activating ligand expression on endothelial and/or lymphoid cells present in tissue sections, which, in the majority of cases, is used as a standard for evaluating the degree of expression of these molecules by malignant cells, does not represent the most appropriate control when the tumor cells being evaluated are of a different lineage. Additional important variables which play a role in the different frequency of HLA class I antigen abnormalities observed in tumors of different histotype include the extent of immune selective pressure imposed on tumor cell populations, their genetic instability, the time length between onset of tumor and diagnosis, the characteristics of the patient population investigated, and the histologic classification of the type of tumor analyzed. In particular, the histologic type of a malignant lesion is likely to be an important source of variability, since differences have been described in the frequency of HLA class I antigen expression between serous and mucinous-adeno ovarian carcinomas (Le et al., 2002) as well as among squamous cell, small cell, adeno, and large cell lung carcinoma (Dammrich et al., 1990). These results emphasize the need to stratify tumors according to their histotype when analyzing HLA antigen expression. To facilitate this stratification, as well as that of patient populations, the HLA and Cancer component of the 13th International Histocompatibility Workshop has developed standardized clinical questionnaires tailored to particular types of malignancies (Campoli et al., 2004). It is hoped that these questionnaires will facilitate the identification of the variables which determine the differences in HLA class I antigen defects in various types of malignancies. As has been indicated, abnormalities in classical HLA class I antigen expression have been detected with low frequency in both leukemia (Demanet et al., 2004; Wetzler et al., 2001) and lymphoma (Amiot et al., 1998; Drenou et al., 2002) cells. This finding is not likely to reflect a lack of genetic instability in leukemic and lymphoma cells, since they, like solid

HLA and NKG2D Ligand Changes in Malignant Cells

199

tumor cells, harbor many genetic and/or epigenetic alterations in their DNA (Bagg, 2004; Gumy-Pause et al., 2004; Kelly and Gilliland, 2002; Smoller et al., 2003; Vega and Medeiros, 2003). Furthermore, in view of the role of immunoselection in the generation of malignant cell populations with classical HLA class I antigen defects (Lehmann et al., 1995; Restifo et al., 1996), lack of leukemic or lymphoma cells-specific, T cell–mediated immunity is unlikely to be a mechanism since (i) leukemic- (Andersen et al., 2001) and lymphoma- (Chaperot et al., 2002) specific T cells have been identified in patients and (ii) a higher frequency of classical HLA class I antigen abnormalities has been found in sporadic diffuse large-cell lymphoma than in immunodeficient and transplant-related lymphomas (List et al., 1993). We favor the possibility that the lack of defects in classical HLA class I antigen expression identified in leukemic and lymphoma cells reflects both the time interval between the onset of disease and its diagnosis as well as the type of immune selective pressure imposed on tumor cell populations. The time interval between the onset of leukemia or lymphoma and its diagnosis is likely to be shorter than that of most, if not all, solid tumors. A short time interval between the onset of tumor and diagnosis may not allow cells to acquire mutations in the gene(s) involved in HLA class I antigen expression and for T cell selective pressure to facilitate the expansion of malignant cells with HLA class I antigen abnormalities. In addition, because of their hematogeneous route of primary tumor and metastasis formation, leukemic and lymphoma cells are targeted by NK cells which are often distributed in the peripheral blood as well as in secondary and tertiary lymphoid sites (i.e., spleen, bone marrow, liver, lung, and lymph nodes) (Morris and Ley, 2004). The role of NK cells in controlling leukemic cells is supported by the association between a decrease in NK cell function and progression of chronic myelogenous leukemia (CML) from chronic phase to blast crisis (Chiorean et al., 2003; Pierson and Miller, 1996; Verfaillie et al., 1990). Therefore, the high HLA class I antigen expression found in leukemia and lymphoma cells may result from the outgrowth of tumor cells with high HLA class I antigen expression because of their reduced susceptibility to NK cell–mediated lysis. Our working hypothesis is supported by the selective HLA-A and HLA-Bw6 allospecificity downregulation which has been found in leukemic cells and which does not increase the susceptibility of target cells to NK cell–mediated lysis (Demanet et al., 2004). In the case of liver carcinoma, normal hepatocytes, which do not express or express very low levels of the transporter associated with antigen processing (TAP), a component of the HLA class I APM, and classical HLA class I antigens acquire the expression of these antigens during malignant transformation (Bi et al., 2002; Fukusato et al., 1986; Kageshita et al., 1992; Kurokohchi et al., 1996). The lack of defects in classical HLA class I antigen expression identified in liver carcinoma cells may reflect the type of immune

200

Chien-Chung Chang et al.

selective pressure imposed on tumor cell populations. In this regard, many NK cells reside in the liver (Morris and Ley, 2004) and the decrease in their activity in patients with chronic liver diseases as well as liver carcinoma speaks for a role of these cells in the control of malignant hepatocytes (Chuang et al., 1990; Ono et al., 1996). Furthermore, expression of the NK cell–activating ligands MICA/B by liver carcinoma cells has been suggested to play an important role in the susceptibility of liver carcinoma cells to NK cell–mediated cytolysis (Jinushi et al., 2003). Therefore, the acquisition of classical HLA class I antigen expression by liver cells during malignant transformation may reflect the selection of cells which are resistant to NK cell–mediated lysis. Studies performed mostly with cell lines in long-term culture have shown that distinct molecular mechanisms underlie total HLA class I antigen loss and/or downregulation in malignant cells and that they are differentially present in different types of tumors. These mechanisms include defects in 2m synthesis, epigenetic alterations involving the HLA class I heavy chain loci, defects in the regulatory mechanisms that control HLA class I antigen expression, and abnormalities in one or more of the APM components.

1. ABNORMALITIES IN 2m EXPRESSION Complete HLA class I antigen loss is generally caused by defects in 2m, which is required for the formation of the HLA class I heavy chain 2m-peptide complex and its transport to the cell membrane (Ferrone, 2002). Since two copies of the 2m gene are present in each cell and only one copy is sufficient for HLA class I antigen expression, complete HLA class I antigen loss is caused by the combination of two events. One involves loss of one copy of the 2m gene because of total or partial loss of chromosome #15, which carries the 2m gene in humans (Goodfellow et al., 1975). The other event involves mutations in the remaining copy of the 2m gene (Seliger et al., 2002). It is not known which of these two events occurs first. It is likely that this information will become available in the near future, since the analysis of tumor cells for loss of heterozygosity (LOH), which is currently widely applied to characterize malignant cells, will determine the frequency of loss of one copy of the 2m gene in malignant cells which express HLA class I antigens. The mutations identified thus far in 2m genes range from large to single nucleotide deletions, which, in most cases, inhibit the translation of mRNA, but do not affect the transcription of 2m gene (Seliger et al., 2002). Although the mutations are distributed randomly in 2m genes, a mutation hot spot has been suggested to be located in the CT repeat region in exon 1 of the 2m gene (Seliger et al., 2002). The available evidence suggests that

HLA and NKG2D Ligand Changes in Malignant Cells

201

the type of mutations in the 2m gene may be influenced by the extent of T cell–mediated selective pressure applied to tumor cell populations, with a high frequency in cell lines derived from patients receiving T cell–based immunotherapy. In this regard, the internal dinucleotide ‘‘CT’’ deletion has been found in three of five melanoma cell lines originating from patients receiving T cell–based immunotherapy (Chang et al., 2005) but only in one (Me1386) (Hicklin et al., 1998) out of five melanoma cell lines established from patients not treated with T cell–based immunotherapy (D’Urso et al., 1991; Wang et al., 1993). If confirmed in a large number of tumor cells of different histotype, this association may suggest a relationship between the outgrowth of tumor cells harboring certain types of genomic instability and the type/extent of immune selective pressure introduced by T cell–based immunotherapy. This possibility is supported by the different frequency of 2m gene mutations identified in different types of colon carcinomas (Bicknell et al., 1994; Gattoni-Celli et al., 1992; Hicklin et al., 1999). The 2m gene CT deletion has been found in microsatellite instability (MSI) (Lengauer et al., 1997) (þ) colon carcinomas but not in chromosomal instability (CIN) (Lengauer et al., 1997) (þ) colon carcinomas. MSI (þ) tumors have been found to be infiltrated by a large number of activated cytotoxic CD8þ T cells, and this infiltration is associated with improved survival in MSI (þ) patients (Di Giorgio et al., 1992; Dolcetti et al., 1999; Ripberger et al., 2003; Smyrk et al., 2001) (Fig. 3). In many other types of tumors, 2m gene mutations have not been or only occasionally been identified. They include HNSCC, breast carcinoma, RCC, and bladder carcinoma lesions with total HLA class I antigen loss and/or downregulation (Chen et al., 1996; Garrido and Algarra, 2001; Giorda et al., 2003; Jimenez et al., 2000). Alternative mechanisms suggested to underlie 2m loss in tumor cells include abnormal post-transcriptional regulation of 2m, which has been described in a drug-resistant breast carcinoma cell line (Ogretmen et al., 1998).

2. DEFECTS IN HLA CLASS I ANTIGEN GENE REGULATION AND/OR APM COMPONENT EXPRESSION At variance with total loss, HLA class I antigen downregulation in malignant cells is generally corrected by cytokines such as IFN-, IFN- , IL-1, and TNF-, since it is usually caused by altered regulation of nonmutated genes, such as those encoding APM components or DNA binding factors (Marincola et al., 2000; Sim and Hui, 1994). Furthermore, a nine base pair negative cis-regulatory element (NRE) has been identified in the 50 flanking region of HLA class I genes and has been shown to act as an inhibitor of the HLA enhancer element in breast carcinoma cell lines (Sim and Hui, 1994). However, the exact functional role of NRE in relationship to HLA class I

202

Chien-Chung Chang et al.

Fig. 3 Potential role of microsatellite instability (MSI) in generating a gene mutational hot spot preferentially selected by strong T cell immune selective pressure. Tumor cells with MSI (MSIþ) because of defective mismatch repair are more prone than those without MSI (MSI
) to small mutations in genes such as the 2m gene and the O-linked N-acetylglucosamine transferase (OGT ) gene, among others. Mutations in the OGT gene result in the generation of a novel HLA-A*0201-restricted CTL-epitope (Ripberger et al., 2003), which may recruit OGT mutant peptide-specific CTL to the tumor site. Under this T cell selective pressure, MSIþ tumor cells with 2m gene mutations can be out-selected over others without 2m gene mutations and become the major population at the tumor site.

antigen gene regulation in vivo remains to be determined. In addition, epigenetic mechanisms, such as hypermethylation of the HLA-A, B, and C gene promoter regions and/or altered chromatin structure of the HLA class I heavy chain gene promoters (Fonsatti et al., 2003; Nie et al., 2001; Serrano et al., 2001), have also been found to underlie total HLA class I antigen downregulation. This phenomenon has been implicated as a major mechanism for transcriptional inactivation of HLA class I antigen genes in esophageal squamous cell carcinoma (Nie et al., 2001) and is also responsible for total HLA class I antigen downregulation in cutaneous melanoma (Fonsatti et al., 2003). An additional frequent cause of HLA class I antigen downregulation is represented by defects in APM components (Seliger et al., 2000) which may affect the generation of peptides from antigens, translocation and loading of peptides onto 2m-HLA class I heavy chain dimers, and expression of stable HLA class I antigen-peptide complexes on the cell surface (Yewdell,

HLA and NKG2D Ligand Changes in Malignant Cells

203

2002). These unstable HLA class I antigen-peptide complexes and/or peptide free- 2m-HLA class I heavy chain dimers may be retained in the endoplasmic reticulum and degraded by proteasome at an increased rate, resulting in the downregulation of HLA class I antigen-peptide complexes on the cell membrane. In spite of the role of APM defects in HLA class I antigen downregulation, there is limited information about APM component expression in malignant lesions. Some components such as calnexin, calreticulin, and ERp57 have not been analyzed and others such as immunoproteasome subunits and TAP subunits have been tested only in a limited number of malignant lesions. In addition, several of the published studies have utilized RT-PCR analysis in order to assess APM component expression in malignant cell lines (Marincola et al., 2000; Seliger et al., 2000). Although these studies are conclusive when mRNA is not detected, they do not provide any information about the level and/or function of the proteins expressed when mRNA is expressed, given the lack of close correlation between transcription and translation of genes encoding APM components in malignant cells (S. Ferrone, unpublished observation). Lastly, the interpretation of the results related to the expression of APM components in malignant cells suffers from the scanty information about their expression in normal tissues. Consequently, it is difficult to determine what constitutes a normal or a pathological phenotype. The paucity of information about APM component expression in normal tissues and in malignant lesions reflects, at least in part, the limited availability of APM component-specific mAb with the appropriate characteristics. It is hoped that these limitations will be overcome by the panel of APM component-specific mAb which have been developed in recent years (Ogino et al., 2003b). To date, among the proteasome and immunoproteasome subunits, only the immunoproteasome subunits LMP2 and LMP7 have been tested for their expression in malignant lesions. It is noteworthy that it is a general assumption that immunoproteasome subunit expression is not constitutive within a cell, but is induced upon exposure to cytokines, i.e., IFN- , (Yewdell, 2002). However, there is limited experimental data to support this assumption and, as a matter of fact, there are several examples in the literature where these subunits have been observed in normal cells of different histology in the absences of apparent exposure to IFN- (Barton et al., 2002; Kuckelkorn et al., 2002; Yewdell, 2002). As a result, it is difficult to determine whether the expression of an immunoproteasome subunit in malignant cells is a normal phenotype and its lack of expression is a downregulation or loss, or whether the lack of expression of an immunoproteasome subunit is a normal phenotype and its expression reflects regulatory defects. Because of these limitations, we will present the data related to LMP expression in tumors as presence or absence without classifying them as normal or abnormal phenotypes. Hopefully, this classification

204

Chien-Chung Chang et al.

will be possible in the future when the distribution of immunoproteasome subunits in normal tissues is defined. LMP2 and LMP7 expression has primarily been analyzed in melanoma lesions where LMP2 and LMP7 have been found to be expressed in 63 and 45% of primary lesions and 81 and 47% of metastatic lesions, respectively (Dissemond et al., 2003a; Kageshita et al., 1999). Analysis of 22 surgically removed formalin-fixed, paraffinembedded breast carcinoma lesions found 100% of the lesions to express LMP2 and LMP7 with variable staining intensity and no association between LMP2 or LMP7 expression and tumor grading (Vitale et al., 1998). Heterogeneous LMP2 expression has also been found in 80, 59, and 19% of colorectal carcinoma (Atkins et al., 2004a), RCC (Atkins et al., 2004b), and cervical carcinoma (Ritz et al., 2001) lesions, respectively, while heterogeneous LMP7 expression has been found in 28 and 6% of RCC (Atkins et al., 2004b) and cervical carcinoma (Ritz et al., 2001) lesions. More recently, LMP2 and LMP7 expression has been noted to be ‘‘downregulated’’ in bladder carcinoma lesions (Romero et al., 2005); however, it is noteworthy that the latter study only utilized RT-PCR analysis to assess APM component expression. The APM component that has been most extensively investigated for its expression in malignant lesions is TAP1. TAP1 downregulation or loss has been found in HNSCC, breast carcinoma, lung carcinoma, colon carcinoma, RCC, bladder carcinoma, ovarian carcinoma, cervical carcinoma, prostate carcinoma, and cutaneous melanoma lesions with a frequency ranging from 10 to 84% (Atkins et al., 2004a,b; Dissemond et al., 2003b; Marincola et al., 2000; Ritz et al., 2001; Romero et al., 2005; Vitale et al., 1998, 2005) (Fig. 2). A few studies have investigated TAP2 expression in breast carcinoma, RCC, ovarian carcinoma, and melanoma lesions (Marincola et al., 2000; Seliger et al., 2003b; Vitale et al., 1998, 2005). The frequency of TAP2 downregulation tends to correlate with that of TAP1 (Seliger et al., 2000). Only during the last few years, taking advantage of newly developed mAb with the appropriate characteristics, has tapasin expression been analyzed in a few types of tumors. Tapasin has been found to be downregulated in HNSCC (Ogino et al., 2003a; Weinman et al., 2003), colon carcinoma (Atkins et al., 2004a), RCC (Seliger et al., 2003b), bladder carcinoma (Romero et al., 2005), and melanoma (Dissemond et al., 2003c) lesions with a frequency ranging between 10 and 65% (Fig. 2). Tapasin downregulation has been found to be correlated with survival in HNSCC (Ogino et al., 2003a). This correlation is likely to reflect the role of tapasin in the selection of peptides presented by HLA class I antigens to CTL (Yewdell, 2002) and its role in the regulation of HLA class I antigen expression (Yewdell, 2002). In the majority of cases, the level of APM components in cells with defective expression of these molecules can be enhanced by incubation with

HLA and NKG2D Ligand Changes in Malignant Cells

205

cytokines, e.g., IFN- or TNF- (Marincola et al., 2000; Seliger et al., 2000), indicating that these abnormalities are regulatory and not structural. Although the molecular mechanisms underlying these regulatory abnormalities have not been defined yet, the available evidence suggests that these defects are at the level of transcription. The rare description of mutations in the genes encoding APM components is likely to reflect not a low frequency of mutations, but technical difficulties to identify them with the currently used antibody-based techniques. In this regard, each APM component is encoded by codominant genes and structural defects in one gene do not appear to cause readily detectable abnormalities in the phenotype of cells and/or in the function of APM. This phenomenon is similar to the one we have described for 2m, which also requires defects in the two encoding genes present on the two chromosomes to generate a pathological phenotype. However, the identified structural abnormalities which result in loss of expression and/or function of APM components are different from those described in cell lines with 2m defects. In the latter, the two mutations causing 2m loss are cancer related and usually consist of the loss of one allele and mutation in the other one. In contrast, in the few cell lines where structural defects have been identified in APM components, cancer-related mutations in one allele have been found to be associated with an inherited or cancer-unrelated mutation in the other allele (Fig. 4). Thus, in one SCLC cell line with TAP1 loss, repression of the wild-type allele has revealed an inherited mutation at position 659 (R659Q) near the ATP-binding site of Tap1 (Chen et al., 1996). Furthermore, in a human melanoma cell line with tapasin loss, LOH of the wild-type allele has unmasked an inherited mutation at nucleotide position 684 present in the other allele (Chang et al., 2003). On the other hand, we do not know at present whether the TAP1 deficiency described in another human melanoma cell line reflects repression or loss of the wild-type allele (Seliger et al., 2001). It is noteworthy that a high frequency of structural TAP2 gene defects have been described in cervical carcinoma lesions (Fowler and Frazer, 2004). However, the molecular nature of these abnormalities has not been characterized yet.

B. Selective HLA Class I Antigen Heavy Chain Loss and/or Downregulation Different types of selective HLA class I antigen loss or downregulation have been identified in malignant cells. They include selective loss or downregulation of one HLA class I allospecificity, loss or downregulation of the gene products of one HLA class I locus, and loss of one HLA class I haplotype (Fig. 1). As observed for total HLA class I antigen defects, the frequency of selective HLA class I antigen abnormalities varies among

206

Chien-Chung Chang et al.

Fig. 4 Inherited mutations in one copy of the TAP1 or tapasin gene revealed by acquired cancer-related mutations in the other copy. At variance with 2m loss which results from two cancer-related mutational events, TAP1 and tapasin loss is caused by one inherited mutation and one acquired cancer-related mutation in malignant cells. The cancer-related mutations involve genetic or epigenetic repression of the wild-type TAP1 allele in the case of TAP1 loss, and LOH at the tapasin locus in the case of tapasin loss.

different types of tumors, being higher in primary cervical carcinoma, prostate carcinoma, and cutaneous melanoma lesions than in primary HNSCC, breast carcinoma, lung carcinoma, RCC, and colon carcinoma lesions (Marincola et al., 2000) (Fig. 2). The reason(s) for these differences is (are) not known. It is likely that the frequency of selective HLA class I antigen defects in malignant cells is higher than that described in the literature, since staining of malignant cells with mAb recognizing monomorphic determinants of HLA class I antigens does not detect selective HLA class I antigen abnormalities (Kageshita et al., 1993) and the expression of some HLA class I allospecificities in malignant lesions has not been assessed because of the lack of appropriate mAb. As discussed for total HLA class I antigen loss and downregulation, different mechanisms underlie selective HLA class I antigen loss and downregulation. While the latter is caused by regulatory abnormalities and is corrected by cytokines, the former is caused by structural defects and therefore can be corrected only by gene transfer. In most of the cases, one mutational event is sufficient to cause selective HLA class I antigen loss and

HLA and NKG2D Ligand Changes in Malignant Cells

207

this explains, at least in part, why the frequency of selective HLA class I antigen loss is higher than that of total HLA class I antigen loss for which two mutational events are required (Marincola et al., 2000). The mechanisms underlying selective HLA class I antigen loss include LOH of chromosome #6, which carries the HLA class I heavy chain genes in humans (Francke and Pellegrino, 1977), mutations which inhibit the transcription or translation of an HLA class I heavy chain gene (Seliger et al., 2002), hypermethylation of the HLA-A, -B, and -C gene promoter regions (Fonsatti et al., 2003; Nie et al., 2001; Serrano et al., 2001), or repression of a gene encoding one HLA class I allospecificity (Griffioen et al., 2000; Imreh et al., 1995; Marincola et al., 1994; Seliger et al., 2002). As in the case of the 2m gene, the mutations found in HLA class I heavy chains range from large deletions to single base deletions (Brady et al., 2000; Geertsen et al., 2002; Jimenez et al., 2001; Koopman et al., 2000; Wang et al., 1999). These mutations appear to occur randomly, although LOH in the short arm of chromosome #6 appears to represent the most frequent mechanism contributing to selective HLA haplotype loss in human tumors (Jimenez et al., 1999; Maleno et al., 2002; McEvoy et al., 2002). This finding may reflect the frequent genetic recombination events at the human MHC located at chromosome 6p21.3, which carries the highest density of genes among all gene loci in human chromosomes (Shiina et al., 1999). Additional mechanisms underlying selective HLA class I antigen loss include yet to be defined post-transcriptional mechanisms and/or genetic alterations at the chromosome 6p21.3 locus, which have been shown to result in partial HLA class I antigen loss in both cervical and prostate carcinoma cells (Chatterjee et al., 2001; Kasahara et al., 2002). Furthermore, a genomic region located on chromosome 1p35-36.1 has been implicated in HLA class I antigen expression and is known as the putative modifier of methylation for HLA class I genes (MEMO-1) (Cheng et al., 1996a,b). Selective HLA class I antigen downregulation may be caused by defective HLA class I locus- or allele-specific transcription. An additional mechanism is represented by instability of certain HLA class I allospecificities caused by defective loading with peptides because of low affinity for TAP-tapasin complexes.

IV. NONCLASSICAL HLA CLASS I ANTIGEN, MICA/B, AND ULBP EXPRESSION BY MALIGNANT CELLS Evidence accumulated during the last few years has convincingly shown that the nonclassical HLA class I antigens HLA-E, -F, and -G may serve as immunosuppressive molecules (O’Callaghan and Bell, 1998), while the

208

Chien-Chung Chang et al.

phylogenetically distant HLA class I MICA/B and ULBP1, ULBP2, and ULBP3 may act as NK cell–activating ligands (Bauer et al., 1999; Cosman et al., 2001). These findings have stimulated interest in the characterization of nonclassical HLA class I antigen and NK cell–activating ligand expression by tumor cells, since their interactions with the host immune system may be affected by these antigens. Here, we review the information about the frequency of nonclassical HLA class I antigen and MIC/ULBP expression in tumor cells and discuss the intriguing findings regarding the preferential in vivo HLA-G expression in malignant cells, as well as the ‘‘doubleedged sword’’ characteristics of MIC in tumors. It is noteworthy that the available information is still limited, since the field is in an early stage and progress in this area is hindered by the lack and/or limited availability of nonclassical HLA class I antigen- and MIC/ULBP molecule-specific mAb.

A. HLA-E, -F, and -G Antigen Expression To the best of our knowledge, HLA-E antigens have been investigated only in cell lines and only by Garrido and his collaborators (Marin et al., 2003) and by ourselves. Garrido and his associates (Marin et al., 2003) have investigated HLA-E antigen expression in a panel of 37 cell lines derived from tumors of different histotype including Burkitt’s lymphoma, histiocytic lymphoma, pre-B-cell leukemia, erythroleukemia, T-cell lymphoma, promyelocytic leukemia, HNSCC, carcinomas of the breast, lung, pancreas, stomach, colon, and cervix, and melanoma (Marin et al., 2003). All 37 cell lines have detectable HLA-E mRNA expression, while only 5 (13%) of them express HLA-E antigens on the cell surface. Three of the latter five cell lines had been derived from leukemia. HLA-F antigen expression has been investigated in over 30 cell lines derived from amnion, placenta, brain, lung, thymus, liver, pancreas, kidney, colon, bone marrow, and skin tumors (Lee and Geraghty, 2003). This study has demonstrated that HLA-F is localized primarily in the cytoplasm of cells. Thus far, only four EBVtransformed lymphoblastoid cell lines and three monocytic cell lines have been found to express HLA-F antigen on the cell surface (Lee and Geraghty, 2003). We have investigated HLA-E and HLA-F antigen expression on eight breast carcinoma cell lines. The preliminary results indicate that HLA-E and HLA-F antigens are expressed in 50 and 25%, respectively, of the eight breast carcinoma cell lines (Chang et al., unpublished). The two cell lines that express HLA-F antigen also express HLA-E antigen. Upon IFN- stimulation, both HLA-E and HLA-F antigen expression were induced or enhanced on seven of the eight cell lines analyzed. On the other hand, HLA-G antigen expression has been investigated extensively both in cell lines and in surgically removed tumors by a number

HLA and NKG2D Ligand Changes in Malignant Cells

209

of investigators as more than 100 cell lines and more than 150 surgically removed lesions have been analyzed for HLA-G antigen expression utilizing binding assays with HLA-G-specific antibodies, immunochemical assays, and RT-PCR (Rouas-Freiss et al., 2003). According to the pattern of HLA-G expression, the cell lines tested can be divided into three groups. The first one includes B chronic lymphocytic leukemia, non-Hodgkin B and T lymphoma, glioma, RCC, and ovarian carcinoma in which HLA-G antigen expression has been convincingly demonstrated. The second group includes carcinoma of breast, pancreas, colon, and prostate in which HLA-G antigen expression has not been detected. The third group includes melanoma in which only a minority of cell lines have been reported to express HLA-G antigen (Chang et al., 2003). Furthermore, conflicting results have been reported by a number of investigators (Chang and Ferrone, 2003). The frequency of HLA-G antigen expression has been reported to be higher in surgically removed lesions than in cell lines (Chang et al., 2003). This difference may be caused by both biological and technical reasons. The latter include mis-scoring of tumor infiltrating HLA-G-bearing macrophages as malignant cells and subjective evaluation of IHC results. The biological reasons for the discrepancy between in vivo and in vitro HLA-G antigen expression are likely to reflect differences in the characteristics of the lesions analyzed, heterogeneity of tumor cell populations, and lack of factors in tissue culture media that are able to induce HLA-G antigen expression in the tumor microenvironment. The latter possibility is supported by the gradual loss of HLA-G antigen expression in a melanoma cell line (N. Rouas-Freiss, personal communication) and in a panel of freshly isolated ovarian carcinoma cells (Malmberg et al., 2002) continuously cultured in vitro. If this finding also applies to other tumor types, one may argue that analysis of cultured cell lines for HLA-G antigen expression underestimates the actual frequency of HLA-G antigen expression in vivo. The mechanism underlying the gradual loss of HLA-G antigen expression by cell lines cultured in vitro is not known, but is thought to be, at least in part, represented by methylation of the HLA-G gene promoter that occurs in tissue culture conditions, as we have demonstrated in the uveal melanoma cell line OCM1-A (Chang et al., unpublished). Like classical HLA class I antigens (Puppo et al., 1997), HLA-G protein is also expressed in a soluble form in sera of healthy donors and of patients with malignant diseases (Rebmann et al., 2003). Elevated soluble HLA-G (sHLA-G) protein levels have been demonstrated in sera of patients with melanoma (Rebmann et al., 2003; Ugurel et al., 2001), glioma (Rebmann et al., 2003), breast carcinoma (Rebmann et al., 2003), ovarian carcinoma (Rebmann et al., 2003), and lymphoproliferative disorders (Amiot et al., 2003). The HLA-G serum level was inversely correlated with survival in glioblastoma multiformis (Rebmann et al., 2003) and was

210

Chien-Chung Chang et al.

directly correlated with advanced disease stage and with tumor load in melanoma (Ugurel et al., 2001). Furthermore, the HLA-G level has been found to be increased in ascites of patients with breast and ovarian carcinoma, suggesting that, in addition to cytology, it may be a useful marker to differentiate malignant from benign ascites (Singer et al., 2003).

B. MICA/B and ULBP Ligand Expression MICA and MICB expression have been examined in glioma (Friese et al., 2003), neuroblastoma (V. Pistoia, personal communication), leukemia (Salih et al., 2003), melanoma (Pende et al., 2002; Vetter et al., 2002), and carcinomas of breast, lung, colon, ovary, prostate, and kidney (Groh et al., 1999). MICA-specific mAb and mAb which recognize determinants shared by MICA and MICB have been utilized in most of the published studies. To the best of our knowledge, only glioma, neuroblastoma, and leukemic cells have been analyzed with MICB-specific mAb. Therefore, the data about frequency of MICB expression by malignant cells have to be interpreted with caution. In carcinomas, the frequency of MICA and MICB expression (or MIC expression) ranges from 20% (breast) to 100% (lung, renal cell, and prostate) of the surgically removed lesions analyzed either by flow cytometry or by IHC staining (Groh et al., 1999). In glioma, MICA and MICB were found to be expressed at a frequency of 100 and 75%, respectively, on 12 long-term cell lines and on 5 short-term cultured cell lines isolated from lesions (Friese et al., 2003). On the other hand, in leukemia, the frequency of MICA expression by long-term cell lines appears to be lower than that by leukemic cells freshly isolated from patients. MICA was found to be expressed by only one of the four leukemic long-term cell lines (Pende et al., 2002), while MICA and MICB were found to be expressed by 44 and 28%, respectively, of the leukemic preparations isolated from 25 patients (Salih et al., 2003). In melanoma, MICA is expressed at a frequency of 77% in cultured metastatic cell lines (Pende et al., 2002). Different frequencies of MICA expression have been found by Vetter (2002) and by ourselves in primary and metastatic surgically removed melanoma lesions. While Vetter and coworkers have reported a frequency of 77 and 60% in primary and metastatic lesions, respectively (Vetter et al., 2002), we have found MICA in 47 and 20%, respectively, of primary and metastatic lesions derived from an Asian population. Whether the difference between the results reported by Vetter (2002) and our own reflects differential sensitivity of the detection system used and/or different characteristics of the patient populations analyzed is not known. In addition, it is not known at present whether the lower MIC expression frequency in metastatic

HLA and NKG2D Ligand Changes in Malignant Cells

211

lesions, as opposed to primary lesions, observed in melanoma applies to other types of tumors. It is also worth noting that MICB is expressed on glioma and leukemic cells, but has not been detected on the surface of tumor cells of other histotypes (Chang et al., unpublished). In preliminary studies, we and others (V. Pistoia, personal communication) have found that MICB is preferentially localized in the cytoplasm of melanoma and neuroblastoma cells. This phenotype probably reflects an impaired intracellular trafficking of MICB protein in cells. Whether the latter phenomenon is governed by a mechanism similar to that of HLA-F is unknown. ULBP expression has been examined only in glioma (Friese et al., 2003), leukemia (Salih et al., 2003), and melanoma cells (Pende et al., 2002). In general, ULBP molecules are expressed less frequently than MIC molecules in malignant cells. All the 12 glioma cell lines tested express both ULBP2 and ULBP3, while only 41% of the cell lines express ULBP1. On the contrary, only ULBP2 was detected in glioma cells freshly isolated from 5 glioma lesions (Friese et al., 2003). Sixteen percent of the 25 leukemic cell preparations isolated from patients expressing ULBP1 and ULBP3 also express ULBP2 (Salih et al., 2003), while 20% of the leukemic cells express only ULBP2. In melanoma, ULBP expression has been analyzed only in cultured cell lines (Pende et al., 2002). In a panel of 18 cultured melanoma cell lines, ULBP2 and ULBP3 were detected in 27 and 50%, respectively, of them, while ULBP1 was detected at a frequency of 5% (Pende et al., 2002). ULBP antigens have not been tested in other types of malignant lesions thus far. It is noteworthy that MIC and ULBP do not appear to be expressed in a coordinated fashion in the tumor cells examined. In addition, the available information indicates that the frequency of MIC and ULBP expression appears to be independent of that of classical and nonclassical HLA class I antigen expression in malignant cells. At variance with the results we have discussed for HLA-G, the frequency of MIC and ULBP expression by cell lines is similar to that found in surgically removed lesions. It is likely that MIC and ULBP expression is closely associated with cell proliferation, since MIC expression is reduced on carcinoma cells (Chang et al., unpublished) and fibroblasts that have entered a resting stage (Zou et al., 2003). Functionally, MIC and ULBP have been convincingly shown to stimulate NK cell responses and co-stimulate CTL through the NKG2D receptor. MIC and ULBP are induced in transformed cells but not in their normal counterparts; therefore, they are considered to be a type of ‘‘induced-self’’ tumor antigen that serves as the target for the host immune surveillance. This possibility is supported by the crucial role of Rae-1, the mouse orthologs of MIC, in NKG2D-bearing  T cells’ control of skin cancer in mice (Girardi et al., 2001). However, several questions remain still to be addressed. First, the MICA- and MICB-null phenotype is present with a frequency of about

212

Chien-Chung Chang et al.

0.1% in Asians and the defective MICA*010 allele is expressed with a frequency of 7% in the general population (Bahram et al., 2000). The lack of a detectable increased frequency of malignancies in individuals with these phenotypes raises the question of whether MIC is dispensable or whether its function can be fully replaced by other NKG2D ligands such as ULBP1, ULBP2, and ULBP3. Second, soluble MIC (sMIC) have been detected in the spent medium of cultured tumor cell lines as well as in sera of patients with malignant diseases such as leukemia (Salih et al., 2003), colon carcinoma (Doubrovina et al., 2003), prostate carcinoma (Wu et al., 2004), and neuroblastoma (V. Pistoia, personal communication) and have been shown to downmodulate CTL and/or NK cell responses in vitro by downregulating the NKG2D receptor (Doubrovina et al., 2003; Groh et al., 2002; Wu et al., 2004) (Fig. 5). Nevertheless, the clinical significance of sMIC remains to be defined, since only in prostate cancer can sMIC serve as a disease progression marker (Wu et al., 2004). Whether this finding reflects a tumor type-specific phenomenon regarding the biological impact of sMIC or reflects the interference of other soluble immunomodulators such as sHLA class I molecules, sICAM-1, sHLA-G, etc., is not known. Clearly, more studies are needed to better define the double-edged sword nature of MIC in malignant diseases.

V. ROLE OF IMMUNE SELECTIVE PRESSURE IN THE GENERATION OF LESIONS WITH DEFECTS IN CLASSICAL AND NONCLASSICAL HLA CLASS I ANTIGENS AND IN NK CELL–ACTIVATING LIGAND EXPRESSION The role of HLA class I antigens in the recognition and destruction of tumor cells by HLA class I antigen restricted, TA-specific CTL implies that defects in HLA class I antigen expression in malignant lesions may have a negative impact on the clinical course of malignant diseases if the latter is influenced by a TA-specific CTL-mediated immune response. Furthermore, defects in HLA class I antigen expression in malignant lesions are expected to counteract the potential beneficial effects of T cell–based immunotherapy, since they may provide tumor cells with a mechanism to evade CTL recognition. These possibilities have stimulated interest in defining the mechanisms which lead to the generation of malignant lesions with HLA class I antigen defects with the expectation that this information may suggest strategies to prevent and/or inhibit their generation. Two events play a role in the generation of lesions with HLA class I antigen defects. One is

HLA and NKG2D Ligand Changes in Malignant Cells

213

Fig. 5 Tumor escape from NK cell control through downregulation or shedding of NKG2D ligands. Tumor cells with increased NK cell–activating ligand MIC and/or ULBP expression in conjunction with classical HLA class I antigen downregulation are sensitive to killing by NK cells. On the other hand, tumor cells with MIC/ULBP downregulation or shedding in spite of increased expression on the cell surface are resistant to lysis by NK cells. KIRs, killer inhibitory receptors.

represented by a mutation which causes a defect in the expression and/or function of HLA class I antigens in a tumor cell. The other is represented by events which favor the outgrowth of malignant cells with HLA class I antigen defects so that they become the predominant population in a malignant lesion. The potential involvement of immune selective pressure in these events is suggested by several lines of clinical and experimental evidence, all of which have in common the possibility that tumor cells with genetic and/or epigenetic defects in HLA class I antigens may be selected for by T cell–mediated selective pressure present in the tumor microenvironment. First, the higher frequency of HLA class I antigen defects in metastases than in primary lesions (Marincola et al., 2000) in various types of malignancies has been suggested to reflect the selection by the host immune system of tumor cells which acquire the ability to escape T cell recognition and therefore can migrate and form metastases in distant organs. Furthermore, the application of T cell–based immunotherapy to an increasing number of patients with malignant disease has provided many examples

214

Chien-Chung Chang et al.

of HLA class I antigen loss in metastases which had recurred in patients following an initial clinical response to therapy (Jager et al., 1997; Lehmann et al., 1995; Restifo et al., 1996; Riker et al., 1999). In these studies, exposure of tumor cells to a strong T cell selective pressure and outgrowth of tumor cells with HLA class I antigen loss are linked by an association and not by a cause–effect relationship. Nevertheless, one might envision that the selective destruction of tumor cells without detectable HLA class I antigen defects by HLA class I antigen–restricted, TA-specific CTL may lead to the outgrowth of malignant cells which escape T cell recognition because of HLA class I antigen defects. This conclusion is supported by results obtained in animal model systems utilizing different experimental conditions (Algarra et al., 2004). In studies which are still in progress, we have taken advantage of a melanoma cell line with a selective HLA-A2 antigen loss to study the role of selective pressure in the generation of lesions with HLA class I antigen defects. Administration to a SCID mouse of a mixture of the HLA-A2 antigen–negative melanoma cell line and of autologous melanoma cells with reconstituted HLA-A2 antigen expression resulted in the formation of a tumor which contained both HLA-A2 antigen–positive and HLA-A2 antigen–negative melanoma cells. However, a selective growth of HLA-A2 negative melanoma cells was observed following administration of an HLA-A2 antigen–restricted, TA-specific CTL line. Similarly, MART-1 negative melanoma cells have been selected when a melanoma tumor with heterogeneous MART-1 expression has been transplanted in SCID mice and then systemically treated with a MART-1-specific CD8þ T cell clone (Lozupone et al., 2003). Additional evidence in support of the role of selective pressure in the generation of malignant lesions with HLA class I antigen defects is provided by the marked differences in the antigenic phenotype of tumor cells grown in immunocompetent and in immunodeficient mice. First, metastatic lung colonies produced by a mouse fibrosarcoma cell line were MHC class I antigen positive when grown in immunodeficient athymic nude/nude mice and MHC class I antigen negative when grown in immunocompetent mice (Algarra et al., 2004). Analysis of the molecular mechanisms underlying the defective MHC class I antigen phenotype demonstrated a high frequency of LMP2, LMP7, LMP10, TAP1, TAP2, tapasin, and calnexin mRNA downregulation that was absent in metastases produced in immunocompetent mice. The antigenic differences we have described are paralleled by the shaping of immunogenicity of naturally occurring tumors by the host immune system, since chemically induced sarcomas in nude and SCID mice are more immunogenic than similar sarcoma cells induced in congenic immunocompetent mice (Algarra et al., 2004). The increased immunogenicity of tumors grown in immunodeficient mice may also contribute to the markedly higher frequency of lymphomas in perforin-deficient mice and of sarcomas

HLA and NKG2D Ligand Changes in Malignant Cells

215

in RAG2 and IFN- receptor deficient mice injected with the chemical carcinogen MCA (Shankaran et al., 2001). The latter results, which have not been corroborated by other investigators (Blankenstein and Qin, 2003; Qin and Blankenstein, 2004), have been taken as evidence to support the cancer immune surveillance theory (Burnet, 1970; Dunn et al., 2004). It is noteworthy that the available clinical and experimental evidence suggests, although does not prove, that selective pressure is imposed on tumor cells which have spontaneously arisen in a host or have been transplanted to a host, not only by CTL but also by NK cells. Thus, the association between poor prognosis and HLA class I antigen loss in malignant lesions described in several malignant diseases (Marincola et al., 2000) as well as the increased ability to metastasize of tumor cells with HLA class I antigen defects raise the possibility that NK cells cannot control tumor growth because the tumor cell population has been immunoselected for lack of NK cell–activating ligand expression by NK cell selective pressure. This possibility is supported by the resistance to NK cell–mediated lysis and lack of NK cell–activating ligand expression found in two melanoma cell lines isolated from recurrent metastases without detectable HLA class I antigen expression (Pende et al., 2002). Furthermore, the mock-transfected, but not the Rae-1-transfected, RMA lymphoma cells form tumors in the syngeneic B6 mice and depletion of NK cells in B6 mice abolishes the control of the growth of Rae-1-transfected RMA cells (Cerwenka et al., 2001). If immunoselective pressure plays a role in the generation of malignant lesions with HLA class I antigen defects, one might ask whether and how a host’s immune system reacts once the selective pressure has facilitated the outgrowth of tumor cells which have developed escape mechanisms from the ongoing immune response. Can the host immune system change the target of its immune response? If so, does this immune response select tumor cells that develop escape mechanisms? To the best of our knowledge, no study has addressed these questions in the clinical setting and in animal model systems. Nevertheless, in the clinical setting, there has been some suggestive evidence for an adaptive change of immune specificity against tumor cells which have developed a second escape mechanism. Coulie and his associates have demonstrated that the antigens recognized by the CTL isolated from the recurrent metastatic melanoma lesions are different from the antigens recognized by the CTL isolated from the primary lesions (Lehmann et al., 1995). Slingluff and his collaborators have shown a shifting of CTL specificity from HLA-A2-MART127–35 to HLA-A2-Tyr369–377D in two sequential metastatic melanoma tumors with distinct HLA class I abnormal phenotypes (Yamshchikov et al., 2005). We have reached similar conclusions by determining whether multiple defects in the molecules involved in the recognition of melanoma

216

Chien-Chung Chang et al.

Fig. 6 Role of immune selective pressure in shaping a tumor cell population bearing multiple classical and nonclassical HLA class I antigen and NK cell–activating ligand abnormalities. Malignant transformation of cells leads to the induction of NK cell–activating ligand expression. Under NK/NKT cell–mediated selective pressure, tumor cells with high levels of NK cell– activating ligand and low levels of classical HLA class I antigens are targeted and destroyed, resulting in the emergence of tumor cells bearing high levels of classical HLA class I antigen expression combined with either shed NK cell–activating ligands or low levels of NK cell– activating ligand expression. In subsequent stages of immune selection mediated by T cell selective pressure, tumor cells with the immunodominant classical HLA class I antigen-TA peptide complex expression are targeted and destroyed. This T cell–mediated selective pressure results in the emergence and expansion of tumor cells with downregulation and/or loss of the immunodominant classical HLA class I antigen-TA peptide complex combined with either shed NK cell–activating ligands or low levels of NK cell–activating ligand expression.

cells by HLA class I antigen–restricted, TA-specific CTL are frequent in melanoma cells. Our working hypothesis is that the presence of multiple defects would be compatible with the possibility that a host’s immune system is able to change its target when a resistant tumor cell population is selected (Fig. 6). The change in the target of the immune response facilitates the outgrowth of a tumor cell population which has developed an additional effective escape mechanism. In five melanoma cell lines derived from lesions exposed to strong T cell selective pressure, we have found

HLA and NKG2D Ligand Changes in Malignant Cells

217

that three of them have multiple defects in APM and HLA class I antigen presentation pathway. One (1259MEL) of these three cell lines possesses 2m loss, HLA-A2 antigen loss, and HLA-B and -C antigen downregulation (Chang et al., 2005). In light of immune system’s ability to reshape tumor cell populations over time, we favor the possibility that the multiple HLA class I antigen defects found in 1259MEL cells may have developed sequentially in the order of HLA-A2 antigen loss, HLA-B and -C antigen downregulation, and 2m loss, although the first two can also occur in the reverse order. The 2m gene mutation is postulated to be a late event, since HLA class I heavy chain abnormalities would provide 1259MEL cells with a mechanism to escape lysis by CTL which recognize immunodominant epitopes before a complete HLA class I antigen loss phenotype is acquired. On the other hand, if total HLA class I antigen loss caused by 2m loss occurs early, abnormalities in heavy chains would not be advantageous to 2m deficient 1259MEL cells because they are not expected to be recognized by CTL. The possible role of NK cell selective pressure in conjunction with T cell selective pressure in selecting for tumor cell variants with no or low NK cell ligand and HLA class I antigen expression is also indicated by the phenotype of the previously mentioned five melanoma cell lines derived from patients receiving T cell–based immunotherapy. Two of the five cell lines, which do not express HLA class I antigens because of 2m loss, also do not express MICA and ULBP antigens (Pende et al., 2002). This phenotype renders the two cell lines resistant to both CTL and NK cell–mediated lysis. In this regard, coexpression of classical HLA class I and the nonclassical HLA class I HLA-E, -F, and -G antigens may also be postulated to represent a tumor phenotype that survived multiple rounds of immune selection, since HLA-G expression on glioma (Wiendl et al., 2002), ovarian carcinoma (Malmberg et al., 2002), and RCC (Bukur et al., 2003) cells renders them resistant to lysis by HLA class I antigen restricted, TA-specific CTL. Finally, it is noteworthy that shedding of MIC antigens, in spite of high levels of expression on the tumor cell surface, may also represent a tumor phenotype that has been selected by NK cell–mediated immune responses, because the presence of sMIC blocks the interaction of NKG2D receptor with tumor membrane-bound MIC. It should be emphasized that the evidence in the human system we have summarized suggests, but does not prove, a cause–effect relationship between multiple rounds of immune selection and the appearance of multiple HLA class I and NK cell ligand abnormalities. Nevertheless, if correct, our view about the role of immune selection in the generation of a malignant cell phenotype implies that a tumor will grow only when it has developed enough escape mechanisms to avoid the range of immune responses a patient’s immune system is able to mount.

218

Chien-Chung Chang et al.

VI. HLA CLASS I ANTIGEN AND HLA CLASS I-TA–DERIVED PEPTIDE COMPLEX EXPRESSION BY MALIGNANT CELLS It is generally assumed that HLA class I antigen expression plays a crucial role in the interaction of tumor cells with HLA class I antigen restricted, TAspecific CTL. This assumption, which is taken as evidence to explain the association between HLA class I antigen downregulation in malignant lesions and poor prognosis described in some malignant diseases (Marincola et al., 2000), derives mainly from the resistance to CTL-mediated lysis of tumor cells which have lost the restricting HLA class I allospecificity. However, this evidence does not address the question of whether HLA class I antigen expression is sufficient for lysis of tumor cells by HLA class I antigen restricted, TA-specific CTL (Chang et al., 2004). Although this seems to be often the case, there are several reports which have documented lack of recognition of tumor cells by HLA class I antigen restricted, TAspecific CTL in spite of the expression of the restricting element at levels higher than the few molecules which are believed to be required for recognition of target cells by CTL. For example, HNSCC cell lines are generally resistant to CTL-mediated lysis, although the restricting HLA class I allospecificity is expressed at levels which should be sufficient for CTLmediated lysis to occur (T. Whiteside and R. L. Ferris, unpublished observations). Similarly, melanoma cells have been described which are resistant to Melan-A-specific CTL in the absence of detectable defects in the restricting HLA-A2 allospecificity (Meidenbauer et al., 2004). In the tumor cells that have been described thus far, resistance to CTL-mediated lysis is associated with abnormalities in some APM components. In some cases, lysis by CTL may be restored by incubating tumor cells with IFN- , which results in the upregulation of not only the level of HLA class I antigen expression but also APM component levels (R. L. Ferris and S. Ferrone, unpublished observations). The molecular mechanisms underlying these findings have not been characterized yet. It is our working hypothesis that the resistance of tumor cells to CTL-mediated lysis reflects the lack of presentation by the restricting HLA class I allospecificity of the peptide targeted by the CTL used as effector cells because of APM component defects. These defects are functional and not structural since they can be corrected by IFN- . While the effect of these defects on the generation of TA-derived peptide in vivo is not known, several lines of evidence suggest that variations in APM component expression can dramatically affect HLA class I-TA peptide complex expression and provide a mechanism for resistance to CTL-mediated lysis of tumor cells without detectable defects in HLA class I antigen expression. First, melanoma cells with no detectable defects in HLA class I antigens,

HLA and NKG2D Ligand Changes in Malignant Cells

219

APM components, and the melanoma-associated antigens MART1 are not recognized by HLA-A2-restricted, MART1-specific CTL following in vitro incubation with IFN- , since LMP2 and LMP7 expression in the immunoproteasome inhibits presentation of the MART1-derived peptide to CTL (Morel et al., 2000). Second, loss of the capacity to up-regulate both LMP immunosubunit and PA28 expression in response to in vitro incubation with IFN- is associated with the loss of the ability to present a peptide derived from the melanoma-associated antigens TRP-2, which requires expression of PA28, a component of the immunoproteasome (Sun et al., 2002). As a result, melanoma cells with this defect are resistant to HLA-A2-restricted, TRP2-specific CTL following exposure to IFN- . Lastly, abnormalities in TAP and/or tapasin expression, which are frequently present in malignant lesions, may have dramatic consequences on the repertoire of peptides presented on the cell membrane, since they each play a role in the qualitative selection of peptides to be transported into the ER and loaded onto HLA class I antigens, respectively. This notion is supported by (i) the increase in presentation of TAP-independent HLA class I antigen-peptide complexes in cells with TAP defects (Anderson et al., 1991; Hosken and Bevan, 1990; Wolfel et al., 2000) and (ii) the significant changes in the repertoire of peptides presented by HLA class I antigens in the absence of tapasin (Williams et al., 2002; Zarling et al., 2003). Complicating matters is the gene polymorphism exhibited by both TAP genes (McCluskey et al., 2004). In rats, TAP1 polymorphisms can lead to changes in the repertoire of peptides presented by MHC class I antigens (McCluskey et al., 2004); whether polymorphisms in TAP and tapasin can be reconciled with any functional variation in humans remains to be determined. The notion that subtle variations in the level of APM components may lead to the lack of presentation of HLA class I antigen-TA derived peptide complexes at sufficient density on tumor cells for recognition by TA-specific CTL parallels the situation observed in viral infections. A number of viruses have been shown to escape immune recognition and destruction by reversing, inhibiting, or redirecting APM component-mediated HLA class I antigen 2m-peptide complex assembly and intracellular transport, thereby reducing the number of HLA class I antigen-virus-derived peptide complexes on the cell surface (Ploegh, 1998). These findings have several implications. First, they may provide an explanation as to why the association of HLA class I antigen expression, as determined by staining malignant lesions with mAb, with a favorable clinical course of the disease is statistically significant but not absolute and why patients with a high level of HLA class I antigen expression in their malignant lesions may have an unfavorable clinical outcome of the disease. Second, the results we have discussed may provide a mechanism for the growth of tumor lesions without detectable HLA class I

220

Chien-Chung Chang et al.

antigen defects in patients who have developed a TA-specific CTL response following active specific immunotherapy. Third, the resistance of malignant cells without detectable HLA class I antigen defects to TA-specific CTL lysis raises the question of whether the level of a HLA class I allospecificity on the cell membrane can be used as a measure of the respective HLA class I allospecificity-TA derived peptide complex which mediates the recognition of tumor cells by HLA class I antigen restricted, TA-specific CTL. Clearly, when all HLA class I antigens or the HLA class I allospecificity presenting a peptide have been lost, the HLA class I allospecificity-TA derived peptide complex of interest is not expressed. There is only scanty information about the relationship between the level of HLA class I allospecificity-TA derived peptide complex and that of HLA class I allospecificity when the latter is expressed. The limited information is caused mostly by the very small number of available antibodies which recognize HLA class I allospecificity-TA derived peptide complexes (Chames et al., 2000; Denkberg et al., 2002; Lev et al., 2002). The results thus far obtained are compatible with the lack of association between level of a HLA class I allospecificity-TA derived peptide complex and that of the corresponding allospecificity. Several mechanisms may be envisioned for this finding. They include defects in TA expression and/ or generation, transport, and/or loading of TA derived peptides on 2m-HLA class I heavy chain complexes because of APM component abnormalities. Lastly, the results we have discussed emphasize the need to monitor the level of HLA class I allospecificity-TA derived peptide complex on tumor cells in order to minimize the limitations of the measurement of the level of a HLA class I allospecificity.

VII. CONCLUSIONS The data we have reviewed conclusively show that changes in the expression of classical and nonclassical HLA class I antigen and of NK cell– activating ligands occur frequently in malignant cells. These changes, which appear to be caused by a combination of mutations and immunoselection events, have been shown, although not in a conclusive way, to be associated with a prolongation of disease-free interval and survival in at least some malignant diseases. In each case, the association has been found to be statistically significant but not absolute. This finding is likely to reflect the interference of several immunological and nonimmunological variables which may confound and/or undermine the impact of HLA class I antigen and NK cell–activating ligand changes on the clinical course of the disease. Although convincingly proven to have profound effects on the interactions of malignant cells with immune cells in vitro, these changes have not yet

HLA and NKG2D Ligand Changes in Malignant Cells

221

been characterized for their effects on the interactions of malignant cells with the host immune system in vivo. This line of research may benefit from the application of real-time imaging and tissue-specific, conditional gene targeting technology. A number of molecular mechanisms underlying changes in HLA antigen phenotype have been identified in malignant cells. This information has not only contributed to our understanding of the mechanism regulating HLA antigen expression, but has also provided a useful background to design strategies to restore expression and/or function of HLA antigens in malignant cells, although their usefulness has been tested in a clinical setting only to a limited extent (Nabel et al., 1993; Stopeck et al., 2001). From a technical viewpoint, significant progress has been made in the development of reagents that can detect most, if not all, HLA antigens in formalin-fixed, paraffin-embedded tissue sections. Although the available IHC technique still suffers from the lack of equipment to quantitate the level of HLA antigen expression in tissue sections, the use of formalin-fixed, paraffin-embedded tissue sections is facilitating the implementation of studies to assess the clinical significance of changes in HLA class I antigen expression associated with malignant transformation of cells. Furthermore, it has been our personal experience that the use of formalin-fixed, paraffin-embedded tissue sections in IHC reactions has greatly facilitated pathologists’ involvement in collaborative studies to analyze HLA molecule expression in surgically removed lesions. It is expected that pathologists’ involvement in these studies will have a positive impact on the assessment of the usefulness of information about HLA antigen expression in malignant lesions for the evaluation of patients with malignant diseases and/or for the selection of those to be enrolled in clinical trials of T cell–based immunotherapy. The information we have reviewed raises a number of questions that need to be addressed in order to fully assess the role of HLA class I antigens and of NK cell–activating ligands in the interactions of malignant cells with the host immune system, in the pathogenesis and clinical course of malignant diseases, and in the outcome of T cell–based immunotherapy. First, which mechanisms underlie HLA class I antigen and APM component downregulation in malignant cells? Are they caused by epigenetic repression of the corresponding genes? Second, why are structural defects in the 2m gene present only in malignant cells of a given histotype, and why is the frequency of total and selective HLA class I antigen loss and downregulation so variable in different types of malignancies? Do these findings reflect the extent and/or type of immune selective pressure imposed on tumor cell populations in different malignancies? If this is the case, does the low frequency of total HLA class I antigen loss or downregulation in RCC and melanoma lesions argue against the current view about the role of

222

Chien-Chung Chang et al.

immunological events in the pathogenesis and clinical course of these malignant diseases? Third, do other variables such as extent and/or type of genetic instability confound the role of immune selective pressure in the generation of malignant lesions with HLA class I antigen defects? Can this explain the high frequency of HLA class I antigen defects in lesions of breast carcinoma and lung carcinoma, two types of cancer without apparent involvement of immunological events in the pathogenesis and clinical course of the disease? Fourth, are some HLA class I allospecificities preferentially lost or downregulated in malignant diseases? If so, does this finding reflect their preferential use as restricting elements in the TA-specific CTL immune response? Can this information facilitate the identification and isolation of clinically relevant TA? Fifth, in breast carcinoma, HER2/neu overexpression results in transcriptional downregulation of various APM components (Herrmann et al., 2004), suggesting a crosstalk between oncogenic factors and genes encoding APM components. Does this phenomenon apply to other tumor types expressing HER2/neu and/or a different oncogenic product? How does one oncogenic product regulate the transcription of various APM genes? Does it involve remodeling of selected chromatin regions through oncogenic product-mediated recruitment of DNA methyltransferases and/or histone deacetylases? Sixth, what is the role of innate immunity in the control of tumor growth and in the expression of nonclassical HLA class I antigens and NK cell–activating ligands by tumor cells? Which mechanisms are involved in the induction of these antigens on malignant cells? How do combined changes in the expression of classical and nonclassical HLA class I antigens and NK cell–activating ligands by malignant cells affect their interactions with the host immune system? Does the combination of defects frequently present in malignant cells reflect changes in the target of the host’s immune response to counteract the multiple escape mechanisms utilized by tumor cells? Lastly, is the existing evidence in support of the role of immunoselection in the generation of immunoresistant tumor cell variants sufficient to revive, in a conclusive way, the cancer immune surveillance theory? As already mentioned, a crucial question to be addressed is related to the clinical significance of changes in classical and nonclassical HLA class I antigen expression and NK cell–activating ligand expression in malignant lesions. It is hoped that the availability of mAb that stain formalin-fixed, paraffin-embedded tissue sections will facilitate studies to provide conclusive answers to these questions. These studies should not be limited to the analysis of histocompatibility antigen expression in malignant lesions, but should also take into account TA expression and TA-specific immune responses mounted by the host. Furthermore, in view of the lack of absolute correlation between HLA class I antigen and HLA class I antigen-TA derived peptide complex expression, as suggested by the previously

HLA and NKG2D Ligand Changes in Malignant Cells

223

mentioned evidence, can the correlation between HLA class I antigen expression in malignant lesions and clinical parameters be improved by measuring the expression of HLA class I antigen–TA derived–peptide complexes in malignant lesions? If this is proven to be the case, it will be important to develop a panel of probes to measure these complexes in malignant lesions. Furthermore, can analysis of classical and nonclassical HLA class I antigen and NK cell–activating ligand expression in malignant lesions provide clinically useful information which is not provided by the currently used diagnostic tests? Lastly, can restoration of HLA class I antigen and NK cell–activating ligand expression in malignant cells improve the clinical course of the disease and the clinical response to T cell–based immunotherapy? Answers to these questions will determine the extent to which malignant lesions will be tested for classical and nonclassical HLA class I antigen and NK cell–activating ligand expression in departments of pathology as well as the extent to which strategies to change the expression of these molecules in malignant cells which will be developed and used in a clinical setting. If proven to be clinically significant, HLA class I antigen and NK cell–activating ligand changes in malignant lesions are likely to provide an impetus for detailed structural and functional studies.

ACKNOWLEDGMENTS The authors thank Charlene DeMont, Marlene Kraebel, and Traci Jackson for their excellent secretarial assistance. This work was supported by PHS grants RO1 CA67108, P30 CA16056, and T32 CA85183 awarded by the National Cancer Institute, DHHS (to S. F.), by a Susan G. Komen Breast Cancer Foundation predoctoral fellowship (to C.-C. C.), and by a Department of Defense predoctoral fellowship (to M. C.).

REFERENCES Algarra, I., Garcia-Lora, A., Cabrera, T., Ruiz-Cabello, F., and Garrido, F. (2004). The selection of tumor variants with altered expression of classical and nonclassical MHC class I molecules: Implications for tumor immune escape. Cancer Immunol. Immunother. 53, 904–910. Amiot, L., Le Friec, G., Sebti, Y., Drenou, B., Pangault, C., Guilloux, V., Leleu, X., Bernard, M., Facon, T., and Fauchet, R. (2003). HLA-G and lymphoproliferative disorders. Semin. Cancer Biol. 13, 379–385. Amiot, L., Onno, M., Lamy, T., Dauriac, C., Le Prise, P. Y., Fauchet, R., and Drenou, B. (1998). Loss of HLA molecules in B lymphomas is associated with an aggressive clinical course. Br. J. Haematol. 100, 655–663. Anastassiou, G., Rebmann, V., Wagner, S., Bornfeld, N., and Grosse-Wilde, H. (2003). Expression of classic and nonclassic HLA class I antigens in uveal melanoma. Invest. Ophthalmol. Vis. Sci. 44, 2016–2019.

224

Chien-Chung Chang et al.

Anderson, K., Cresswell, P., Gammon, M., Hermes, J., Williamson, A., and Zweerink, H. (1991). Endogenously synthesized peptide with an endoplasmic reticulum signal sequence sensitizes antigen processing mutant cells to class I-restricted cell-mediated lysis. J. Exp. Med. 174, 489–492. Andersen, M. H., Pedersen, L. O., Becker, J. C., and Straten, P. T. (2001). Identification of a cytotoxic T lymphocyte response to the apoptosis inhibitor protein survivin in cancer patients. Cancer Res. 61, 869–872. Atkins, D., Breuckmann, A., Schmahl, G. E., Binner, P., Ferrone, S., Krummenauer, F., Storkel, S., and Seliger, B. (2004a). MHC class I antigen processing pathway defects, ras mutations, and disease stage in colorectal carcinoma. Int. J. Cancer 109, 265–273. Atkins, D., Ferrone, S., Schmahl, G. E., Storkel, S., and Seliger, B. (2004b). Down-regulation of HLA class I antigen processing molecules: An immune escape mechanism of renal cell carcinoma? J. Urol. 171, 885–889. Bagg, A. (2004). Role of molecular studies in the classification of lymphoma. Expert Rev. Mol. Diagn. 4, 83–97. Bahram, S. (2000). MIC genes: From genetics to biology. Adv. Immunol. 76, 1–60. Barton, L. F., Cruz, M., Rangwala, R., Deepe, G. S., Jr., and Monaco, J. J. (2002). Regulation of immunoproteasome subunit expression in vivo following pathogenic fungal infection. J. Immunol. 169, 3046–3052. Bauer, S., Groh, V., Wu, J., Steinle, A., Phillips, J. H., Lanier, L. L., and Spies, T. (1999). Activation of NK cells and T cells by NKG2D, a receptor for stress-inducible MICA. Science 285, 727–729. Bi, Y., Zhang, J., Hong, Z., Huang, J., Chen, J., Wang, X., and Ferrone, S. (2002). Association of HLA class I antigen expression in human hepatocellar carcinoma with improved prognosis. Tissue Antigens 2, 105–106. Bicknell, D. C., Rowan, A., and Bodmer, W. F. (1994). Beta 2-microglobulin gene mutations: A study of established colorectal cell lines and fresh tumors. Proc. Natl. Acad. Sci. USA 91, 4751–4755. Blankenstein, T., and Qin, Z. (2003). Chemical carcinogens as foreign bodies and some pitfalls regarding cancer immune surveillance. Adv. Cancer Res. 90, 179–207. Brady, C. S., Bartholomew, J. S., Burt, D. J., Duggan-Keen, M. F., Glenville, S., Telford, N., Little, A. M., Davidson, J. A., Jimenez, P., Ruiz-Cabello, F., Garrido, F., and Stern, P. L. (2000). Multiple mechanisms underlie HLA dysregulation in cervical cancer. Tissue Antigens 55, 401–411. Bortin, M. M., and Truitt, R. L. (1980). Proceedings of the 1st International Symposium on Alien Histocompatibility Antigens on Cancer Cells. Transplant. Proc. 12, 1–215. Bortin, M. M., and Truitt, R. L. (1981). Proceedings of the 2nd International Symposium on Alien Histocompatibility Antigens on Cancer Cells. Transplant. Proc. 13, 1751–2001. Bukur, J., Rebmann, V., Grosse-Wilde, H., Luboldt, H., Ruebben, H., Drexler, I., Sutter, G., Huber, C., and Seliger, B. (2003). Functional role of human leukocyte antigen-G up-regulation in renal cell carcinoma. Cancer Res. 63, 4107–4111. Burnet, F. M. (1970). The concept of immunological surveillance. Prog. Exp. Tumor Res. 13, 1–27. Cabrera, T., Lopez-Nevot, M. A., Gaforio, J. J., Ruiz-Cabello, F., and Garrido, F. (2003a). Analysis of HLA expression in human tumor tissues. Cancer Immunol. Immunother. 52, 1–9. Cabrera, T., Pedrajas, G., Cozar, J. M., Garrido, A., Vicente, J., Tallada, M., and Garrido, F. (2003b). HLA class I expression in bladder carcinomas. Tissue Antigens 62, 324–327. Campoli, M., Cheney, R., Rebmann, V., Grosse-Wilde, H., Worsham, M., Cabrera, T., Garrido, F., Tait, B., and Ferrone, S. (2004). HLA antigen changes and cancer: A report from the 13th International Histocompatibility Workshop. In ‘‘HLA 2004: Immunobiology of the Human

HLA and NKG2D Ligand Changes in Malignant Cells

225

MHC. Proceedings of the 13th International Histocompatibility Workshop and Congress’’ (J. Hansen and B. Dupont, Eds.), Vol. II, pp. 445–452. IHWG Press, Seattle, Washington. Carosella, E. D., Moreau, P., Le Maoult, J., Le Discorde, M., Dausset, J., and Rouas-Freiss, N. (2003). HLA-G molecules: From maternal-fetal tolerance to tissue acceptance. Adv. Immunol. 81, 199–252. Cerwenka, A., Baron, J. L., and Lanier, L. L. (2001). Ectopic expression of retinoic acid early inducible-1 gene (RAE-1) permits natural killer cell-mediated rejection of a MHC class I-bearing tumor in vivo.. Proc. Natl. Acad. Sci. USA 98, 11521–11526. Chames, P., Hufton, S. E., Coulie, P. G., Uchanska-Ziegler, B., and Hoogenboom, H. R. (2000). Direct selection of a human antibody fragment directed against the tumor T-cell epitope HLA-A1-MAGE-A1 from a nonimmunized phage-Fab library. Proc. Natl. Acad. Sci. USA 97, 7969–7974. Chan, J. K. C. (2000). Advances in immunohistochemistry: Impact on surgical pathology. In ‘‘Seminars in Diagnostic Pathology: Immunohistochemistry in Tumor Diagnosis’’ (D. J. Santa Criz, Ed.), Vol. 17, No. 3, pp. 170–177. W. B. Saunders Company, Philadelphia, DA. Chang, C. C., Campoli, M., and Ferrone, S. (2004). HLA class I antigen expression in malignant cells: Why does it not always correlate with CTL-mediated lysis? Curr. Opin. Immunol. 16, 644–650. Chang, C. C., Campoli, M., Restifo, N. P., Wang, X., and Ferrone, S. (2005). Immune selection of hot-spot B2M gene mutations, HLA-A2 allospecificity loss, and antigen processing machinery component downregulation in melanoma cells derived from recurrent metastases following immunotherapy. J. Immunol. in press. Chang, C. C., and Ferrone, S. (2003). HLA-G in melanoma: Can the current controversies be solved? Semin. Cancer Biol. 13, 361–369. Chang, C. C., Murphy, S. P., and Ferrone, S. (2003). Differential in vivo and in vitro HLA-G expression in melanoma cells: Potential mechanisms. Hum. Immunol. 64, 1057–1063. Chang, C. C., Pirozzi, G., Lombardi, M. L., and Ferrone, S. (2003). Identification of a tapasin gene mutation in a human melanoma cell line with HLA class I antigen downregluation. FASEB J. 17, C326. Chaperot, L., Manches, O., Mi, J. Q., Moine, A., Jacob, M. C., Gressin, R., Molens, J. P., Sotto, J. J., Leroux, D., Bensa, J. C., and Plumas, J. (2002). Differentiation of anti-tumor cytotoxic T lymphocytes from autologous peripheral blood lymphocytes in non-Hodgkin’s lymphomas. Br. J. Haematol. 119, 425–431. Chatterjee, A., Pulido, H. A., Koul, S., Beleno, N., Perilla, A., Posso, H., Manusukhani, M., and Murty, V. V. (2001). Mapping the sites of putative tumor suppressor genes at 6p25 and 6p21.3 in cervical carcinoma: Occurrence of allelic deletions in precancerous lesions. Cancer Res. 61, 2119–2123. Chen, H. L., Gabrilovich, D., Tampe, R., Girgis, K. R., Nadaf, S., and Carbone, D. P. (1996). A functionally defective allele of TAP1 results in loss of MHC class I antigen presentation in a human lung cancer. Nat. Genet. 13, 210–213. Chen, H. L., Gabrilovich, D., Virmani, A., Ratnani, I., Girgis, K. R., Nadaf-Rahrov, S., Fernandez-Vina, M., and Carbone, D. P. (1996). Structural and functional analysis of beta2 microglobulin abnormalities in human lung and breast cancer. Int. J. Cancer 67, 756–763. Cheng, N. C., Beitsma, M., Chan, A., Op den Camp, I., Westerveld, A., Pronk, J., and Versteeg, R. (1996a). Lack of class I HLA expression in neuroblastoma is associated with high N-myc expression and hypomethylation due to loss of the MEMO-1 locus. Oncogene 13, 1737–1744. Cheng, N. C., Chan, A. J., Beitsma, M. M., Speleman, F., Westerveld, A., and Versteeg, R. A. (1996b). Human modifier of methylation for class I HLA genes (MEMO-1) maps to chromosomal bands 1p35-36.1. Hum. Mol. Genet. 5, 309–317.

226

Chien-Chung Chang et al.

Chiorean, E. G., Dylla, S. J., Olsen, K., Lenvik, T., Soignier, Y., and Miller, J. S. (2003). BCR/ ABL alters the function of NK cells and the acquisition of killer immunoglobulin-like receptors (KIRs). Blood 101, 3527–3533. Chuang, W. L., Liu, H. W., and Chang, W. Y. (1990). Natural killer cell activity in patients with hepatocellular carcinoma relative to early development and tumor invasion. Cancer 65, 926–930. Cordon-Cardo, C., Fuks, Z., Drobnjak, M., Moreno, C., Eisenbach, L., and Feldman, M. (1991). Expression of HLA-A,B,C antigens on primary and metastatic tumor cell populations of human carcinomas. Cancer Res. 51, 6372–6380. Cosman, D., Mullberg, J., Sutherland, C. L., Chin, W., Armitage, R., Fanslow, W., Kubin, M., and Chalupny, N. J. (2001). ULBPs, novel MHC class I-related molecules, bind to CMV glycoprotein UL16 and stimulate NK cytotoxicity through the NKG2D receptor. Immunity 14, 123–133. Crowley, N. J., Darrow, T. L., Quinn-Allen, M. A., and Seigler, H. F. (1991). MHC-restricted recognition of autologous melanoma by tumor-specific cytotoxic T cells. Evidence for restriction by a dominant HLA-A allele. J. Immunol. 146, 1692–1699. Dabbs, D. J. (2002). ‘‘Diagnostic Immunohistochemistry.’’ Churchill Livingstone, New York. Dammrich, J., Muller-Hermelink, H. K., Mattner, A., Buchwald, J., and Ziffer, S. (1990). Histocompatibility antigen expression in pulmonary carcinomas as indication of differentiation and of special subtypes. Cancer 65, 1942–1954. Demanet, C., Mulder, A., Deneys, V., Worsham, M. J., Maes, P., Claas, F. H., and Ferrone, S. (2004). Down-regulation of HLA-A and HLA-Bw6, but not HLA-Bw4, allospecificities in leukemic cells: An escape mechanism from CTL and NK attack? Blood 103, 3122–3130. Denkberg, G., Cohen, C. J., Lev, A., Chames, P., Hoogenboom, H. R., and Reiter, Y. (2002). Direct visualization of distinct T cell epitopes derived from a melanoma tumor-associated antigen by using human recombinant antibodies with MHC-restricted T cell receptor-like specificity. Proc. Natl. Acad. Sci. USA 99, 9421–9426. Di Giorgio, A., Botti, C., Tocchi, A., Mingazzini, P., and Flammia, M. (1992). The influence of tumor lymphocytic infiltration on long-term survival of surgically treated colorectal cancer patients. Int. Surg. 77, 256–260. Dissemond, J., Goette, P., Moers, J., Lindeke, A., Goos, M., Ferrone, S., and Wagner, S. N. (2003a). Immunoproteasome subunits LMP2 and LMP7 downregulation in primary malignant melanoma lesions: Association with lack of spontaneous regression. Melanoma Res. 13, 371–377. Dissemond, J., Gotte, P., Mors, J., Lindeke, A., Goos, M., Ferrone, S., and Wagner, S. N. (2003b). Association of TAP1 downregulation in human primary melanoma lesions with lack of spontaneous regression. Melanoma Res. 13, 253–258. Dissemond, J., Kothen, T., Mors, J., Weimann, T. K., Lindeke, A., Goos, M., and Wagner, S. N. (2003c). Downregulation of tapasin expression in progressive human malignant melanoma. Arch. Dermatol. Res. 295, 43–49. Dolcetti, R., Viel, A., Doglioni, C., Russo, A., Guidoboni, M., Capozzi, E., Vecchiato, N., Macri, E., Fornasarig, M., and Boiocchi, M. (1999). High prevalence of activated intraepithelial cytotoxic T lymphocytes and increased neoplastic cell apoptosis in colorectal carcinomas with microsatellite instability. Am. J. Pathol. 154, 1805–1813. Doubrovina, E. S., Doubrovin, M. M., Vider, E., Sisson, R. B., O’Reilly, R. J., Dupont, B., and Vyas, Y. M. (2003). Evasion from NK cell immunity by MHC class I chain-related molecules expressing colon adenocarcinoma. J. Immunol. 171, 6891–6899. Drenou, B., Le Friec, G., Bernard, M., Pangault, C., Grosset, J. M., Lamy, T., Fauchet, R., and Amiot, L. (2002). Major histocompatibility complex abnormalities in non-Hodgkin lymphomas. Br. J. Haematol. 119, 417–424.

HLA and NKG2D Ligand Changes in Malignant Cells

227

Dunn, G. P., Bruce, A. T., Ikeda, H., Old, L. J., and Schreiber, R. D. (2002). Cancer immunoediting: From immunosurveillance to tumor escape. Nat. Immunol. 3, 991–998. Dunn, G. P., Old, L. J., and Schreiber, R. D. (2004). The three Es of cancer immunoediting. Annu. Rev. Immunol. 22, 329–360. D’Urso, C. M., Wang, Z. G., Cao, Y., Tatake, R., Zeff, R. A., and Ferrone, S. (1991). Lack of HLA class I antigen expression by cultured melanoma cells FO-1 due to a defect in B2m gene expression. J. Clin. Invest 87, 284–292. Ferrone, S. (Guest Ed.) (2002). Tumor immune escape. Semin. Cancer Biol. 12, 1–86. Francke, U., and Pellegrino, M. A. (1977). Assignment of the major histocompatibility complex to a region of the short arm of human chromosome 6. Proc. Natl. Acad. Sci. USA 74, 1147–1151. Friese, M. A., Platten, M., Lutz, S. Z., Naumann, U., Aulwurm, S., Bischof, F., Buhring, H. J., Dichgans, J., Rammensee, H. G., Steinle, A., and Weller, M. (2003). MICA/NKG2Dmediated immunogene therapy of experimental gliomas. Cancer Res. 63, 8996–9006. Fonsatti, E., Sigalotti, L., Coral, S., Colizzi, F., Altomonte, M., and Maio, M. (2003). Methylation-regulated expression of HLA class I antigens in melanoma. Int. J. Cancer 105, 430–431. Fowler, N. L., and Frazer, I. H. (2004). Mutations in TAP genes are common in cervical carcinomas. Gynecol. Oncol. 92, 914–921. Fukusato, T., Gerber, M. A., Thung, S. N., Ferrone, S., and Schaffner, F. (1986). Expression of HLA class I antigens on hepatocytes in liver disease. Am. J. Pathol. 123, 264–270. Garrido, F., and Algarra, I. (2001). MHC antigens and tumor escape from immune surveillance. Adv. Cancer Res. 83, 117–158. Garrido, F., Cabrera, T., Accolla, R. S., Bensa, J. C., Bodmer, W., Dohr, G., Drouet, M., Fauchet, R., Ferrara, G. B., Ferrone, S., Giacomini, P., Kageshita, T., Koopman, L., Maio, M., Marinicola, M., Mazzilli, C., Morel, P. A., Murray, A., Papasteriades, C. R. H., Salvaneschi, L., Stern, P. L., and Ziegler, A. (1997). HLA and cancer: Twelfth International Histocompatibility Workshop study. In ‘‘Genetic Diversity of HLA. Functional and Medical Implication. Proc. Twelfth International Histocompatibility Workshop and Conference’’ (D. Charron, Ed.), Vol. II, pp. 445–452. EDK, Sevres, France. Gattoni-Celli, S., Kirsch, K., Timpane, R., and Isselbacher, K. J. (1992). Beta 2-microglobulin gene is mutated in a human colon cancer cell line (HCT) deficient in the expression of HLA class I antigens on the cell surface. Cancer Res. 52, 1201–1204. Geertsen, R., Boni, R., Blasczyk, R., Romero, P., Betts, D., Rimoldi, D., Hong, X., Laine, E., Willers, J., and Dummer, R. (2002). Loss of single HLA class I allospecificities in melanoma cells due to selective genomic abbreviations. Int. J. Cancer 99, 82–87. Giorda, E., Sibilio, L., Martayan, A., Moretti, S., Venturo, I., Mottolese, M., Ferrara, G. B., Cappellacci, S., Eibenschutz, L., Catricala, C., Grammatico, P., and Giacomini, P. (2003). The antigen processing machinery of class I human leukocyte antigens: Linked patterns of gene expression in neoplastic cells. Cancer Res. 63, 4119–4127. Girardi, M., Oppenheim, D. E., Steele, C. R., Lewis, J. M., Glusac, E., Filler, R., Hobby, P., Sutton, B., Tigelaar, R. E., and Hayday, A. C. (2001). Regulation of cutaneous malignancy by gammadelta T cells. Science 294, 605–609. Goodfellow, P. N., Jones, E. A., Van Heyningen, V., Solomon, E., Bobrow, M., Miggiano, V., and Bodmer, W. F. (1975). The beta2-microglobulin gene is on chromosome 15 and not in the HLA region. Nature 254, 267–269. Griffioen, M., Ouwerkerk, I. J., Harten, V., and Schrier, P. I. (2000). HLA-B locus-specific downregulation in human melanoma requires enhancer A as well as a sequence element located downstream of the transcription initiation site. Immunogenetics 52, 121–128. Groh, V., Rhinehart, R., Secrist, H., Bauer, S., Grabstein, K. H., and Spies, T. (1999). Broad tumor-associated expression and recognition by tumor-derived gamma delta T cells of MICA and MICB. Proc. Natl. Acad. Sci. USA 96, 6879–6884.

228

Chien-Chung Chang et al.

Groh, V., Wu, J., Yee, C., and Spies, T. (2002). Tumor-derived soluble MIC ligands impair expression of NKG2D and T-cell activation. Nature 419, 734–738. Guerry, D. (1998). The cellular immunobiology of melanoma. In ‘‘Cutaneous Oncology: Pathophysiology, Diagnosis, and Management’’ (S. J. Miller and M. E. Maloney, Eds.), pp. 211–217. Blackwell Science, Maiden, MA. Gumy-Pause, F., Wacker, P., and Sappino, A. P. (2004). ATM gene and lymphoid malignancies. Leukemia 18, 238–242. Herrmann, F., Lehr, H. A., Drexler, I., Sutter, G., Hengstler, J., Wollscheid, U., and Seliger, B. (2004). HER-2/neu-mediated regulation of components of the MHC class I antigen-processing pathway. Cancer Res. 64, 215–220. Hicklin, D. J., Marincola, F. M., and Ferrone, S. (1999). HLA class I antigen downregulation in human cancers: T-cell immunotherapy revives an old story. Mol. Med. Today 5, 178–186. Hicklin, D. J., Wang, Z., Arienti, F., Rivoltini, L., Parmiani, G., and Ferrone, S. (1998). beta2microglobulin mutations, HLA class I antigen loss, and tumor progression in melanoma. J. Clin. Invest 101, 2720–2729. Hosken, N. A., and Bevan, M. J. (1990). Defective presentation of endogenous antigen by a cell line expressing class I molecules. Science 248, 367–370. Huland, E., and Heinzer, H. (2003). Renal cell carcinoma: Novel treatments for advanced disease. Curr. Opin. Urol. 13, 451–456. Ikeda, H., Lethe, B., Lehmann, F., van Baren, N., Baurain, J. F., de Smet, C., Chambost, H., Vitale, M., Moretta, A., Boon, T., and Coulie, P. G. (1997). Characterization of an antigen that is recognized on a melanoma showing partial HLA loss by CTL expressing an NK inhibitory receptor. Immunity 6, 199–208. Imreh, M. P., Zhang, Q. J., de Campos-Lima, P. O., Imreh, S., Krausa, P., Browning, M., Klein, G., and Masucci, M. G. (1995). Mechanisms of allele-selective down-regulation of HLA class I in Burkitt’s lymphoma. Int. J. Cancer 62, 90–96. Jager, E., Ringhoffer, M., Altmannsberger, M., Arand, M., Karbach, J., Jager, D., Oesch, F., and Knuth, A. (1997). Immunoselection in vivo: Independent loss of MHC class I and melanocyte differentiation antigen expression in metastatic melanoma. Int. J. Cancer 71, 142–147. Jimenez, P., Cabrera, T., Mendez, R., Esparza, C., Cozar, J. M., Tallada, M., Lopez-Nevot, M. A., Ruiz-Cabello, F., and Garrido, F. A. (2001). Nucleotide insertion in exon 4 is responsible for the absence of expression of an HLA-A*0301 allele in a prostate carcinoma cell line. Immunogenetics 53, 606–610. Jimenez, P., Canton, J., Collado, A., Cabrera, T., Serrano, A., Real, L. M., Garcia, A., RuizCabello, F., and Garrido, F. (1999). Chromosome loss is the most frequent mechanism contributing to HLA haplotype loss in human tumors. Int. J. Cancer 83, 91–97. Jimenez, P., Canton, J., Concha, A., Cabrera, T., Fernandez, M., Real, L. M., Garcia, A., Serrano, A., Garrido, F., and Ruiz-Cabello, F. (2000). Microsatellite instability analysis in tumors with different mechanisms for total loss of HLA expression. Cancer Immunol. Immunother. 48, 684–690. Jinushi, M., Takehara, T., Tatsumi, T., Kanto, T., Groh, V., Spies, T., Kimura, R., Miyagi, T., Mochizuki, K., Sasaki, Y., and Hayashi, N. (2003). Expression and role of MICA and MICB in human hepatocellular carcinomas and their regulation by retinoic acid. Int. J. Cancer 104, 354–361. Kageshita, T., Hirai, S., Ono, T., Hicklin, D. J., and Ferrone, S. (1999). Downregulation of HLA class I antigen-processing molecules in malignant melanoma: Association with disease progression. Am. J. Pathol. 154, 745–754. Kageshita, T., Ono, T., Hirai, S., Yoshii, A., Kimura, T., Nakakuma, H., Horikawa, K., Takatsuki, K., and Ferrone, S. (1992). Ganglioside, adhesion molecule, and HLA antigen expression in basal cell carcinoma lesions. Cancer Res. 52, 3201–3207.

HLA and NKG2D Ligand Changes in Malignant Cells

229

Kageshita, T., Wang, Z., Calorini, L., Yoshii, A., Kimura, T., Ono, T., Gattoni-Celli, S., and Ferrone, S. (1993). Selective loss of human leukocyte class I allospecificities and staining of melanoma cells by monoclonal antibodies recognizing monomorphic determinants of class I human leukocyte antigens. Cancer Res. 53, 3349–3354. Kasahara, K., Taguchi, T., Yamasaki, I., Kamada, M., Yuri, K., and Shuin, T. (2002). Detection of genetic alterations in advanced prostate cancer by comparative genomic hybridization. Cancer Genet. Cytogenet. 137, 59–63. Kelly, L. M., and Gilliland, D. G. (2002). Genetics of myeloid leukemias. Annu. Rev. Genomics Hum. Genet. 3, 179–198. Khong, H. T., and Restifo, N. P. (2002). Natural selection of tumor variants in the generation of ‘‘tumor escape’’ phenotypes. Nat. Immunol. 3, 999–1005. Klein, B., Klein, T., Konichezky, M., Nyska, A., Livini, E., Levine, I., Zamir, R., Kooperman, O., and Lurie, H. (1990). The expression of HLA class I antigens in germ cell testicular cancer. Am. J. Clin. Pathol. 93, 202–207. Koopman, L. A., Corver, W. E., van der Slik, A. R., Giphart, M. J., and Fleuren, G. J. (2000). Multiple genetic alterations cause frequent and heterogeneous human histocompatibility leukocyte antigen class I loss in cervical cancer. J. Exp. Med. 191, 961–976. Kuckelkorn, U., Ruppert, T., Strehl, B., Jungblut, P. R., Zimny-Arndt, U., Lamer, S., Prinz, I., Drung, I., Kloetzel, P. M., Kaufmann, S. H., and Steinhoff, U. (2002). Link between organspecific antigen processing by 20S proteasomes and CD8 (þ) T cell-mediated autoimmunity. J. Exp. Med. 195, 983–990. Kurokohchi, K., Carrington, M., Mann, D. L., Simonis, T. B., Alexander-Miller, M. A., Feinstone, S. M., Akatsuka, T., and Berzofsky, J. A. (1996). Expression of HLA class I molecules and the transporter associated with antigen processing in hepatocellular carcinoma. Hepatology 23, 1181–1188. Lampson, L. A., Fisher, C. A., and Whelan, J. P. (1983). Striking paucity of HLA-A, -B, -C and beta 2-microglobulin on human neuroblastoma cell lines. J. Immunol. 30, 2471–2478. Le, Y. S., Kim, T. E., Kim, B. K., Park, Y. G., Kim, G. M., Jee, S. B., Ryu, K. S., Kim, I. K., and Kim, J. W. (2002). Alterations of HLA class I and class II antigen expressions in borderline, invasive, and metastatic ovarian cancers. Exp. Mol. Med. 34, 18–26. Lee, N., and Geraghty, D. E. (2003). HLA-F surface expression on B cell and monocyte cell lines is partially independent from tapasin and completely independent from TAP. J. Immunol. 171, 5264–5271. Lehmann, F., Marchand, M., Hainaut, P., Pouillart, P., Sastre, X., Ikeda, H., Boon, T., and Coulie, P. G. (1995). Differences in the antigens recognized by cytolytic T cells on two successive metastases of a melanoma patient are consistent with immune selection. Eur. J. Immunol. 25, 340–347. Lengauer, C., Kinzler, K. W., and Vogelstein, B. (1997). Genetic instability in colorectal cancers. Nature 386, 623–627. Lev, A., Denkberg, G., Cohen, C. J., Tzukerman, M., Skorecki, K. L., Chames, P., and Hoogenboom, H. R. (2002). Isolation and characterization of human recombinant antibodies endowed with the antigen-specific, major histocompatibility complex-restricted specificity of T cells directed toward the widely expressed tumor T-cell epitopes of the telomerase catalytic subunit. Cancer Res. 62, 3184–3194. List, A. F., Spier, C. M., Miller, T. P., and Grogan, T. M. (1993). Deficient tumor-infiltrating T-lymphocyte response in malignant lymphoma: Relationship to HLA expression and host immunocompetence. Leukemia 7, 398–403. Lopez-Nevot, M. A., Esteban, F., Ferron, A., Gutierrez, J., Oliva, M. R., Romero, C., Huelin, C., Ruiz-Cabello, F., and Garrido, F. (1989). HLA class I gene expression on human primary tumors and autologous metastases: Demonstration of selective losses of HLA antigens on colorectal, gastric, and laryngeal carcinomas. Br. J. Cancer 59, 221–226.

230

Chien-Chung Chang et al.

Lozupone, F., Rivoltini, L., Luciani, F., Venditti, M., Lugini, L., Cova, A., Squarcina, P., Parmiani, G., Belardelli, F., and Fais, S. (2003). Adoptive transfer of an anti-MART-1(2735)-specific CD8þ T cell clone leads to immunoselection of human melanoma antigen-loss variants in SCID mice. Eur. J. Immunol. 33, 556–566. Maleno, I., Lopez-Nevot, M. A., Cabrera, T., Salinero, J., and Garrido, F. (2002). Multiple mechanisms generate HLA class I altered phenotypes in laryngeal carcinomas: High frequency of HLA haplotype loss associated with loss of heterozygosity in chromosome region 6p21. Cancer Immunol. Immunother. 51, 389–396. Malmberg, K. J., Levitsky, V., Norell, H., de Matos, C. T., Carlsten, M., Schedvins, K., Rabbani, H., Moretta, A., Soderstrom, K., Levitskaya, J., and Kiessling, R. (2002). IFNgamma protects short-term ovarian carcinoma cell lines from CTL lysis via a CD94/ NKG2A-dependent mechanism. J. Clin. Invest 110, 1515–1523. Marin, R., Ruiz-Cabello, F., Pedrinaci, S., Mendez, R., Jimenez, P., Geraghty, D. E., and Garrido, F. (2003). Analysis of HLA-E expression in human tumors. Immunogenetics 54, 767–775. Marincola, F. M., Jaffee, E. M., Hicklin, D. J., and Ferrone, S. (2000). Escape of human solid tumors from T cell recognition: Molecular mechanisms and functional significance. Adv. Immunol. 74, 181–273. Marincola, F. M., Shamamian, P., Simonis, T. B., Abati, A., Hackett, J., O’Dea, T., Fetsch, P., Yannelli, J., Restifo, N. P., and Mule, J. J. (1994). Locus-specific analysis of human leukocyte antigen class I expression in melanoma cell lines. J. Immunother. Emphasis. Tumor Immunol. 16, 13–23. McCluskey, J., Rossjohn, J., and Purcell, A. W. (2004). TAP genes and immunity. Curr. Opin. Immunol. 16, 651–659. McEvoy, C. R., Seshadri, R., Morley, A. A., and Firgaira, F. A. (2002). Frequency and genetic basis of MHC, beta-2-microglobulin and MEMO-1 loss of heterozygosity in sporadic breast cancer. Tissue Antigens 60, 235–243. Meidenbauer, N., Zippelius, A., Pittet, M. J., Laumer, M., Vogl, S., Heymann, J., Rehli, M., Seliger, B., Schwarz, S., Gal, F. A., Dietrich, P. Y., Andreesen, R., Romero, P., and Mackensen, A. (2004). High frequency of functionally active Melan-a-specific T cells in a patient with progressive immunoproteasome-deficient melanoma. Cancer Res. 64, 6319–6326. Morel, S., Levy, F., Burlet-Schiltz, O., Brasseur, F., Probst-Kepper, M., Peitrequin, A. L., Monsarrat, B., Van Velthoven, R., Cerottini, J. C., Boon, T., Gairin, J. E., and Van den Eynde, B. J. (2000). Processing of some antigens by the standard proteasome but not by the immunoproteasome results in poor presentation by dendritic cells. Immunity 12, 107–117. Morris, M. A., and Ley, K. (2004). Trafficking of natural killer cells. Curr. Mol. Med. 4, 431–438. Nabel, G. J., Nabel, E. G., Yang, Z., Fox, B. A., Plautz, G. E., Gao, X., Huang, L., Shu, S., Gordon, D., and Chang, A. E. (1993). Direct gene transfer with DNA-liposome complexes in melanoma: Expression, biologic activity, and lack of toxicity in humans. Proc. Natl. Acad. Sci. USA 90, 11307–11311. Natali, P. G., Viora, M., Nicotra, M. R., Giacomini, P., Bigotti, A., and Ferrone, S. (1983). Antigenic heterogeneity of skin tumors of nonmelanocyte origin: Analysis with monoclonal antibodies to tumor-associated antigens and to histocompatibility antigens. J. Natl. Cancer. Inst. 71, 439–447. Nie, Y., Yang, G., Song, Y., Zhao, X., So, C., Liao, J., Wang, L. D., and Yang, C. S. (2001). DNA hypermethylation is a mechanism for loss of expression of the HLA class I genes in human esophageal squamous cell carcinomas. Carcinogenesis 22, 1615–1623. O’Callaghan, C. A., and Bell, J. I. (1998). Structure and function of the human MHC class Ib molecules HLA-E, HLA-F and HLA-G. Immunol. Rev. 163, 129–138.

HLA and NKG2D Ligand Changes in Malignant Cells

231

Ogino, T., Bandoh, N., Hayashi, T., Miyokawa, N., Harabuchi, Y., and Ferrone, S. (2003a). Association of tapasin and HLA class I antigen downregulation in primary maxillary sinus squamous carcinoma lesions with reduced patients’ survival. Clin. Cancer. Res. 9, 4043–4051. Ogino, T., Wang, X., Kato, S., Miyokawa, N., Harabuchi, Y., and Ferrone, S. (2003b). Endoplasmic reticulum chaperone-specific monoclonal antibodies for flow cytometry and immunohistochemical staining. Tissue Antigens 62, 385–393. Ogretmen, B., McCauley, M. D., and Safa, A. R. (1998). Molecular mechanisms of loss of beta 2-microglobulin expression in drug-resistant breast cancer sublines and its involvement in drug resistance. Biochemistry 3, 11679–11691. Ono, K., Yamanaga, Y., Yamamoto, K., Koga, S. I., Nishimura, J., and Nawata, H. (1996). Natural killing activities in chronic liver diseases and hepatocellular carcinoma. J. Clin. Immunol. 16, 41–45. Pardoll, D. (2003). Does the immune system see tumors as foreign or self? Annu. Rev. Immunol. 21, 807–839. Parmiani, G., Castelli, C., Dalerba, P., Mortarini, R., Rivoltini, L., Marincola, F. M., and Anichini, A. (2002). Cancer immunotherapy with peptide-based vaccines: What have we achieved? Where are we going? J. Natl. Cancer Inst. 94, 805–818. Paul, P., Rouas-Freiss, N., Moreau, P., Cabestre, F. A., Menier, C., Khalil-Daher, I., Pangault, C., Onno, M., Fauchet, R., Martinez-Laso, J., Morales, P., Villena, A. A., Giacomini, P., Natali, P. G., Frumento, G., Ferrara, G. B., McMaster, M., Fisher, S., Schust, D., Ferrone, S., Dausset, J., Geraghty, D., and Carosella, E. D. (2000). HLA-G, -E, -F preworkshop: Tools and protocols for analysis of nonclassical class I genes transcription and protein expression. Hum. Immunol. 61, 1177–1195. Pende, D., Rivera, P., Marcenaro, S., Chang, C. C., Biassoni, R., Conte, R., Kubin, M., Cosman, D., Ferrone, S., Moretta, L., and Moretta, A. (2002). Major histocompatibility complex class I-related chain A and UL16-binding protein expression on tumor cell lines of different histotypes: Analysis of tumor susceptibility to NKG2D-dependent natural killer cell cytotoxicity. Cancer Res. 62, 6178–6186. Perosa, F., Luccarelli, G., Prete, M., Favoino, E., Ferrone, S., and Dammacco, F. (2003). Beta 2-microglobulin-free HLA class I heavy chain epitope mimicry by monoclonal antibody HC-10-specific peptide. J. Immunol. 171, 1918–1926. Pierson, B. A., and Miller, J. S. (1996). CD56þbright and CD56þdim natural killer cells in patients with chronic myelogenous leukemia progressively decrease in number, respond less to stimuli that recruit clonogenic natural killer cells, and exhibit decreased proliferation on a per-cell basis. Blood 88, 2279–2287. Ploegh, H. L. (1998). Viral strategies of immune evasion. Science 280, 248–253. Puppo, F., Indiveri, F., Scudeletti, M., and Ferrone, S. (1997). Soluble HLA antigens: New roles and uses. Immunol. Today 18, 154–155. Qin, Z., and Blankenstein, T. (2004). A cancer immunosurveillance controversy. Nat. Immunol. 5, 3–4. Rebmann, V., Regel, J., Stolke, D., and Grosse-Wilde, H. (2003). Secretion of sHLA-G molecules in malignancies. Semin. Cancer Biol. 13, 371–377. Restifo, N. P., Marincola, F. M., Kawakami, Y., Taubenberger, J., Yannelli, J. R., and Rosenberg, S. A. (1996). Loss of functional beta 2-microglobulin in metastatic melanomas from five patients receiving immunotherapy. J. Natl. Cancer Inst. 88, 100–108. Riker, A., Cormier, J., Panelli, M., Kammula, U., Wang, E., Abati, A., Fetsch, P., Lee, K. H., Steinberg, S., Rosenberg, S., and Marincola, F. (1999). Immune selection after antigenspecific immunotherapy of melanoma. Surgery 126, 112–120. Ripberger, E., Linnebacher, M., Schwitalle, Y., Gebert, J., and von Knebel Doeberitz, M. (2003). Identification of an HLA-A0201-restricted CTL epitope generated by a tumor-specific

232

Chien-Chung Chang et al.

frameshift mutation in a coding microsatellite of the OGT gene. J. Clin. Immunol. 23, 415–423. Ritz, U., Momburg, F., Pilch, H., Huber, C., Maeurer, M. J., and Seliger, B. (2001). Deficient expression of components of the MHC class I antigen processing machinery in human cervical carcinoma. Int. J. Oncol. 19, 1211–1220. Romero, J. M., Jimenez, P., Cabrera, T., Cozar, J. M., Pedrinaci, S., Tallada, M., Garrido, F., and Ruiz-Cabello, F. (2005). Coordinated downregulation of the antigen presentation machinery and HLA class I/beta2-microglobulin complex is responsible for HLA-ABC loss in bladder cancer. Int. J. Cancer 113, 605–610. Rosenberg, S. A., Packard, B. S., Aebersold, P. M., Solomon, D., Topalian, S. L., Toy, S. T., Simon, P., Lotze, M. T., Yang, J. C., Seipp, C. A., Simpson, C., Carter, C., Bock, S., Schwartzentruber, D., Wei, J. P., and White, D. E. (1988). Use of tumor-infiltrating lymphocytes and interleukin-2 in the immunotherapy of patients with metastatic melanoma. A preliminary report. N. Engl. J. Med. 319, 1676–1680. Ruiz-Cabello, F., Lopez Nevot, M. A., Gutierrez, J., Oliva, M. R., Romero, C., Ferron, A., Esteban, F., Huelin, C., Piris, M. A., and Rivas, C. (1989). Phenotypic expression of histocompatibility antigens in human primary tumors and metastases. Clin. Exp. Metastasis 7, 213–226. Rouas-Freiss, N., Moreau, P., Menier, C., and Carosella, E. D. (2003). HLA-G in cancer: A way to turn off the immune system. Semin. Cancer Biol. 5, 325–336. Salih, H. R., Antropius, H., Gieseke, F., Lutz, S. Z., Kanz, L., Rammensee, H. G., and Steinle, A. (2003). Functional expression and release of ligands for the activating immunoreceptor NKG2D in leukemia. Blood 102, 1389–1396. Scupoli, M. T., Sartoris, S., Tosi, G., Ennas, M. G., Nicolis, M., Cestari, T., Zamboni, G., Martignoni, G., Lemoine, N. R., Scarpa, A., and Accolla, R. S. (1996). Expression of MHC class I and class II antigens in pancreatic adenocarcinomas. Tissue Antigens 48, 301–311. Seliger, B., Abken, H., and Ferrone, S. (2003a). HLA-G and MIC expression in tumors and their role in anti-tumor immunity. Trends Immunol. 24, 82–87. Seliger, B., Atkins, D., Bock, M., Ritz, U., Ferrone, S., Huber, C., and Storkel, S. (2003b). Characterization of human lymphocyte antigen class I antigen-processing machinery defects in renal cell carcinoma lesions with special emphasis on transporter-associated with antigenprocessing down-regulation. Clin. Cancer Res. 9, 1721–1727. Seliger, B., Cabrera, T., Garrido, F., and Ferrone, S. (2002). HLA class I antigen abnormalities and immune escape by malignant cells. Semin. Cancer Biol. 12, 3–13. Seliger, B., Maeurer, M. J., and Ferrone, S. (2000). Antigen-processing machinery breakdown and tumor growth. Immunol. Today 21, 455–464. Seliger, B., Ritz, U., Abele, R., Bock, M., Tampe, R., Sutter, G., Drexler, I., Huber, C., and Ferrone, S. (2001). Immune escape of melanoma: First evidence of structural alterations in two distinct components of the MHC class I antigen processing pathway. Cancer Res. 61, 8647–8650. Sernee, M. F., Ploegh, H. L., and Schust, D. J. (1998). Why certain antibodies cross-react with HLA-A and HLA-G: Epitope mapping of two common MHC class I reagents. Mol. Immunol. 35, 177–188. Serrano, A., Tanzarella, S., Lionello, I., Mendez, R., Traversari, C., Ruiz-Cabello, F., and Garrido, F. (2001). Reexpression of HLA class I antigens and restoration of antigen-specific CTL response in melanoma cells following 5-aza-20 -deoxycytidine treatment. Int. J. Cancer 94, 243–251. Shankaran, V., Ikeda, H., Bruce, A. T., White, J. M., Swanson, P. E., Old, L. J., and Schreiber, R. D. (2001). IFNgamma and lymphocytes prevent primary tumor development and shape tumor immunogenicity. Nature 410, 1107–1111. Shiina, T., Tamiya, G., Oka, A., Takishima, N., Yamagata, T., Kikkawa, E., Iwata, K., Tomizawa, M., Okuaki, N., Kuwano, Y., Watanabe, K., Fukuzumi, Y., Itakura, S., Sugawara, C.,

HLA and NKG2D Ligand Changes in Malignant Cells

233

Ono, A., Yamazaki, M., Tashiro, H., Ando, A., Ikemura, T., Soeda, E., Kimura, M., Bahram, S., and Inoko, H. (1999). Molecular dynamics of MHC genesis unraveled by sequence analysis of the 1,796,938-bp HLA class I region. Proc. Natl. Acad. Sci. USA 96, 13282–13287. Sim, B. C., and Hui, K. M. (1994). A HLA class I cis-regulatory element whose activity can be modulated by hormones. Int. J. Cancer 59, 646–656. Singer, G., Rebmann, V., Chen, Y. C., Liu, H. T., Ali, S. Z., Reinsberg, J., McMaster, M. T., Pfeiffer, K., Chan, D. W., Wardelmann, E., Grosse-Wilde, H., Cheng, C. C., Kurman, R. J., and Shih, I. M. (2003). HLA-G is a potential tumor marker in malignant ascites. Clin. Cancer Res. 9, 4460–4464. Smoller, B. R., Santucci, M., Wood, G. S., and Whittaker, S. J. (2003). Histopathology and genetics of cutaneous T-cell lymphoma. Hematol. Oncol. Clin. North Am. 17, 1277–1311. Smyrk, T. C., Watson, P., Kaul, K., and Lynch, H. T. (2001). Tumor-infiltrating lymphocytes are a marker for microsatellite instability in colorectal carcinoma. Cancer 91, 2417–2422. Stam, N. J., Spits, H., and Ploegh, H. L. (1986). Monoclonal antibodies raised against denatured HLA-B locus heavy chains permit biochemical characterization of certain HLA-C locus products. J. Immunol. 137, 2299–2306. Stopeck, A. T., Jones, A., Hersh, E. M., Thompson, J. A., Finucane, D. M., Gutheil, J. C., and Gonzalez, R. (2001). Phase II study of direct intralesional gene transfer of allovectin-7, an HLA-B7/beta2-microglobulin DNA-liposome complex, in patients with metastatic melanoma. Clin. Cancer Res. 7, 2285–2291. Stutman, O. (1974). Tumor development after 3-methylcholanthrene in immunologically deficient athymic-nude mice. Science 183, 534–536. Stutman, O. (1979). Chemical carcinogenesis in nude mice: Comparison between nude mice from homozygous and heterozygous matings and effect of age and carcinogen dose. J. Natl. Cancer Inst. 2, 353–358. Sun, Y., Sijts, A. J., Song, M., Janek, K., Nussbaum, A. K., Kral, S., Schirle, M., Stevanovic, S., Paschen, A., Schild, H., Kloetzel, P. M., and Schadendorf, D. (2002). Expression of the proteasome activator PA28 rescues the presentation of a cytotoxic T lymphocyte epitope on melanoma cells. Cancer Res. 62, 2875–2882. Taylor, C. R. (1986). ‘‘Immunomicroscopy: A diagnostic Tool for the Surgical Pathologist. Major Problems in Pathology.’’ W.B. Saunders Company, Philadelphia, PA. Temponi, M., Kekish, U., Hamby, C. V., Nielsen, H., Marboe, C. C., and Ferrone, S. (1993). Characterization of anti-HLA class II monoclonal antibody LGII-612.14 reacting with formalin fixed tissues. J. Immunol. Methods 161, 239–256. Torres, M. J., Ruiz-Cabello, F., Skoudy, A., Berrozpe, G., Jimenez, P., Serrano, A., Real, F. X., and Garrido, F. (1996). Loss of an HLA haplotype in pancreas cancer tissue and its corresponding tumor-derived cell line. Tissue Antigens 47, 372–381. Ugurel, S., Rebmann, V., Ferrone, S., Tilgen, W., Grosse-Wilde, H., and Reinhold, U. (2001). Soluble human leukocyte antigen—G serum level is elevated in melanoma patients and is further increased by interferon-alpha immunotherapy. Cancer 92, 369–376. Vega, F., and Medeiros, L. J. (2003). Chromosomal translocations involved in non-Hodgkin lymphomas. Arch. Pathol. Lab. Med. 127, 1148–1160. Verfaillie, C., Kay, N., Miller, W., and McGlave, P. (1990). Diminished A-LAK cytotoxicity and proliferation accompany disease progression in chronic myelogenous leukemia. Blood 76, 401–408. Vetter, C. S., Groh, V., thor Straten, P., Spies, T., Brocker, E. B., and Becker, J. C. (2002). Expression of stress-induced MHC class I related chain molecules on human melanoma. J. Invest. Dermatol. 118, 600–605. Vitale, M., Pelusi, G., Taroni, B., Gobbi, G., Micheloni, C., Rezzani, R., Donato, F., Wang, X., and Ferrone, S. (2005). HLA class I antigen downregulation in primary ovary carcinoma lesions. Association with disease. Clin. Can. Res. in press.

234

Chien-Chung Chang et al.

Vitale, M., Rezzani, R., Rodella, L., Zauli, G., Grigolato, P., Cadei, M., Hicklin, D. J., and Ferrone, S. (1998). HLA class I antigen and transporter associated with antigen processing (TAP1 and TAP2) down-regulation in high-grade primary breast carcinoma lesions. Cancer Res. 58, 737–742. Vivier, E., Tomasello, E., and Paul, P. (2002). Lymphocyte activation via NKG2D: Towards a new paradigm in immune recognition? Curr. Opin. Immunol. 14, 306–311. Wang, Z., Cao, Y., Albino, A. P., Zeff, R. A., Houghton, A., and Ferrone, S. (1993). Lack of HLA class I antigen expression by melanoma cells SK-MEL-33 caused by a reading frameshift in beta 2-microglobulin messenger RNA. J. Clin. Invest 91, 684–692. Wang, Z., Marincola, F. M., Rivoltini, L., Parmiani, G., and Ferrone, S. (1999). Selective histocompatibility leukocyte antigen (HLA)-A2 loss caused by aberrant pre-mRNA splicing in 624MEL28 melanoma cells. J. Exp. Med. 190, 205–215. Weinman, E., Roche, P., Kasperbauer, J., Cha, S., Sargent, D., Cheville, J., Murphy, L., Chen, L., Wettstein, P., Gostout, B., Ferrone, S., and Strome, S. (2003). Characterization of antigen processing machinery and survivin expression in tonsillar squamous cell carcinoma. Cancer 97, 2203–2211. Wetzler, M., Baer, M. R., Stewart, S. J., Donohue, K., Ford, L., Stewart, C. C., Repasky, E. A., and Ferrone, S. (2001). HLA class I antigen cell surface expression is preserved on acute myeloid leukemia blasts at diagnosis and at relapse. Leukemia 15, 128–133. Wiendl, H., Mitsdoerffer, M., Hofmeister, V., Wischhusen, J., Bornemann, A., Meyermann, R., Weiss, E. H., Melms, A., and Weller, M. (2002). A functional role of HLA-G expression in human gliomas: An alternative strategy of immune escape. J. Immunol. 168, 4772–4780. Williams, A. P., Peh, C. A., Purcell, A. W., McCluskey, J., and Elliott, T. (2002). Optimization of the MHC class I peptide cargo is dependent on tapasin. Immunity 16, 509–520. Wolfel, C., Drexler, I., Van Pel, A., Thres, T., Leister, N., Herr, W., Sutter, G., Huber, C., and Wolfel, T. (2000). Transporter (TAP)- and proteasome-independent presentation of a melanoma-associated tyrosinase epitope. Int. J. Cancer 88, 432–438. Wolfel, T., Klehmann, E., Muller, C., Schutt, K. H., Meyer zum Buschenfelde, K. H., and Knuth, A. (1989). Lysis of human melanoma cells by autologous cytolytic T cell clones. Identification of human histocompatibility leukocyte antigen A2 as a restriction element for three different antigens. J. Exp. Med. 170, 797–810. Wu, J. D., Higgins, L. M., Steinle, A., Cosman, D., Haugk, K., and Plymate, S. R. (2004). Prevalent expression of the immunostimulatory MHC class I chain-related molecule is counteracted by shedding in prostate cancer. J. Clin. Invest 114, 560–568. Yamshchikov, G. V., Mullins, D. W., Chang, C. C., Ogino, T., Thompson, L., Presley, J., Galavotti, H., Aquila, W., Deacon, D., Ross, W., Patterson, J. W., Engelhard, V. H., Ferrone, S., and Slingluff, C. L., Jr. (2005). Sequential immune escape and shifting of T cell responses in a long-term survivor of melanoma. J. Immunol. in press. Yewdell, J. (Guest Ed.) (2002). Generating peptide ligands for MHC class I molecules. Mol. Immunol. 39, 125–259. Zarling, A. L., Luckey, C. J., Marto, J. A., White, F. M., Brame, C. J., Evans, A. M., Lehner, P. J., Cresswell, P., Shabanowitz, J., Hunt, D. F., and Engelhard, V. H. (2003). Tapasin is a facilitator, not an editor, of class I MHC peptide binding. J. Immunol. 171, 5287–5295. Zou, Y., Mirbaha, F., and Stastny, P. (2003). Contact inhibition causes strong down-regulation of expression of MICA and decreased NK cell killing of human fibroblast. The 29th American Society for Histocompatibility and Immunogenetics Annual Meeting Abstracts #256.

Index

2-deoxyglucose uptake, 15, 17, 26, 37 4E-BP1, 37–38 9,10-dimethyl-1,2-benzanthracene (DMBA), 171 12-O-tetradecanoylphorbol-13-acetate (TPA), 171 40S ribosomal subunit, 35

A Acid-soluble pool, 31 Actinomycin, 10 D, 10 Adenocarcinomas, 39 Agar, 41–42, 45, 173 Alkalinization, cellular, 34 Allele, 205 wild-type, 205 Amiloride, 34 Amino acid, 26 AMP cyclic, 41 dibutyryl cyclic, 41 Angiogenesis, 171 and chemokines, 174, 176 Antigens, 161 human leukocyte (HLA), 190–94, 201–5 -anti, 192, 194 class 1, 191, 193, 195–207, 213, 216–23 -DM, 131, 136–38, 142–44 -DO, 131, 138–39, 141, 144 -F, 211 -G -bearing macrophages, 209 soluble (sHLA-G), 209 melanoma-associated MART1, 219 MHC class I, 219 MICA, 217 minor histocompatibility, 146

processing machinery (APM), 194, 201–5, 216, 220–22 transporter associated with (TAP), 199, 203, 207 ULBP, 217 very late (VLA), 163 viral, 145 Ascites, differentiating malignant from benign, 210 ATP, 37–38, 44, 49 -binding site, 205 concentration, 38

B Bacteria, stringent response of to amino acid starvation, 3

C Cadherins, 39–40 Calcium, 29–33 Cancer, human adeno, 198 anogenital, 60 bladder, 143, 204 lesions, 204 bone sarcomas, 172 breast, 64, 169–70, 173, 175, 197–98, 201, 204–5, 208–10, 222 cell lines, 209 lesions, paraffin-embedded, 204 cervical, 60–61, 68, 71–73, 80, 143, 197, 204–5, 207–8 in situ, 64 lesions, 204–5 colorectal, 204 APC mutations in, 46

235

236 Cancer, human (continued) lesions, 204 colon, 64, 169, 197, 201, 204–5, 208–10, 212 esophageal squamous, 202 head and neck squamous cell (HNSCC), 60–61, 65–66, 68, 70–71, 73, 75, 76–77, 80, 197, 204–5, 208, 218 and HPV, 77–78 hepatoma, 197 hypopharynx, squamous cell, 67, 71–72 immune surveillance theory, 215 intestinal, 39 kidney, 210 larynx, 67, 71–72, 78 liver, 200 lung, 171, 174, 197, 204–5, 208, 210, 222 large, 198 Lewis, 170 lymphomas, 172, 214 mammary carcinomas, 172 melanoma, 197, 208 myelomas, 172 nonmelanoma skin, 60 oral cavity, 64, 67, 71–72 oropharyngeal (OSCC), 60–61, 63–64, 70–71 oropharynx, 72 ovarian, 173, 175, 204, 209–10, 217 serous and mucinous-adeno, 198 pancreas, 208–9 prostate, 197–98, 204–5, 207, 209–10, 211–2 renal, 143, 210 stomach, 208 tongue, 64, 67, 78 tonsillar, 59–89 HPV in, 59–89 and correlation to cell-cycle proteins, 70–73 and genetic instability, 73–77 differences of gains and losses of HPV-positive and negative cancer, 76 disease-free 3 years after diagnosis, HPV-positive vs. HPV-negative, 74 squamous cell carcinomas, HPV and CGH data of, 75 and radiosensitivity, 69–70

Index frequency and type of, 62–63 patient features, 63–65 prognosis, 65–69 disease-free patients at 3 years after diagnosis, number & percent of, 66 disease-specific survival in patients, Kaplan-Meier graph, 67 treatment of, 61 vulvar, 76 Carbohydrate metabolism, velocity of the kinase reactions of, 38 Carcinoma adeno, 198 bladder, 204 lesions, 204 breast, 197–98, 201, 204–5, 208–10, 222 cell lines, 208 lesions, paraffin-embedded, 204 cervical, 197, 204–5, 207–8 lesions, 204–5 colon, 197, 201, 204–5, 208–10, 212 colorectal lesions, 204 esophageal squamous cell, 202 head and neck squamous cell (NNSCC), 197 hepatoma, 197 kidney, 210 liver, 197, 199–200 lung, 197, 204–5, 208, 210, 222 large cell, 198 ovarian, 204, 209–10, 217 serous and mucinous-adeno, 198 pancreas, 208–9 prostate, 197–98, 204–5, 207, 209–10, 212 renal cell (RCC), 197–98, 209–10, 217, 221 small cell, 198 squamous cell, 198 stomach, 208 Cation (s), 24, 39, 49 bound, 16–17 stoichiometric changes in, 16–17 cellular, 21 divalent, 17–18, 28, 44–45 binding capacity of, 5 externally bound, 44 ionophore A23187, 27 exchange, 17, 25 cytosol by, 18

Index external, 24 externally bound, 16, 44 intracellular, 18, 28 changes 16 to 17 hours after adding mitogen, 16 monovalent, 34 pumps, 45 surface-bound, 16 Cell (s) 3T3, 30 Balb/c, quiescent confluent, 27, 30, 43 bombesin-stimulated, 23 confluent, inhibition of, 31 ouabain-treated, 29 quiescent, 27 confluent Swiss, 23 Swiss, 27–29 DNA synthesis in, 34 quiescent, 34 acceleration of through the G1 period into the S period, 43 adhesions between, 33 aneuploid, 74 B, 135–36, 138, 140–41, 143–46 bone marrow-derived stem, 161 carcinoma (s), 211 colon, 197, 201, 204–5, 208–10, 212 human, 39 renal (RCC), 197–98, 209–10 small, 198 squamous, 39 contact-inhibited, 33 cortical tissue endothelial, 134 cultures, quiescent, 16 cycling, 7 cytoplasm of, 208 dendritic (DC), 140, 143, 161–62, 168 density-dependent, 33 diploid, 74 division cycle (CDC), 12 kinase, 12 related peptide, 13 embryonic stem (ES), 115 endothelial, 162–63, 165, 168, 175, 194, 198 chemotaxis, 174 human umbilical vein, 174 tumor, 172 vascular, 163, 165 epithelial, 160, 162, 165, 170, 172

237 fibroblast, 160, 168 basic, growth factor (bFGF), 171 in a resting stage, 211 quiescent, 13 stromal, 165 glioma, 211 head and neck squamous cell (HNSCC), 60–61, 65–66, 68, 70–71, 73, 75, 76–77, 80, 197, 204–5, 208, 218 and HPV, 77–78 HeLa, 10 synchronized, 13 hemapoietic normal, 130 stem, 175 hypopharynx, squamous cell, 67, 71–72 immune, 168 in vitro, 220 leukemic, 198–99, 210–11 lines, 217 breast, 208 carcinoma, 208 cultured, 211 HNSCC (head and neck squamous cell cancer), 218 lymphoblastoid, 208 melanoma, 209, 214–15, 217 cultured, 211 with tapasin loss, 205 metastatic, 210 monocytic, 208 tumor, cultured, 212 liver, 199 lymphoblastoid, 208 lymphoid, 161, 194, 198 lymphoma, 23, 198–99 RMA, 215 malignant, 189–234 lesions, 192 NK cell activating ligand expression in, 223 organ-specific metastasis of, 175 resistance of, 220 transformation of, 221 mammalian quiescent cultured, 38 somatic, 44 mast (MCs), 161–62, 168, 170–71 melanoma, 172–73, 198, 201, 211, 216, 218–19 autologous, 214

238 Cell (s) (continued) cytoplasm of, 211 with tapasin loss, 205 membrane, 203, 219–20 motility of, 33 mesenchymal, 162 mesothelial, 170 metastatic, 175 migration, 2, 33 correlated with proliferation, 4 mitogen-stimulated, 48 monocytes, 140 monolayered confluent sheet of normal, 3 mutant derivative, 2 mutual adhesion of, 39 natural killer (NK), 191, 195, 199–200, 215 activating ligand, 215, 220–22 abnormalities, 216 -anti, 192 expression in malignant lesions, 223 ligand, 193–94, 217 abnormalities, 217 activating, 192, 194, 198, 200, 208, 215 mediated cytolysis, 200 immune responses, 217 lysis, 199–200, 215, 217 responses, 211 selective pressure, 215 neoplastic, 3, 39, 168–69, 172–73, 175–76 non, 39 neuroblastoma, cytoplasm of, 211 owl monkey kidney (OMK), 93 professional antigen-presenting, 144 proliferation, 3, 30, 32, 42, 47, 169, 211 control, 45 regulation of, 12, 33 specific and nonspecific stimulators of, 2–5 quiescent, 3, 5, 20, 25–26, 43 confluent, 31 membrane changes, serum-starved, 4 rate of glycolysis in, 33 renal cell (RCC), 197–98, 209–10, 217, 221 RMA, 215 sarcoma, 214 selective pressure, 217 sentinel, 161

Index small, 198 squamous, 198 stained, 194 stromal, 168, 172 T-cells; See T-cells transendothelial, migration (diapedesis), 162–63 transformed, 45–46 density-inhibited, 41 tumor, 198–201, 208–9, 211–12, 214–15, 217–18, 220 prostate, 917 renal cell (RCC), 197, 201 basal, 197 germ, 197 lysis of, 218 CTL-mediated, 218 population, 221 resistant, 216 stained, 194 variants, immunoresistant, 222 vascular, 160, 162, 168, 174 Chemokine, 161–63, 164–68, 170–76 altered regulation of in cancer, 173–74 and angiogenesis, 174 and metastasis, 175–76 fold, 164 -GAG bound, 168 interactions, 165 posttranslationally modified by proteases, 166–67 receptors, 173 Chemotaxis, 175 endothelial cell, 174 Chicken embryo fibroblasts (CEF), 5, 11, 17–18, 27, 30, 33, 42, 44–45 confluent, 43 cultures, 6, 14–15 confluent, serum-free medium on, 16 lowering pH of, 43 hexose uptake in, 5 insulin-stimulated, 17 proliferation, 42 of in serum, 29 quiescent, 30, 34, 48 with serum, stimulation of, 38, 45 Choline, 26, 34 treatment of with insulin, 23

Chromosomal aberrations, average number of (ANCA), 74 instability (CIN), 201 Class II-associated Ii peptide (CLIP), 130, 133–38 Clone (s), nontransformed and spontaneously transformed, 40 transformed for DNA synthesis, 41 Colony stimulating factor-1 (CSF-1), 170 Comparative genomic hybridization (CGH), 73–74 Concentration external, 31 extracellular, 30 intracellular, 30–31 of protons, 32 pyrophosphate, 42 reduction of ATP, 35 serum, 39 surface proton, 33 Confluence, 4 quiescent, 4 Contact inhibition, 2–3, 7–8 Coordinate control, 4 responses, 4 Cortisol, 17, 44 CTL, 218 -mediated lysis, 218 Melan-A-specific, 218 TA-specific, 218, 220 Cyclin (s), 12 and CDKs, specific proteins required for transition from G1 to stages of the cell cycle, 11 and proteins, ubiquitin ligases for, 13 controlled check-points, 13 D1, 13 E, 13 associated protein kinase activity, 13 dependent kinase, 13 dependent kinases (CDKs), 13 Cyclinlike (CYL) expression, 13 gene function during S phase commitment, 13 protein, 13 Cycloheximide, 10, 42–43

Cystatin, 141 Cysteine, 165 proteases, 168 Cytokines, 162–68, 171, 173, 201, 203, 206 angiogenic, 176 anti-inflammatory, 161 chemotactic, 161 pro-inflammatory, 161, 164 Cytology, 210 Cytoplasmic domain, 35 Cytosine monophosphate, 12 Cytosol, 5, 45

D dCMP, 12 Dephosphorylation, 13 Density dependence, 41 multilayered saturation, 41 saturation, 42 Destruction box motif (RXXL), 134, 136 Diapedesis (transendothelial cells, migration), 162–63 Dihydroxycortisone molecule bound, 17 DNA aberration (diploid or aneuploid), 73 amplified, 69 binding factors, 201 contact mutations, 73 content value, 74, 77 damage, 72 double-stranded circular, 62 episomal, 93 gains and losses, 77 genomic viral, 95 H-, 95, 98 host cell chromosomal, 97 human papillomavirus (PV), 62 HVS, 114 HPV, 62 in situ hybridization, 63 incorporation of radioactive thymidine into, 3 L-, 95, 105, 114, 116 HVS, 96, 103 M-, 95 methyltransferases &/or histone deacetylases, oncogenic

240 DNA (continued) oxidative, damage in proliferating cells, 169 recombinant, 195 stability, 169 synthesis, 3–6, 11–13, 17–23, 26, 29–31, 42–44, 48, 100 after growth stimulation, 7 and mitosis, 5, 45, 49 decreases in, 15 effects of inhibitors of protein synthesis on initiation of, 10–11 in Swiss 3T3 cells, 34 increase in, 16 inhibition of protein synthesis by cycloheximide and its effect on the initiation of, 11 initiation of, 31–33 necessity for prolonged stimulation and increased protein synthesis to induce, 5–7 onset of, 27, 34, 38 purified HVS virion, 92 reduction in, 28 requirement of RNA and protein synthesis for the initiation of, 7–10 S period of, 3, 6 transformed clone for, 41 synthetic phase, 10 viral genomic, 98

E EDTA, 25, 44 EGTA, 30, 32 eIF-4E, 35 Electrostatic surface potential, 33 Endoplasmic reticulum (ER), 4, 131–32, 133, 135–36 Endotoxin, 160 Enzymes proteolytic, 5 Eosinophils, 162, 168, 170 Epidermal growth factor (EGF), 8, 23, 32, 35 stimulation of myocytes by, 34 Epithelia, 12 quiescent, confident mammary, 23 Epstein-Barr virus (EBV), 130, 145–46 antigen, 144

Index -induced B cell non-Hodgkin lymphoma, 143, 145 -transformed B cells, 144 Erythroleukemia, 208 Extracellular matrix (ECM), 160, 162, 165, 168 interactions of leukocytes with, 176 proteins, 174 remodeling, 169

F Farnesylation, 47 Floccules, 32, 42 Flocculent precipitates with calcium, 5 Fluorescence in situ hybridization (FISH), 68 Foci, discrete dense of multilayered, 3

G G1, 7–8, 18, 23, 44 deprivation of a growth factor during, 13 period, 5–6, 10, 12, 14, 32, 34, 43, 48 phase, 13, 30 protein synthesis in, 14 stage into the S period of DNA synthesis, 3, 14 stage of the cell cycle, 3 to S, transition from, 11–12 to S periods, 10 G1-S border, 49 boundary, 6 phase, 13 transition, 13 G2, 13 period, 12 Glioblastoma multiforms, 209 Glioma, 209–11, 217 Glucocorticoids, 42 Glucose analogs, 26 increased transport of, 32 measured with radioactive analogs, 3 utilization of in energy metabolism, 32 Glycolysis, 35, 42–43, 45 Glycosaminoglycans (GAGs), 162–63, 165 -bound chemokines, 168 Gonadotropin, 24–25, 44 stimulated oocyte growth, 24

241

Index Granulocyte macrophage-colony stimulating factor (GM-CSF), 162, 171 polymorphonuclear, 161 Growth regulation molecular pathways of, 35 possible roles of Kþ, Ca2þ, pH, Naþ in, 27–35

H Hepatitis virus, 130 Hepatocytes from streptozotocin-induced diabetic rats, 23, 44 malignant, 199–200 Heterozygosity, loss of (LOH), 200, 205, 207 Hexoses, 43, 45 Histamine, 161 HIV (human immunodeficiency virus) entry into leukocytes, 176 -infected patients, 64, 79 Homeostasis, tissue and organ, 160 Hormones, polypeptide, 2, 5, 42 Human leukocyte antigen (HLA), 190–91, 193–94, 201–5, 217, 219 -anti, 192, 194 class 1, 191, 193, 195–207, 213–14, 219, 222 -DM, 131, 136–38, 142–44 -DO, 131, 138–39, 141, 144 Human papillomavirus (HPV), 60–62, 68–69, 72 -16, 61, 65–67, 68, 80 vaccines, anti, 81 and other tumors of the head and neck, 77–78 -associated anogenital malignancies, 64 in tonsillar cancer, 59–81 infections, vaccinations against, 62 load high, 70, 78, 80 low, 78, 81 mutations in, 72 tumors negative, 60, 65, 69–71, 73–76, 77–78, 80 positive, 60, 65, 68–71, 73–74, 76, 77–78, 80 survival for patients with, 76

type 16, 63 oncogenic, 64 vaccines, 78–80 VLPs, 80 Hydrocortisone, 44

I Ii, 131, 133–36, 141, 143–44 Image Cytometry (ICM), 73 Immune response, TA-specific, 222 system impaired cellular-mediated, 64 in vivo, host, 221 involved in tumor surveillance, 142–43 Immunoglobulin (Ig), 164 Immunohistochemistry (IHC), 69, 71–73, 190–91, 192–95, 197–98, 209–10, 221 staining, 197 Immunoproteasome, 219 Inflammation in tumors, acute vs. chronic, 159–88 Insulin, 3, 16–18, 23, 27, 30, 34, 35, 37–38, 43–45, 48 treatment, 4–5 Intracellular adhesion molecule 1/2 (ICAM-1/2), 163 contact, 40 sites, 25 Interferon (IFNs), 168, 215 Ionic sites, 24 Isotherm data, 25

K Kahler’s disease, 143 Kaplan-Meier survival analysis, 69 Kaposi’s sarcoma (KS), 173 herpesvirus (KSHV), 97–98, 102, 111, 115–16 G-protein coupled receptor (KSHV-GPCR), 173 Kinetics, 21 Krebs cycle, 42

L L1 capsid protein, 62–63, 79 Labile protein, 11

242 Lactic acid, 26 production, 15 Lesions bladder, 204 cancer, 204 carcinoma, tumor, 201 breast cancer, paraffin-embedded, 204 cervical carcinoma, 204–5 colorectal carcinoma, 204 malignant, 218–19, 222–23 cell, 192 HLA antigen expression in, 221 NK cell activating ligand expression in, 223 melanoma, 204, 206, 210–11, 214, 219, 221 cutaneous, 206 metastatic, 215 metastatic, 204, 210 primary, 215 tumor, 219 Leukemia, 130, 143, 146, 197–99, 208, 210, 212 B chronic lymphocytic, 209 B-cell, 143–44 pre, 208 chronic lymphocytic, 143 myelogenous (CML), 199 myelocytic, 208 Leukocytes, 160–61, 168, 170–71, 173–76 and inflammation, 161–68 chemokine receptors, 162 infiltration, 169 migration, 172 recruitment, 163, 173 resident, 165 Leupeptin, 141 Lipid mediators, 162 Lipopolysaccharide (LPS), 160 Liver diseases, chronic, 200 Long control region (LCR), 68–69 Lymphocytes B, 161 T, 161–62, 170 TA-specific cytotoxic (CTC), 190 Lymphoma, 197–99, 172, 214 Burkitt, 143, 208

Index EBV-induced B cell non-Hodgkin, 143 follicular, 143 histiocytic, 208 large-cell, 199 non-Hodgkin B and T, 209 T-cell, 208 acute, 92 acute peripheral T-cell, 92–93 pleomorphic peripheral, 93 transplant-related, 199 Lymphoproliferative disorders, 209

M Macromolecular synthesis, 20–21 Macrophage, 161–62, 168, 170, 173 HLA-G-bearing, 209 recruitment, 173 tumor-associated (TAMs), 169–70 Malignant cells, 189–234 conclusion, 220–23 HLA class I antigen and HLA class I-TA derived peptide complex expression by, 218–20 HLA class I antigen, MICA/B, & ULBP expression by malignant cells, nonclassical, 207–12 HLA-E, -F, and -G antigen expression, 208–10 MICA/B and ULBP ligand expression, 210–12 introduction, 190–92 lesions, 212, 215 HLA antigen and NK cell activating ligand expression in, detection of, 192–94 HLA class I antigen phenotypes in, frequency & molecular mechanisms underlying abnormal, 195–207 heavy chain loss and/or downregulation, selective, 205–7 inherited mutations in one copy of the TAP1 or tapasin gene, 206 loss and marked downregulation, total, 195–205 abnormalities in 2m expression, 200–1

Index gene regulation and/or APM component expression, defects in, 201–5 TAP1 and tapasin downregulation in malignant lesions of different embryological origin, frequency of, 197 phenotypes identified in, molecular mechanisms underlying abnormal, 196 role of immune selective pressure in the generation of, 212–17 in shaping a tumor cell population, 216 Matrix metalloproteinase-9 (MMP-9), 168, 171, 174–75 Melanoma, 130, 143, 146, 169, 198, 209–11 -associated antigens MART1, 219 cell line, 209, 214, 215, 217 cultured, 211 with tapasin loss, 205 cells, 211, 216, 218, 219 autologous, 214 cytoplasm of, 211 negative, 214 cutaneous, 202, 204 lesions, 206 lesions, 204, 206, 210–11, 214, 221 metastatic, 215 tumors, metastatic, 215 Membrane (s) binding capacity of, 39 change, conformational, 17 components, phosphorylation of, 47 Magnesium, Mitosis (MMM), 45 plasma, 47 Metabolism carbohydrate, kinases of, 49 cellular, 45 Metalloproteinases (MPs), 165 Metastatic cell lines, 210 lesions, 204, 210 Metastasis and chemokines, 175–76 cell, 175 cancer, 175 neoplastic, 175 Methylcholanthrene, 39 treated epidermis, 39

243 Mg2þ and Kþ concentration dependencies of globin synthesis in vitro, 9 availability of, major features of a model for coordinate control of intermediary metabolism and growth by, 46 concentrations, protein and DNA synthesis in very low Ca2þ with variations in, 18–21 content and rates of protein synthesis in Balb/c 3T3, parallel decreases with time in, 22 cytosolic free, mitogen-induced increases in, 23–25 dependence of DNA synthesis and cation content in CEF, 15 deprivation of, normalization of transformed cells by, 40 effects on diverse cellular responses to growth factors, 26–27 in cell growth regulation and transformation, 1–57 conclusions, 42–49 intracellular, 16 rates of protein and DNA synthesis in Balb/c cells as a function of, 20 limitation in physiological CA2þ, kinetics of cellular responsiveness to, 21–22 role of in growth regulation, 14–18 Mg, total cellular, 23 MHC, 207 class I, 130–31, 144, 146 A, 210 antigens, 217 B, 200, 207–12 -related chain (MIC), 191, 208, 210 soluble (sMIC), 212 blocks, 217 specific, mAB, 210 /ULBP, 208 MHC class II containing compartments (MIIC), 131, 134–36, 138–42 MHC class II molecules in cancer, 129–58 antigen presentation, multiple steps in, 130–40 a brief introduction to the process of, 130–31 from the MIIC to the plasma membrane, 139–40

244 MHC class II molecules in cancer (continued) HLA-DM, the editor for antigenic peptide loading of MHC class II molecules, 136–38 the structure of, 137 HLA-DO, the chaperone of chaperones, 138–39 invariant chain, more than just an MHC class II chaperone, 135–36 transporting pseudopeptide, the invariant chain, 131–34 peptide loading of MHC class II and MHC class I molecules, 132 structural overview of the p31 isoform of the human invariant chain, 134 concluding remarks, 146 in oncogenesis, 142–46 expression and tumor development, 143–45 immune system involved in tumor surveillance, 142–43 toward MHC class II-restricted tumor immunotherapy, 145–46 interfering with antigen presentation, 140–42 promoting antigen presentation, 140–41 introduction, 129–30 Microsatellite instability (MSI), 201 Microtubule organizing center (MTOC), 139 Microvilli, 4 Mitochondrial inhibitor, 38 Mitogen, 38, 45, 48, 110 Mitogenic response, 28–29 stimulation of nontransformed animal cells, nonspecific, 4 Mitogenesis, 34 Mitosis, 5, 8, 45, 49 Molecules, immunosuppressive, 207 Monocytes, 161–62 Monoclonal antibodies (mAb), 190–95, 198, 204, 206, 210, 219 APM component-specific, 203 MICA-specific, 210 mRNA, 14, 105, 192, 200, 203, 214 levels, 13 protein synthesis on, 35 mTOR, 37, 47 kinase, 35, 49 phosphorylates, 35

Index phosphorylation, 38, 49 of and the increase in the initiation of protein synthesis, 37 Mucosa, adjacent normal, 39 Mucosal addressin cell adhesion molecule 1 (MAdCAM-1), 163 Multiplication stimulating activity (MSA), 6 Mycobacterium Leprae, 142 Tuberculosis, 142 Myocytes, 32, 34 Myristic acid, 47 Myristylation, 47

N Negative cisregulatory element (NRE), 201 Neoplastic cells, 3, 39, 168–69, 172 non, 39 development, 174 environment, 172 processes, 170 progression, 171–73 tissue, 172 transformation, role of cations in, 38–42 Neuroblastoma, 210, 212 cells, cytoplasm of, 211 Neutrophils, 162, 168, 170

O Oligomers 35 Oncogenes, 40 transfection of, 47 Oncogenic P3K retroviruses code, 47 potential, 60 Oocytes, 25 of Xenopus laevis, 24 quiescent, 24 Orthophosphate, inorganic, 32 Osmotic substitute, 34 Ouabain, 27–28 treated cultures, 28

P Papillomavirus bovine (BPV), 79 canine oral (COPV), 79

Index cottontail rabbit (CRPV), 79 human, 130 Pathogens, 141–42 PDGF, 29, 34, 35 Peptide chain, elongation of, 10 growth factors, 28 pH, 6, 43 and sodium, 33–35 cytosolic, 34 intracellular, 33 Phenotype, transformed, 42, 47 PI 3-K signaling pathway, simplified model of the activation of the, 36 Phosphate, inorganic, 42 Phosphofructo-kinase, 33, 45 Phosphoglucomutase, 43, 48 Phospholipids, 14 Phosphoryl transfer reaction, 38 Phosphorylation, 20, 26, 31, 35, 37 detectable, 38 oxidative, 43 upstream, of the PI 3-K pathway, 38 Plant lectins, 48 Plasma, 30 membrane, 14, 32 Pleiotypic response, 3, 5 Polymerase chain reaction (PCR), 63, 66–69, 78 -SSCP, 72 Polyoma middle T (PyMT), 170 Polypeptide (s) chain, 43 growth factors, 35 purified, 29 Polysomal structures, 8 Polysome formation, 8, 14 in vitro, bacterial, 10 free cytoplasmic, 8 Potassium, 27–29 Proangiogenic factors, 171 Proliferative capacity, 2 Prostaglandins, 173 Proteases, 162, 165, 166–67 extracellular, 170–71 Protein (s), 2, 14 cell surface, mediation between, 39 cyclinlike (CYL), 13 ECM, 174

245 Gag, 47 gammaherpesvirus signaling, structural diversity of the major, 106 kinases, 12, 35, 47, 49 of the PI 3-K pathway, 38 latent membrane (LMP), 108, 203 2 and 7, 204 immunoproteasome subunits, 203 leupeptin-induced (LIP), 134 small (SLIP), 134 macrophage inflammatory (MIP), 113 monocyte chemotactic (MCP), 169 ribosomal, 49 S6, 35 saimiri transformation-associated (STP), 105–6 of sub-group C strains (stpC), 103–5, 109, 111, 114 T-cell transformation and oncogenesis, 105–6, 116 src, 47 synthesis, 7–8, 10–12, 14–15, 18–22, 24–29, 32–34, 38, 42–45, 47–49 in vitro, 19, 25, 31, 43 in vivo regulation of, 48 gonadotropin-induced increase in, 25 hormone-induced increase in, 25 on mammalian polysomes in vitro, 9 regulation of by the PI 3-K and mTOR pathways, 35–38 response of to polypeptide growth factors, 35 S6, 35 tumor-specific, 145–46 tyrosine kinase interacting (tip), 103–5, 108–9, 111, 114, 116 UL16-binding (ULBP), 191, 207–12 MIC, 208 Proteinases, extracellular, 176 Proteoglycans, 162 Proteolysis, regulated intramembrane (RIP), 136 Proteolytic activity, 46 Protooncogenes, 47 Pyrophosphate concentration, 42 inorganic, 14–15, 32, 42

246

Q Quiescence, 28

R Reactive oxygen species (ROS), 168 Remission, complete (CR), 69 without, (non-CR), 69 Retroviruses, transformation by, 47 Reverse transcription-polymerase chain reaction (RT-PCR), 192, 203–4, 209 Ribonucleotide reductase, 12 Ribosomes, 8, 32, 43 of transformed cells, 39 Ribosomal monomers, 8, 14 proteins, synthesis of, 35, 37 structure, 7 subunit, 10 RNA, 3, 7–8, 32, 43, 164–65 export, nuclear, 97 in situ hybridization, 63 synthesis, 10, 15, 26–27 unspliced, 97 Rous sarcoma virus (RSV), 29 src gene product of, 47

S S period, 3, 11–12, 29, 43–44 phase, 6–7, 10, 28, 30, 43 S6 kinase, 35, 37–38 protein, 35 Salmonella Typhimurium, 142 Sarcomas, 215 bone, 172 chemically induced, 214 Kaposi’s (KS), 173 Rous virus (RSV), 29 Saturation density, 2–3, 40–41 Selectins, 163 Serine, 165 phosphorylating, 49 proteases, 168 /threonine kinases, 35 Serum, 2–3, 6, 12, 13, 17–18, 21, 26–27, 29–32, 34, 41, 43, 45 animal, 42 concentration of, 33, 39, 41

Index containing medium, 30 free medium, 28–29 growth factor, 28, 31 heterologous (calf), 30 molecules, 39 proliferation of CEF in, 29 rate, 7 stimulation, 21, 27–28 Signal transducer and activator of transcription (STAT), 105, 108, 113 Sodium pyrophosphate, 5 Staining IHC, 197 intensity, 194 Stoichiometric exchange, 44 Strontium, 32 Sulfate chondroitin, 165 dermatin, 165 heperan, 165 Superantigens (Sag), 97, 102 murine, 97 Substratum, 46

T Tapasin, 219 downregulation, 197 gene, 206 loss, melanoma cells with, 205 T-cell, 130–31, 140 -based immunotherapy, 190, 201, 212–13, 217, 221, 223 cytotoxic, 130, 145 mediated immunity, leukemic and lymphoma cells-specific, 199 selective pressure, 213 recognition, 213 responses, 144 selective pressure and outgrowth of tumor cells with HLA class I antigen loss, 214 T-cell transformation and oncogenesis, 91–128 conclusions, 115–17 gene therapy, rhadinovirus vectors for, 114–15 rhesus monkey rhadinovirus (RRV), 115

Index genome structure, 95–96 growth transformation of human T-cells by rhadinoviruses, 110–14 application of T-cell growth transformation, 113–14 establishing transformed T-cell lines, 110–11 phenotype of rhadinovirus transformants, 111–13 oncogenesis, 103–10 gammaherpesvirus signaling proteins, structural diversity of the major, 106 Herpesvirus saimiri & Herpesvirus ateles L-DNA left terminal regions, comparison of, 104 rhadinoviruses, transformation-related proteins of, 103–5 STP, the transformation-associated protein, 105–6, 116 T-lymphotropic rhadinoviruses, replication and persistence of, 96–98 immediate-early (IE) and early genes, 96–97 latency-associated nuclear antigen, 97–98 viral homologs to cellular genes, 98–103 antiapoptotic viral proteins, 101 complement regulatory proteins, 100 complement control protein homologue (CCPH), 100 homologs to enzymes of the nucleotide metabolism and growth regulation, 100 lytic gene expression cascade, the LANA protein as an upstream regulator of, 99 viral cytokines and receptor proteins, 101–2 G-protein coupled receptors (GPCR), 101 viral superantigen homolog and a new signaling adaptor protein, 102–3 major histocompatibility complex (MHC) class II molecules on accessory cells, 102 restricted antigen-specificity, 116 restriction elements, 110

247 mouse mammary tumor virus (MMTV), 102 viruses and pathology, natural occurrence of the, 92–94 herpesvirus ateles (HVA), 92–93, 97, 100–1, 103–4, 109, 116 in spider monkeys (Ateles spp., A.), 92 herpesvirus saimiri (HVS), 92–93, 95–98, 101, 103–5, 108–14, 116 genomes, episomal, 93 in common marmosets (Callithrix jacchus, C.), 92–93 in cottontop tamarins (S. oedipus), 93, 102 in cynomolgus monkeys (M. fascicularis), 93 in macaques, 93, 113–14, 117 in owl monkeys (Aorus trivirgatus), 92 in rhesus monkeys (Macaca mulatta), 93, 113 in squirrel monkeys (Saimiri sciureus), 92, 109 in tamarins (Saguinus spp., S.), 92–93 transformed human, 102 KS-associated herpesvirus (KSHV), 97–98, 102, 111, 115–16 rhadinovirus, 92, 95, 98 genome structure of selected, 94 Threonines phosphorylating, 49 /serine kinases, 35 Thrombin, 5 Thymidine, 17 kinase, 26 uptake of, 26 Thyroglobulin domain (TGD), 134–135 Titrations, null point, 23 Tonsil, palatine, 60 Transforming growth factor beta (TGf- ), 171 Transmembrane domain (TM), 134 Transplant patients, 64, 70, 142 ‘‘Trigger cell division,’’ 34, 49 Tumor basal cell, 197 bladder, 197 carcinoma, 197 bladder, lesions, 201 breast, 197, 201 drug-resistant, 201 cervical, 197

248 Tumor (continued) colon, 197 leukemia, 197 lung, 197 lymphoma, 197 melanoma, 197 prostate, 917 renal cell (RCC), 197, 201 germ cell, 197 head and neck squamous cell carcinoma (HNSCC), 197, 201 hepatoma, 197 human papillomavirus (HPV) negative, 60, 65, 69–71, 73–76, 77–78, 80 positive, 60, 65, 68–71, 73–74, 76, 77–78, 80 survival for patients with, 76 lesions, 219 melanoma, metastatic, 215 necrosis factor-alpha (TNF-), 164 ovarian, 197 pancreas, 197 stomach, 197 Tumor development, soluble mediators of inflammation during, 159–88 concluding remarks, 176 inflammation and tumor progression, 168–76 TNF- and cancer development, 171–72 chemokines altered regulation of in cancer, 173–74 and metastasis, 175–76 promote neoplastic progression, 172–73 introduction, 160–61 leukocytes and inflammation, 161–68 adherence and immobilization at sites of tissue damage, 162–63

Index cytokines, chemokines, and leukocyte recruitment, 164–68 neoplastic progression associated with inflammation and angiogenesis, 160 Two-on-one (tio) gene, 103–5, 109 Tyrosine autophosphorylate on, 35 kinase, 23 residues, 47 Trypsin, 5, 42 Trypsinization, 6

U Uridine, 3, 17, 26 kinase, 26, 31 phosphorylation of in cells, 27, 31–32 uptake, 20–21, 27, 31–32

V Vascular cell adhesion molecule-1 (VCAM-1), 162–63 endothelial growth factor (VEGF), 169–71 Vasopressin, 34 Virus replication, 100 tumor-causing, 130 Viruslike particles (VLPs), 79 HPV, 80

W Waldeyer’s ring, 60

X Xenopus laevis, 24, 29, 32, 44 oocytes of, 24