Lentiviral Vector Systems for Gene Transfer

MEDICAL INTELLIGENCE UNIT 31

Gary L. Buchschacher, Jr.

Lentiviral Vector Systems for Gene Transfer

MEDICAL INTELLIGENCE UNIT 31

Lentiviral Vector Systems for Gene Transfer Gary L. Buchschacher, Jr., M.D., Ph.D. Division of Hematology/Oncology Department of Medicine University of California-San Diego La Jolla, California, U.S.A.

EUREKAH.COM GEORGETOWN, TEXAS U.S.A.

LENTIVIRAL VECTOR SYSTEMS FOR GENE TRANSFER Medical Intelligence Unit 31 Eurekah.com Landes Bioscience Designed by Jesse Kelly-Landes Georgetown, Texas, U.S.A. Copyright ©2003 Eurekah.com All rights reserved. No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher. Printed in the U.S.A. Please address all inquiries to the Publishers: Eurekah.com / Landes Bioscience, 810 South Church Street, Georgetown, Texas, U.S.A. 78626 Phone: 512/ 863 7762; FAX: 512/ 863 0081 ISBN: 1-58706-094-9 While the authors, editors and publisher believe that drug selection and dosage and the specifications and usage of equipment and devices, as set forth in this book, are in accord with current recommendations and practice at the time of publication, they make no warranty, expressed or implied, with respect to material described in this book. In view of the ongoing research, equipment development, changes in governmental regulations and the rapid accumulation of information relating to the biomedical sciences, the reader is urged to carefully review and evaluate the information provided herein.

Library of Congress Cataloging-in-Publication Data Lentiviral vector systems for gene transfer / [edited by] Gary L. Buchschacher, Jr. p. ; cm. -- (Medical intelligence unit ; 31) Includes bibliographical references and index. ISBN 1-58706-094-9 (hardcover) -- ISBN 1-58076-095-7 (softcover) 1. Genetic vectors. 2. Lentiviruses. 3. Genetic transformation. [DNLM: 1. Gene Transfer Techniques. 2. Gene Therapy. 3. Genetic Vectors. 4. Lentivirus. QZ 50 L574 2001] I. Buchschacher, Gary L. II. Series. QH442.2 .L46 2001 571.9'648--dc21

in memoria di mia nonna Elisabetta Michelotti Mattaini n. 11 aprile 1902 Fiano, Italia m. 7 agosto 2000 Bessemer, Michigan e mio nonno Pietro Luigi Mattaini n. 23 luglio 1897 Sesona di Vergiate, Italia m. 17 giugno 1976 Bessemer, Michigan

—GB

CONTENTS Preface ................................................................................................. vii 1. Introduction to Retroviruses and Retroviral Vectors .............................. 1 Gary L. Buchschacher, Jr. Abstract ....................................................................................................... 1 Introduction ................................................................................................ 1 General Background and Classification of Retroviruses ................................ 1 The Retrovirus Genome and Virion Structure ............................................. 2 The Retrovirus Replication Cycle ................................................................ 3 Elements of Retroviral Vector Systems ......................................................... 6 Summary ..................................................................................................... 9 Acknowledgements .................................................................................... 10

2. HIV-1 Replication................................................................................ 13 Eric O. Freed Introduction .............................................................................................. 13 The HIV-1 Replication Cycle .................................................................... 14 Role of the HIV Accessory Proteins in Virus Replication ........................... 28 Concluding Remarks ................................................................................. 30

3. Determinants for Lentiviral Infection of Non-Dividing Cells ............... 35 Marie A. Vodicka Abstract ..................................................................................................... 35 Introduction .............................................................................................. 35 Nuclear Transport ..................................................................................... 36 Transit of Non-Retroviral Particles ............................................................ 36 Retroviral Preintegration Complexes .......................................................... 39 Assays for Infection of Non-Dividing Cells ................................................ 40 Viral Protein Determinants for HIV-1 Infection of Non-Dividing Cells ........................................................................... 40 Viral Genome Structural Determinants ..................................................... 43 Cellular Transport Pathways and the Cytoskeleton .................................... 45 Summary and Conclusions ........................................................................ 45

4. HIV-1 Vector Systems .......................................................................... 51 Narasimhachar Srinivasakumar Abstract ..................................................................................................... 51 Introduction .............................................................................................. 51 HIV-1 Vector System Development .......................................................... 53 Clinical Applications of HIV-1 Vectors ..................................................... 68 Anticipated Developments in HIV-1 Based Packaging Systems and Possible Confounding Factors ......................................................... 69 Conclusions ............................................................................................... 72 Acknowledgements .................................................................................... 72

5. HIV-2 and SIV Vector Systems ............................................................ 83 James R. Gilbert and Flossie Wong-Staal Abstract ..................................................................................................... 83

Introduction .............................................................................................. 83 Classification and Distribution .................................................................. 84 Pathology and Viral Replication ................................................................ 85 Genome Organization and Regulation ....................................................... 87 Vector Systems .......................................................................................... 91 Requiem and Prospectus ............................................................................ 94

6. FIV Vector Systems .............................................................................. 99 Sybille L. Sauter and Mehdi Gasmi Abstract ..................................................................................................... 99 Epidemiology and Pathogenesis of FIV Infection ....................................... 99 Development of FIV-Based Vectors ......................................................... 107 Acknowledgements .................................................................................. 118

7. EIAV, CAEV and Other Lentivirus Vector Systems ........................... 131 John C. Olsen Abstract ................................................................................................... 131 Introduction ............................................................................................ 131 Biological Similarities and Differences of Lentiviruses .............................. 131 Recent Vector Developments ................................................................... 134

8. Safety Considerations in Vector Development ................................... 147 John C. Kappes and Xiaoyun Wu Abstract ................................................................................................... 147 Introduction ............................................................................................ 147 Current Safety Design ............................................................................. 149 Advancing Lentiviral Vectors Toward the Clinic ..................................... 154

9. Prospects for Gene Therapy Using HIV-Based Vectors ...................... 159 Jiing-Kuan Yee and John A. Zaia Abstract ................................................................................................... 159 Advantages of HIV Vectors Support Development .................................. 159 Attempts to Engineer Biosafety into HIV Vectors.................................... 160 Potential Clinical Applications of HIV Vectors........................................ 163 Regulatory Issues ..................................................................................... 166 Acknowledgements .................................................................................. 168

10. Ethical Considerations in the Use of Lentiviral Vectors for Genetic Transfer ........................................................................... 175 Ina Roy Abstract ................................................................................................... 175 Introduction ............................................................................................ 175 Gene Therapy: Potentially Problematic Types of Application .................. 176 General Ethical Considerations: Research and Clinical Settings ............... 182 Summary ................................................................................................. 190

Index .................................................................................................. 193

EDITOR Gary L. Buchschacher, Jr., M.D., Ph.D. Division of Hematology/Oncology Department of Medicine University of California-San Diego La Jolla, California, U.S.A. Chapter 1

CONTRIBUTORS Eric O. Freed Laboratory of Molecular Microbiology NIAID, NIH Bethesda, Maryland, U.S.A Chapter 2 Mehdi Gasmi University of California-San Diego La Jolla, California, U.S.A. Chapter 6 James R. Gilbert University of California-San Diego La Jolla, California, U.S.A. Chapter 5 John C. Kappes Department of Medicine University of Alabama at Birmingham Birmingham, Alabama, U.S.A. Chapter 8 John C. Olsen Cystic Fibrosis/Pulmonary Research and Treatment Center University of North Carolina Chapel Hill, North Carolina, U.S.A. Chapter 7 Ina Roy Center for Bioethics University of Pennsylvania Philadelphia, Pennsylvania, U.S.A. Chapter 10 Sybille L. Sauter GenStar Therapeutics San Diego, California, U.S.A. Chapter 6

Narasimhachar Srinivasakumar Division of Hematology/Oncology Department of Medicine Vanderbilt University Nashville, Tennessee, U.S.A. Chapter 4 Marie A. Vodicka Human Biology Fred Hutchinson Cancer Research Center Seattle, Washington, U.S.A. Chapter 3 Flossie Wong-Staal University of California-San Diego La Jolla, California, U.S.A. Chapter 5 Xiaoyun Wu Department of Medicine University of Alabama at Birmingham Birmingham, Alabama, U.S.A. Chapter 8 Jiing-Kuan Yee Department of Virology Beckman Research Institute City of Hope Duarte, California, U.S.A. Chapter 9 John A. Zaia Department of Virology Beckman Research Institute City of Hope Duarte, California, U.S.A. Chapter 9

PREFACE

T

his volume is designed to summarize recent progress in the development of lentiviral vectors for in vitro gene transfer studies and for subsequent animal and human clinical studies. Although lentiviral vector systems may offer some advantage over previously studied viral vector systems for use in potential gene therapy applications, such vectors also present technological hurdles that must be overcome and raise unique concerns that must be addressed as the vector systems are developed and potentially applied to clinical gene transfer studies. Applied gene transfer studies using viral vector systems are being undertaken by more and more people. For this reason, the first chapter presented here is designed to assist those interested primarily in the application of lentiviral vectors for gene transfer, but who may lack a background in basic retrovirology, by presenting some general information about retroviral replication and retroviral vector systems that will provide a basis for the topics that are discussed in detail in the remainder of the book. The final section of this book, discussing ethical considerations relevant to lentiviral vector systems, is important for those studying either vector development or potential clinical gene transfer applications; it is recommended that this chapter be read first, so that topics discussed in other chapters can be thought of with these ethical considerations in mind. Because the various lentiviruses share many features of replication and because human immunodeficiency virus type 1 (HIV-1) is the most thoroughly studied of the lentiviruses, one chapter is devoted to a discussion of HIV-1 replication. The following chapter discusses in detail the feature that makes lentiviral vectors attractive as gene transfer vehicles—the ability of lentiviruses to efficiently infect non-dividing cells. Subsequent chapters discuss progress in development of vector systems based on a variety of lentiviruses—HIV-1, HIV-2, SIV, FIV, EIAV and so on. Potential approaches to and issues regarding some clinical applications of the vector systems are then presented, along with discussions dealing with various safety issues that are raised when developing and considering these systems for clinical use. It is hoped that the contents of the book will leave readers, either those involved in the study of basic virology or vector system development or those involved in gene transfer applications, with an understanding of both the current status of lentiviral vector technology and the issues which must be addressed as these systems are further developed and used. Clearly, regardless of how the various vector systems eventually might or might not be used for clinical applications, the value of study of the vector systems is evident by the amount of information on the basic science of viral replication, cell biology and gene transfer that continues to be obtained. Gary L. Buchschacher, Jr., M.D., Ph.D.

CHAPTER 1

Introduction to Retroviruses and Retroviral Vectors Gary L. Buchschacher, Jr.

Abstract

A

s various viral vector systems for gene transfer are developed, interest in using such systems in applied settings continues to grow. This Chapter is designed to provide background information for readers interested in learning about lentiviral vector systems for gene transfer applications but who lack a background in retrovirology. To assist those readers who are unfamiliar with retroviral vector systems, basic outlines of the retroviral replication cycle and of characteristics of retroviral vector systems are introduced here in order to present and define concepts and terms that are discussed in subsequent Chapters.

Introduction The development of vector systems derived from the lentivirus genus of retroviruses has potential for possible clinical gene transfer applications.1 Retroviral vector systems used previously in gene transfer applications have been derived from the oncogenic retrovirus genus. The oncoretroviruses, sometimes (though not necessarily correctly) referred to as “simpler” retroviruses in comparison to lentiviruses, had been studied for years and knowledge gained from those studies enabled development of vector systems based on the parent viruses. The concepts governing development of oncoretrovirus-derived vector systems also guide the development of lentiviral vector systems. Therefore, an understanding of the retroviral replication cycle and of the basic features of retroviral vector systems based on oncogenic retroviruses is necessary in order to fully understand the complexities and issues associated with the development of lentiviral vector systems discussed later in this book. In this first Chapter, the retroviral replication cycle and concepts related to development of retroviral vector systems are outlined. Since a comprehensive review of the various retroviruses and vectors derived from them is obviously beyond the scope of this section, the information presented will describe generic features shared by retroviruses and vector systems in general, rather than focus on one or more particular viruses. Readers who are familiar with these concepts are advised to skip this introductory Chapter and to move on to Chapter 2.

General Background and Classification of Retroviruses Retroviruses are single-stranded RNA viruses that replicate through a double-stranded DNA intermediate.2 Various retroviruses have been found that infect a number of organisms, including humans and many other mammals. The earliest retroviruses studied were isolated from mice and birds. Examples of such retroviruses include murine leukemia virus (MLV) and Lentiviral Vector Systems for Gene Transfer, edited by Gary L. Buchschacher, Jr. ©2003 Eurekah.com.

2

Lentiviral Vector Systems for Gene Transfer

mouse mammary tumor virus (MMTV), and the avian pathogens Rous sarcoma virus (RSV), spleen necrosis virus (SNV), and avian leukosis virus (ALV). These viruses were discovered and were of interest because of their association with the development of tumors in their host organisms. Study of these viruses eventually led to the discovery and development of the oncogene theory of tumorigenesis: some of the viruses actually contained oncogenes within their genomes, while others interacted with oncogenes in either a direct or indirect way to contribute to tumor formation.3,4 Over time, other retroviruses that had in common with the previously isolated viruses many features of their genome organization and overall replication strategy were discovered. Historically, because of their pattern of pathogenicity, these viruses were grouped into three subfamilies: 1) the acutely oncogenic retroviruses, or oncoretroviruses (such as those described above); 2) the lentiviruses (associated with “slow” diseases or those with long latent periods); 3) the spumaviruses (“foamy” viruses, named because of the pathogenic changes observed in infected cells). Viruses were also grouped (Types A-D) according to the electron microscopic appearance of their nucleocapsid structures.5 Further study of these viruses enabled detailed comparison of their genome structures and nucleic acid sequences, which resulted in a further refinement of the retrovirus classification system.6 This led to a revised classification of the Retroviridae family into seven genera: mammalian type B retroviruses, mammalian type C retroviruses, avian type C retroviruses, mammalian type D retroviruses, HTLV/BLV type retroviruses, lentiviruses, and spumaviruses. The remainder of this Chapter will focus on describing general features of the retrovirus replication cycle and retroviral vector systems, focusing on “simple” oncoretroviruses. Although vector systems derived from MLV are the most commonly used retroviral-based systems in gene transfer applications, this discussion will not focus on MLV per se, but will describe prototypical features characteristic of the retroviruses in general.

The Retrovirus Genome and Virion Structure Genome Structure The retrovirus genome contained within viral particles consists of two identical singlestranded RNA molecules (for this reason, retroviruses are referred to as being “pseudodiploid”) of positive polarity that are replicated through a double-stranded DNA intermediate (reviewed in refs. 7, 8). The organization of the RNA and DNA forms of the genome are shown in Figure 1. The 5’ end of the genomic RNA begins with the “r” (for repeat) and “u5” (for unique 5’ region) segments, followed by the viral genes gag, pol, and env. The 3’ end of the genomic RNA terminates with the u3 (for unique 3’ region) and r (identical to the 5’ r region) regions and a polyA tail. Following reverse transcription of the RNA genome into a double-stranded DNA molecule, the DNA form of the viral genome is integrated into the host cell chromosomal DNA, where it is thereafter referred to as a “provirus.” Because of the mechanism used for reading and utilizing the viral RNA template during reverse transcription,8-10 the u3 and u5 regions are duplicated such that the 5’ and 3’ ends of the proviral genome differ in structure from the ends of the RNA genome (Fig. 1). Each end of the proviral genome is made up of regions called long terminal repeats (LTRs) which contain the proviral U3, R, and U5 regions. This rearrangement of both termini of the viral genome enables appropriate expression of the viral genes. The U3 region of the 5’ LTR (copied from the u3 region at the 3’ end of the RNA genome) contains the viral promoter and enhancers responsible for initiation of transcription of the viral genome at the 5’ U3/R junction. The viral gag and pol genes are expressed from an unspliced transcript while the env gene is expressed from a spliced transcript (the splice donor is located between the 3’ end of the 5’ LTR and the 5’ end of the gag gene, with the splice acceptor located at the 5’ end of the env gene). The 3’ end of the genome contains the transcription termination signal, with the

Introduction to Retroviruses and Retroviral Vectors

3

Fig. 1. Genome structure of a prototypical retrovirus. The genomic viral RNA, represented by a single black line, is shown at the top of the figure, with the structure of the resulting provirus after reverse transcription below. The locations of the open reading frames gag, pol, and env are shown. Reverse transcription of the RNA results in rearrangement of the termini of the genome, resulting in the structures of the LTRs (long terminal repeats) as indicated. Cis-acting sequences of the viral genome are shown in more detail in Figure 3.

polyadenylation signal located in the 3’ LTR. The gag gene encodes viral core structural proteins: matrix (MA), capsid (CA), and nucleocapsid (NC). Some retroviruses also encode various other small proteins or peptides within the gag open reading frame. The pol gene encodes the viral replication enzymes: protease (PR), reverse transcriptase (RT), and integrase (IN). The env gene encodes the envelope glycoprotein (Env) which is processed into transmembrane (TM) and surface (SU) subunits.

Virion Structure Retrovirus particles consist of the viral protein core and the surrounding viral envelope that is made up of a cellular membrane-derived lipid bilayer and viral-encoded envelope glycoproteins. The structural core of the virion contains the two RNA molecules (associated as a dimer) in association with the nucleocapsid protein that, in turn, is surrounded by a shell of capsid protein. The matrix protein is located outside of the viral capsid and appears to interface with the inner part of the viral envelope that surrounds the core. The viral-encoded replication enzymes, including reverse transcriptase and integrase, are located within the viral core.

The Retrovirus Replication Cycle The retroviral replication cycle (Fig. 2) commences when a virion begins to infect a cell via the interaction of the envelope surface glycoprotein with a specific cellular receptor (or receptors). A subsequent conformational change in the viral envelope surface glycoprotein results in exposure of a fusion domain contained within the envelope transmembrane glycoprotein. Subsequent fusion of the viral envelope with the cellular membrane occurs, which results in release of the virus core into the cytoplasm of the cell.7 In most retroviruses, this membrane fusion occurs directly at the cell surface, but in some retroviruses the virion, after binding to the cellular receptor, is internalized via endocytosis; fusion of the viral envelope with the endosome then results in release of the viral core into the cytoplasm. Exactly how the next series of steps occur in relation to each other is not completely clear. However, it is known that the viral genome is uncoated, reverse transcribed into double-stranded DNA, and integrated into the genome of the host cell where it resides permanently as the provirus. The provirus is then passed on to daughter cells following cellular division, just like a cellular gene would be. A brief description of the general steps of retroviral reverse transcription will be given here; the steps of reverse transcription of the HIV-1 genome are discussed in detail in Chapter 2 (refs. 7,8). Viral reverse transcriptase, in order to initiate minus strand DNA synthesis using the RNA genome as a template, utilizes as a primer a cellular tRNA molecule that is bound to the primer binding site (pbs) located just downstream of the 5’ LTR. The identical r regions of the

4

Lentiviral Vector Systems for Gene Transfer

Fig. 2. The retrovirus replication cycle. Following recognition of a specific cellular receptor by the viral envelope glycoprotein and adsorption of the virion to the cell surface, fusion of viral and cellular membranes results in release of the viral core into the cell cytoplasm (usually at the cell surface, but for some retroviruses this occurs following endocytosis of virions). The viral RNA is uncoated and reverse transcribed into a double-stranded DNA copy. Following breakdown of the nuclear envelope, the DNA copy is integrated into a chromosome where it resides as the provirus. Transcription of the provirus results in production of copies of viral genomic RNA, which can either be translated to produce viral proteins or packaged by viral proteins to form new core particles. Cores form and mature while budding from the cell surface, during which time they obtain their envelopes consisting of cellular membrane and viral envelope glycoproteins (modified from ref. 1).

viral RNA genome facilitate a strand transfer of the reverse transcriptase and nascent minus strand DNA that is necessary for DNA synthesis to continue.9 Utilization of the polypurine tract (ppt) located near the 3’ end of the genome to initiate synthesis of the DNA plus strand followed by a second strand-switching event results in the completion of reverse transcription and the existence of a full-length double-stranded DNA copy of the viral genome containing LTRs at each end. This genome is integrated, in an apparent random fashion, into a chromosome of the host cell where it resides as the provirus. The integration reaction11,12,13 is mediated by the viral integrase protein; the termini of the DNA genome are processed (two nucleotides on each end are removed), there is a break in and a short duplication made of the cellular DNA sequence, the viral DNA is inserted, and DNA repair enzymes complete the integration process. Because oncoretroviruses, unlike lentiviruses, lack a nuclear transport function to move the proviral preintegration complex into an intact nucleus, the integration process occurs only in cells where the nuclear envelope has broken down as a part of the mitotic process.14

Introduction to Retroviruses and Retroviral Vectors

5

Once integration of the viral genome is complete, the provirus is maintained in the cell just like any cellular gene. A cellular RNA polymerase uses the viral promoter/enhancer in the 5’ LTR to initiate transcription of proviral DNA. RNA transcripts are polyadenylated and transported from the nucleus for translation as any cellular mRNA would be. The viral Gag and Pol proteins are translated from a full-length unspliced RNA that extends from the 5’ LTR to the 3’ LTR. Usually, during translation of the retroviral RNA, translation terminates after reading of the gag open reading frame by the ribosomal complexes, resulting in significant Gag production. However, about 5% of the time there is production of a Gag-Pol fusion polyprotein precursor [some studies have indicated that the stoichiometry of Gag and Pol production appears to be important for efficient and correct virion assembly15,16]. The mechanism of production of the Gag-Pol precursor varies among the different retroviruses: in some retroviruses, production of Gag-Pol during translation is a result of a ribosomal frameshift event that puts the pol gene into the same reading frame as the gag gene;17 in other retroviruses, Gag-Pol production results from a readthrough event mediated by a suppressor tRNA that prevents termination of translation at the end of the gag gene.18 The envelope protein is expressed from a spliced message. This protein precursor undergoes post-translational modification including extensive glycosylation and processing by a cellular protease to generate the two TM and SU glycoprotein subunits that make up the functional Env protein. Full-length transcripts of the viral genome, in addition to being translated into viral proteins, also can be incorporated into newly forming viral particles.19 This “packaging” (or “encapsidation”) of the viral genome into newly forming capsids is mediated by an RNA packaging (encapsidation) signal located on full-length viral transcripts. This predominate packaging signal, sometimes referred to as “ψ” or “E,” is generally located near the 5’ end of the genome between the splice donor and the gag start codon (Fig. 3). Because it is typically located within an intron, the packaging signal is removed during splicing, insuring that only fulllength viral transcripts and not spliced transcripts (containing only env) are incorporated into progeny virions. Although the locations of retroviral packaging sequences are generally thought to be as described above, further study of the packaging signals of various retroviruses has revealed that the actual situation is, not surprisingly, more complex.20,21 In some cases, sequences upstream of the splice donor have been shown to be important in RNA packaging. In other cases, sequences extending past the gag start codon have been shown to increase efficiency of RNA encapsidation (these “extended” packaging signals are oftentimes referred to as “ψ+” or “E+”). Still, in other cases, sequences distant from the major packaging signal have been suggested to affect encapsidation of viral RNA. It is generally believed that the secondary structure of the RNA within the packaging signal, and not the primary RNA sequence itself, plays the more important role in RNA encapsidation. The process and relative timing of many events of virus particle assembly are not understood completely. Newly synthesized Gag proteins recognize RNA that is to be packaged into virions and incorporate the RNA into capsids formed of multimers of Gag molecules. Recognition of the viral encapsidation sequence is thought to be mediated primarily by the NC portion of the Gag protein precursor. Viral core formation is thought to be driven by intermolecular Gag-Gag interactions, with the Pol protein being incorporated into the forming virion via Gag molecules interacting with the Gag portion of the Gag-Pol precursor. Viral core assembly and processing of the Gag and Gag-Pol protein precursors to form mature, infectious virions appears to occur during and just after budding of virus particles from the cell.22 The viralencoded protease (PR, encoded by pol) self-cleaves the Gag-Pol precursor and also further processes Gag into the MA, CA, and NC proteins and Pol into the RT and IN proteins. The

6

Lentiviral Vector Systems for Gene Transfer

Fig. 3. Retroviral vector cis-acting elements. In this example of a “typical” retroviral vector, the viral genes have been removed and replaced with a foreign gene of interest. The viral sequences that remain as part of the vector construct are necessary in cis during various steps of the retroviral replication cycle. These sequences are necessary for vector production and for successful reverse transcription and integration of the vector genome, followed by expression of the foreign gene. Although in the example shown the foreign gene is expressed directly from the LTR, other strategies for expressing foreign genes exist and are illustrated in Figure 4. att, attachment site; pbs, primer binding site; ppt, polypurine tract. Modified from ref. 1.

virion envelope, made up of cellular membrane and viral envelope glycoproteins, is obtained during budding of the virus from the cell.

Elements of Retroviral Vector Systems General Concepts Retroviral vectors are derivatives of viruses that have been engineered to carry a foreign gene of interest into a target cell. Generally, for studies involving gene transfer or examination of the basic viral replication cycle, the vectors are engineered to be replication-defective, being able to complete only a single round of the retroviral replication cycle (though in some instances vectors capable of continued self-propagation might be used; ref. 23). Because of the way replication-defective retroviral vectors are designed, virus particles containing vector genomes can be produced and can be used to infect target cells. The vector genome then undergoes reverse transcription and integration into the cell’s genome, where it can express the foreign gene(s) of interest, but is unable to be replicated an additional time and spread to other cells; the vectors can undergo only a single round of replication. For studies of basic viral biology, this property enables researchers to study in detail many aspects of the replication cycle; for gene transfer studies, it enables foreign genes to be permanently introduced into cells without exposing them to replicating virus. Building a replication-defective vector from the parental retrovirus necessitates separating the cis- and trans-acting sequences of the viral genome. In a practical sense, this entails removal of the trans-acting gag, pol, and env genes from the virus (and replacing them with a foreign gene of interest), leaving on the genome only those cis-acting regions that are recognized by viral and cellular proteins during the various stages of the viral replication cycle—reverse transcription, integration, transcription, encapsidation—as reviewed above. These cis-acting regions are shown in Figure 3. Obviously, necessary viral proteins that make up the physical structure of the virion and that perform enzymatic functions need to be provided in order to produce infectious vector

Introduction to Retroviruses and Retroviral Vectors

7

virus. This requirement can be satisfied by expression of the gag, pol, and env genes, removed from the parental virus during vector construction, in cells that also express the vector construct. In this way, Gag, Pol, and Env can be provided in trans but the gag, pol, and env genes will not be carried along with the vector when it is harvested and used to infect target cells. Various strategies24 for expression of vector constructs and viral genes and for production of vector virus are discussed below. Vector virus that is produced by this trans-complementation can then be used to transduce (infect and express a foreign gene in) target cells. For experiments using vectors to study a single round of viral replication, the foreign gene is generally some type of marker gene that can be used to screen for transduction and to quantify the vector virus that is produced; typical markers used to titer vector produced include lacz or genes encoding GFP or a molecule conferring antibiotic resistance.

Design of Vectors Once the basic vector backbone has been constructed (Fig. 3), there are several different ways in which to express a foreign gene (Fig. 4). In the simplest case, the foreign gene is expressed directly from the promoter located in the 5’ LTR of the vector construct. Two or more genes can be expressed from the LTR if the second gene is expressed from a spliced message or if the second gene is translated using an internal ribosome entry site (IRES) (ref. 25). Although using the vector LTR to drive expression of the foreign gene(s) is useful in a number of settings, at times other strategies are used in order to attempt to control or to increase the level of gene expression. By designing vectors that contain internal, heterologous promoters, oftentimes foreign gene expression can be increased greatly compared to levels that could be obtained using the promoter in the vector LTR (this is attributable to low LTR promoter activity in the target cell, which typically is not the natural target cell of the parental virus). Frequently, the heterologous promoter is another viral promoter, such as cytomegalovirus (CMV), or it may be a tissue-specific promoter. Of course, the years of experience with retroviral vector development have made it clear that the design of vectors to deliver and express foreign genes in cells is often not as simple as one might think from the description above. For example, sustained expression of the foreign genes has been a great problem that can have many possible etiologies. For instance, although use of a heterologous promoter can increase foreign gene expression, addition of an additional promoter on the vector sometimes can actually decrease gene expression. This phenomenon, termed “promoter interference,” is incompletely understood and not always predictable: often it can be affected by the precise or relative location of the promoters or even by the foreign gene itself.26,27 Efforts to overcome promoter interference have included attempts to express the foreign gene from a heterologous promoter/foreign gene cassette placed on the vector in an anti-sense orientation relative to the LTRs. This strategy has had mixed success, probably since the same number of promoters are still present on the vector and possibly because the production of anti-sense transcripts may interfere with vector production or gene expression. The development of self-inactivating (SIN) vectors may decrease the problem of promoter interference and also offers a potential safety advantage over traditional retroviral vectors. SIN vectors are named as such because they are engineered to generate, following reverse transcription of the vector RNA into the DNA form, a defective, and thus inactive, promoter in the 5’ LTR. This is accomplished by engineering constructs that have a defective u3 region at the 3’ end of the viral RNA form of the genome (this defective u3 is duplicated as a defective U3 of the 5’ LTR during reverse transcription; see above and Chapter 2). In this scenario, there would be no active promoter in the proviral 5’ LTR to interfere with internal, heterologous promoters. It also, in theory, offers additional safety advantages in that it would be more difficult to re-generate a wild-type parental retrovirus via recombination and also it may decrease

8

Lentiviral Vector Systems for Gene Transfer

Fig. 4. Vector design strategies for expression of foreign genes. Examples of various strategies for expression of foreign genes delivered by retroviral vectors are illustrated. Other combinations of the strategies shown here also can be used. A. The foreign gene is expressed directly from the promoter located in the vector LTR. B. Expression of two or more foreign genes might be expressed using the promoter in the vector LTR by utilizing a spliced message to express the second gene. C. A foreign gene can be expressed from a heterologous promoter located in the middle of the vector; sometimes this promoter/foreign gene cassette is placed in the anti-sense orientation relative to the vector LTR. D. An internal ribosome entry site (IRES) can be utilized to express a second foreign gene. LTR, long terminal repeat; S.D., splice donor; S.A., splice acceptor; Pr, promoter. Arrows indicate the location and direction of transcription initiation.

the chances of insertional mutagenesis by a promoter insertion mechanism after the provirus is integrated into the cell’s genome.

Design of Packaging Systems There are three basic strategies for expressing viral proteins for trans-complementation of a replication-defective vector. These include co-infection of vector-producing cells with wildtype virus (“helper virus”), transient transfection of cells with plasmids expressing vector and protein-coding constructs (“packaging plasmids”), and the use of cells (“packaging cells” or “helper cells”) that stably express viral proteins. If cells containing a vector genome are infected with replication competent wild-type virus, the wild-type virus will complete the viral replication cycle as described above. The viral proteins produced as part of this process, however, also will recognize the vector RNA and will incorporate vector RNA genomes into virus particles. Both wild-type virus and vector virus will be released from cells. This vector virus can be harvested and used to infect target cells; however, because these preparations would be contaminated with wild-type virus, any target cells transduced with the vector would also be infected with wild-type virus. Therefore, continued replication and spread of the vector to other cells would occur. In certain experimental situations this could be a useful phenomenon, but in most cases a mixture of vector virus and wild-type virus would confound experimental results and make them difficult or impossible to interpret. Obviously, the presence of wild-type virus in vector preparations would make them unacceptable for clinical gene transfer studies. In order to limit vector replication to a single round, separate viral protein-coding constructs can be used to trans-complement the vector. The gag, pol, and env genes are removed from the virus and expressed on separate plasmids using heterologous promoters. These packaging and vector plasmids can be co-transfected into cells. The viral proteins produced will

Introduction to Retroviruses and Retroviral Vectors

9

package vector RNA, and vector virus that is released from these transfected cells can be harvested for transduction of target cells. Because the helper packaging plasmids do not contain the cis-acting sequences necessary for propagation (see above), they will not be packaged or transferred to the target cells; therefore, there will be only a single round of vector replication and the foreign gene can be introduced into target cells in a relatively predictable manner. In order to decrease the chances of the re-generation of wild-type virus (replication-competent retrovirus, RCR) during vector production via recombination between vector and packaging constructs, use of a “split genome” approach to expressing viral proteins generally has been adopted. In such an approach, the vector is expressed on one construct, the gag and pol genes on another, and the env gene on yet another. Separate expression of the env gene also enables the use of Env from heterologus viruses in place of the native Env. The use of Env from heterologous viruses, termed pseudotyping, is extremely valuable because it allows the cellular tropism of the vector virus to be different from that of the parental virus from which the vector was derived. In addition, different envelope proteins have different physical properties that can be an advantage during vector preparation.28,29 The split genome approach decreases the frequency of RCR generation, given that multiple recombination events would need to take place. However, it does not reduce it to zero, especially when using transient transfection of cells, an inherently recombinagenic procedure, to produce vector virus. For this reason, packaging cell lines often are used to generate vector virus. Packaging cells are cell lines that have been engineered to stably maintain viral genes and to express viral proteins.30 The viral genes are expressed from heterologous promoters and there are no cis-acting sequences associated with the viral genes, just as when packaging plasmids are used in transient transfections as described above; usually the split genome approach is used to express gag, pol, and env. When a vector construct is introduced into the packaging cells, usually by transfection, vector virus is produced and can be harvested (Fig. 5). Generally, the use of packaging cells was thought to be preferential to the use of a transient transfection protocol for vector production: packaging cells can be well characterized and it was thought that the chances of RCR formation through recombination was less than during a transient co-transfection of vector and packaging plamsids. The use of packaging cells was generally a more direct way to produce vector virus and yielded higher vector titers than those obtained from transient transfections; however, improvements in protocols for vector production have made this factor less of a concern. One final word on vector production and design of packaging systems: the biggest safety concern31 during vector production is that RCR might inadvertently be generated.32 Therefore, improvements designed to reduce the amount of sequence homology among constructs used in vector and packaging systems has remained a priority. Still, all vector preparations used need to be extensively tested for the presence of RCR.

Summary Vector systems based on oncoretroviruses have been important tools for understanding basic retroviral replication and for applied studies involving gene transfer. Development of these vector systems enabled detailed study of the basic biology of the parent retrovirus from which they were derived and vice versa, enabling advances in vector systems and gene transfer technology to be made. Use of retroviral vectors for gene transfer has been limited for a number of years by several problems including low vector titers, inability to achieve sustained foreign gene expression in target cells, inability to target vector virus to transduce the desired cell type, the theoretical possibility of insertional mutagenesis by the vector upon integration, and the possibility of the re-generation wild-type virus during vector production. Over time, advances in the understanding of retroviral vector design and gene transfer have occurred and progress has been made in overcoming what had been identified as system limitations.

Lentiviral Vector Systems for Gene Transfer

10

Fig. 5. Vector production using a packaging cell line. Packaging cells are cell lines that have been engineered to stably express viral proteins from heterologous promoters; these viral protein-coding sequences lack cisacting sequences that are necessary for propagation and are therefore not transmitted to other cells. Following introduction of a vector into the packaging cell, usually by transfection, vector RNA is produced and packaged into vector virus particles that are then released from the packaging cell. These vector virus particles can be harvested and used to infect fresh target cells. Following reverse transcription and integration of the vector genome, the foreign gene is expressed in the target cells. Because viral proteins are not produced in the target cells, further vector production and propagation does not occur (1).

Recently, vector systems derived from several different lentiviruses, which could offer some advantages over oncoretroviral vector systems, have been under development. At least some of these systems are anticipated to have use in future clinical gene transfer applications. Certainly, the immense experience obtained with oncoretroviral vector systems and the basic tenets that guided their development has benefited and will continue to benefit the development of lentiviral vector systems which, because of their more complex genome and their potential to be human pathogens, raise technical, practical, and ethical issues that must be addressed.

Acknowledgements I thank Kathleen Boris-Lawrie of the Ohio State University for comments on the manuscript and Scott Buchschacher for preparation of figures.

References 1. Buchschacher GL, Jr., Wong-Staal F. Development of lentiviral vectors for gene therapy for human diseases. Blood 2000; 9:2499-2504. 2. Temin HM, Mizutani S. RNA-dependent DNA polymerase in virions of Rous sarcoma virus. Nature 1970; 226:1211-1213. 3. Poeschla EM, Buchschacher GL, Jr., Wong-Staal F. Etiology of cancer: RNA viruses. In: DeVita VT, Jr., Hellman S, Rosenberg SA, eds. Cancer: Principles and Practice of Oncology. 6th ed. Philadelphia: Lippincott, Williams & Wilkins, 2000:149-158. 4. Fearon ER. Oncogenes and tumor suppressor genes. In: Abeloff MD, Armitage JO, Lichter AS, Niederhuber JE, eds. Clinical Oncology. 2nd ed. Philadelphia: Churchill Livingstone, 2000:77-118.

Introduction to Retroviruses and Retroviral Vectors

11

5. Nermut MV, Hockley DJ. Comparitive morphology and structural classification of retroviruses. New Biol 1992; 4:1-24. 6. Coffin JM. Structure and classification of retroviruses. In: Levy JA, ed. The Retroviridae. New York: Plenum Press, 1992:19-50. 7. Coffin JM. Retroviridae: The viruses and their replication. In: Fields BN, Knipe DM, Howley PM et al, eds. Fundamental Virology. 3rd ed. Philadelphia: Lippincott-Raven Publishers, 1996:763-843. 8. Flint SJ, Enquist LW, Krug RM et al. Reverse transcription and integration: Hallmarks of the retroid viruses. In: Principles of Virology-Molecular Biology, Pathogenesis, and Control. Washington DC: ASM Press, 2000:199-234. 9. Panganiban AT, Fiore D. Ordered interstrand and intrastrand DNA transfer during reverse transcription. Science 1988; 241:1064-1069. 10. Hu Wei-Shau, Temin HM. Retroviral recombination and reverse transcription. Science 1990; 250:1227-1233. 11. Grandgenett DP, Mumm SR. Unraveling retrovirus integration. Cell 1990; 60:3-4. 12. Panganiban AT. Retroviral reverse transcription and integration. 1990; 1:187-194. 13. Skalka AM. Retroviral integration. In: Advances in virus research. Maramorosch K, Murphy FA, Shatkin AJ, eds. Seminars in Virology: Retroviral DNA Integration. Vol 52. New York: Academic Press, 1999. 14. Lewis PF, Emerman M. Passage through mitosis is required for oncoretroviruses but not for the human immunodeficiency virus. J Virol 1994; 68:510-516. 15. Schwartzberg P, Colicelli J, Gordon ML, Goff SP. Mutations in the gag gene of Moloney murine leukemia virus: Effects on production of virions and reverse transcriptase. J Virol 1984; 49:918-924. 16. Felsenstein KM, Goff S. Expression of the gag-pol fusion protein of Moloney murine leukemia virus without gag protein does not induce virion formation or proteolytic processing. J Virol 1988; 62:2179-2182. 17. Jacks T, Power FR, Masiarz PA et al. Characterization of ribosomal frameshifting in HIV-1 gagpol expression. Nature 1988; 331:280-283. 18. Weaver TA, Talbot KJ, Panganiban AT. Spleen necrosis virus gag polyprotein is necessary for particle assembly and release but not for proteolytic processing. J Virol 1990; 64:2642-2652. 19. Butsch M, Boris-Lawrie K. Translation is not required to generate virion precursor RNA in human immunodeficiency virus type 1-infected T cells. J Virol 2000; 74:11531-11537. 20. McBride MS, Panganiban AT. The human immunodeficiency virus type 1 encapsidation site is a multipartite RNA element composed of functional hairpin structures. J Virol 1996; 70:2963-2973. 21. Jewell NA, Mansky LM. In the beginning: Genome recognition, RNA encapsidation and the initiation of complex retrovirus assembly. J Gen Virology 2000; 81:1889-1899. 22. Vogt VM. Proteolytic processing and particle formation. Curr Top Microbiol Immunol 1996; 214:95-132. 23. Kucherova L, Altanerova V, Altaner C, Boris-Lawrie K. Bovine leukemia virus structural gene vectors are immunogenic and lack pathogenicity in a rabbit model. J Virol 1999; 73:8160-8166. 24. Boris-Lawrie KA, Temin HM. Recent advances in retrovirus vector technology. Curr Opin Gen Dev 1993; 3:102-109. 25. Adam MA, Ramesh N, Miller AD et al. Internal initiation of translation in retroviral vectors carrying picornavirus 5’ nontranslated regions. J Virol 1991; 65:4895-4990. 26. Emerman M, Temin HM. Genes with promoters in retrovirus vectors can be independently suppressed by an epigenetic mechanism. Cell 1984; 39:459-467. 27. Emerman M, Temin HM. Comparison of promoter suppression in avian and murine retrovirus vectors. Nucleic Acids Res 1986; 14:9381-9396. 28. Burns JC, Friedmann T, Driever W et al. Vesicular stomatitis virus G glycoprotein pseudotyped retroviral vectors: Concentration to very high titer and efficient gene transfer into mammalian and nonmammalian cells. Proc Natl Acad Sci USA 1993; 90:8033-8037. 29. Hopkins N. High titers of retrovirus (vesicular stomatitis virus) pseudotypes, at last. Proc Natl Acad Sci USA 1993; 90:8759-8760. 30. Miller AD. Retrovirus packaging cells. Human Gene Ther 1990; 1:5-14. 31. Temin HM. Safety considerations in somatic gene therapy of human disease with retrovirus vectors. Hum Gene Ther 1990; 1:111-123. 32. Chong H, Starkey W, Vile RG. A replication-competent retrovirus arising from a split-function packaging cell line was generated by recombination events between the vector, one of the packaging constructs, and endogenous retroviral sequences. J Virol 1998; 72:2663-2670.

HIV-1 Replication

13

CHAPTER 2

HIV-1 Replication Eric O. Freed

Introduction

I

n general terms, the replication cycle of lentiviruses, including HIV-1, closely resembles that of other retroviruses.1 There are, however, a number of unique aspects of HIV replication; for example, the HIVs and SIVs target receptors and coreceptors distinct from those used by other retroviruses. Lentiviruses encode a number of regulatory and accessory proteins not encoded by the genomes of the prototypical “simple” retroviruses. Of particular interest from the gene therapy perspective, lentiviruses possess the ability to productively infect some types of non-dividing cells. This chapter, while reiterating certain points discussed in Chapter 1, will attempt to focus on issues unique to HIV-1 replication. The HIV-1 genome encodes the major structural and non-structural proteins common to all replication-competent retroviruses (Fig. 1, and Chapter 1). From the 5'- to 3'-ends of the genome are found the gag (for group-specific antigen), pol (for polymerase), and env (for envelope glycoprotein) genes. The gag gene encodes a polyprotein precursor whose name, Pr55Gag , is based on its molecular weight. Pr55Gag is cleaved by the viral protease (PR) to the mature Gag proteins matrix (also known as MA or p17), capsid (CA or p24), nucleocapsid (NC or p7), and p6. Two spacer peptides, p2 and p1, are also generated upon Pr55Gag processing. The pol-encoded enzymes are initially synthesized as part of a large polyprotein precursor, Pr160GagPol, whose synthesis results from a rare frameshifting event during Pr55Gag translation. The individual pol-encoded enzymes, PR, reverse transcriptase (RT), and integrase (IN), are cleaved from Pr160GagPol by the viral PR. The envelope (Env) glycoproteins are also synthesized as a polyprotein precursor (Fig. 1). Unlike the Gag and Pol precursors, which are cleaved by the viral PR, the Env precursor, known as gp160, is processed by a cellular protease during Env trafficking to the cell surface. gp160 processing results in the generation of the surface (SU) Env glycoprotein gp120 and the transmembrane (TM) glycoprotein gp41. gp120 contains the determinants that interact with receptor and coreceptor, while gp41 not only anchors the gp120/gp41 complex in the membrane (Fig. 2), but also contains domains that are critical for catalyzing the membrane fusion reaction between viral and host lipid bilayers during virus entry. Comparison of env sequences from a large number of virus isolates revealed that gp120 is organized into five conserved regions (C1-C5) and five highly variable domains (V1-V5). The variable regions tend to be located in disulfide-linked loops. gp41 is composed of three major domains: the ectodomain (which contains determinants essential for membrane fusion), the transmembrane anchor sequence, and the cytoplasmic tail. In addition to the gag, pol, and env genes, HIV-1 also encodes a number of regulatory and accessory proteins. Tat is critical for transcription from the HIV-1 LTR and Rev plays a major Lentiviral Vector Systems for Gene Transfer, edited by Gary L. Buchschacher, Jr. ©2003 Eurekah.com.

14

Lentiviral Vector Systems for Gene Transfer

Fig. 1. Organization of the HIV-1 genome. The relative locations of the HIV-1 open reading frames gag, pol, env, vif, vpr, vpu, nef, tat, and rev are indicated. The 5' and 3' LTRs are shown, with U3, R, and U5 regions noted. The ψ indicates the position of the RNA packaging signal. The major Gag domains (MA, CA, NC, p6) and the Gag spacer peptides (p2 and p1) are shown under the gag gene. The site of Gag N-terminal myristylation is denoted as “myr”. Under the pol gene are indicated the PR, RT (p66 and p51 subdomains), and IN coding regions. The SU and TM Env glycoproteins (gp120 and gp41, respectively) are enlarged to show the position in gp120 of the major conserved (C1-C5) and variable (V1-V5) regions and in gp41 the location of the fusion peptide, the N- and C-helices, membrane-spanning domain, and the cytoplasmic tail.

role in the transport of viral RNAs from the nucleus to the cytoplasm. Vpu, Vif, Vpr and Nef have been termed “accessory” or “auxiliary” proteins to reflect the fact that they are not uniformly required for virus replication. The functions of these very interesting proteins will be discussed in more detail at the end of this chapter. HIV replication proceeds in a series of events that can be divided into two overall phases: “early” and “late” (Fig. 3).1 Although some events occur in a concerted or simultaneous fashion, the replication cycle can be viewed most simply as proceeding in an ordered, step-wise manner. In this chapter, each step in virus replication will be considered; additional information can be obtained from the more detailed reviews and primary references that are cited.

The HIV-1 Replication Cycle Virus Entry CD4 Binding Early in the AIDS epidemic it was observed that HIV specifically targets and depletes the CD4+ subset of T-lymphocytes in the peripheral blood of infected individuals. By the mid 1980s, it was well established that the CD4 molecule was the primary cell-surface receptor for SIV/HIV, making it the first retroviral receptor identified.2 Following its identification as the principal HIV receptor, numerous studies attempted to identify the domains of both CD4 and gp120 responsible for the CD4/Env interaction. The high-affinity binding site on CD4 for gp120 maps to a small segment of the N-terminal extracellular domain. The region in Env involved in CD4 binding is located primarily in the third (C3) and fourth (C4) conserved domains of gp120, although residues elsewhere in gp120 also play a role. The recently obtained

HIV-1 Replication

15

Fig. 2. Schematic representation of a mature HIV-1 particle. Positions of the major viral proteins, the lipid bilayer, and the genomic RNA are indicated. Modified from Freed, 1998 (ref. 22)

crystal structure of the gp120 “core” reveals two major domains, the “inner” and “outer” domains, connected by a “bridging sheet”.3 One of the interesting features of the gp120 structure evident from this work is that the CD4 binding site in gp120 is deeply recessed and flanked by heavily glycosylated variable regions. A number of direct gp120/CD4 contacts are resolved in this structure. As a consequence of the high affinity interaction between gp120 and CD4, these two molecules associate during their transport to the surface of infected cells in the late phase of the replication cycle.4 This intracellular Env/CD4 interaction leads to the downmodulation of CD4 from the cell surface, rendering infected cells partially resistant to further infection by HIV Env-bearing viruses. This Env-mediated “superinfection interference” is not operative against virions bearing heterologous Env glycoproteins [e.g., those of amphotropic murine leukemia virus (A-MuLV) or the vesicular stomatitis virus G glycoprotein (VSV-G)] as such pseudotyped viruses bind and enter the target cell in a CD4-independent manner.

Coreceptor Interactions After the identification of CD4 as the major HIV receptor, it was soon appreciated that CD4 was not sufficient for HIV Env-mediated membrane fusion and virus entry. This conclusion was based in part on the observation that primary virus isolates from infected individuals display variable tropism for CD4+ cells. Certain isolates, referred to as macrophage-tropic (or “M”-tropic) replicate efficiently in primary macrophage cultures, whereas other isolates, referred to as T-cell-line tropic (“T”- or “TCL”-tropic) cannot productively infect macrophages but replicate to high levels in T-cell lines. Both M- and TCL-tropic isolates replicate in activated peripheral blood mononuclear cells (PBMC). A decade-long search ultimately identified members of the G protein-coupled receptor superfamily of seven-transmembrane domain proteins as coreceptors for HIV entry. 5,6 These molecules serve as receptors for the α and β

16

Lentiviral Vector Systems for Gene Transfer

Fig. 3. Schematic representation of the HIV-1 life cycle. The major steps in the early and late stages of the replication cycle (described in detail in the text) are indicated.

chemokines. Two coreceptors appear to be predominately important in vivo: the α-chemokine receptor CXCR4 (originally designated fusin) and the β-chemokine receptor CCR5. The identification of the coreceptors largely explains the basis for the differential cell-type tropism mentioned above: T-cell lines typically express CXCR4 but not CCR5; primary lymphocytes express both CXCR4 and CCR5, and macrophages express CCR5. To reflect the importance of coreceptors in HIV biology, a system of nomenclature was developed based on coreceptor usage: strains (generally TCL-tropic) that preferentially use CXCR4 are named X4 viruses, isolates (generally M-tropic) that utilize CCR5 are denoted R5 isolates, and dual-tropic strains that utilize both CCR5 and CXCR4 are termed R5X4 isolates. The V3 loop of gp120 plays a major role in determining HIV-1 tropism; in fact, exchanging the V3 region between isolates can confer M-tropism upon TCL-tropic clones.7 The V1/ V2 region also appears to influence coreceptor usage. In addition to the V1/V2 and V3 variable loops, highly conserved regions of gp120 take part in coreceptor binding. Antibodies whose epitopes are exposed upon CD4 binding block gp120/CCR5 interaction; these epitopes lie in part within the gp120 “bridging sheet”. It has been observed that binding of gp120 to CD4 changes the conformation of gp120 such that its affinity for coreceptor is increased. This finding suggests the following series of events leading up to membrane fusion: gp120 first binds CD4; a ternary complex composed of gp120, CD4, and coreceptor then forms, and finally conformational changes in gp41 ultimately trigger membrane fusion 5-7 (Fig. 4). Interestingly, certain HIV and SIV isolates are able to bind coreceptor and infect cells in the absence of CD4.8 Thus, it appears that the gp120 in these isolates is in a constituitively “activated” conformation. The importance of coreceptors in vivo is illustrated by a number of studies indicating that genetic heterogeneity at coreceptor alleles can affect the susceptibility of an individual to HIV infection, or can alter the course of the disease following infection. The best characterized example of this phenomenon is the so-called CCR5/∆32 mutation; individuals homozygous for this mutant allele (which encodes a truncated form of the CCR5 protein) are almost

HIV-1 Replication

17

Fig. 4. Steps leading to membrane fusion induced by HIV-1 Env. Panel 1. The gp120/gp41 complex in its “resting” configuration prior to interaction with CD4 and coreceptor. Panel 2. gp120 binds CD4 and interacts with coreceptor, leading to conformational changes in both gp120 and gp41. Panel 3. The gp41 ectodomain adopts a hypothetical extended conformation; the fusion peptide at the N-terminus of gp41 inserts directly into the target cell lipid bilayer. Panel 4. The N- and C-helices of the gp41 ectodomain fold into a highly stable six-helix bundle, bringing the membranes in apposition and allowing membrane fusion to occur. Modified from Freed and Martin, ©2001 (see ref. 1).

completely resistant to HIV infection.9-11 These observations suggest that coreceptors might be effective targets for antiviral therapy.

Membrane Fusion The membrane fusion reaction that takes place between the lipid bilayers of the viral envelope and the host cell plasma membrane enables the viral core to gain access to the cytoplasm and is thus central to the infection process. As mentioned above, membrane fusion is catalyzed by the gp120/gp41 Env glycoprotein complex (Fig. 4). The role of gp120 is primarily to bind the target cell, interact with coreceptor, and induce conformational changes in gp41 that allow it to directly promote the fusion event. A number of domains in both gp120 and gp41 participate in CD4/coreceptor binding or the membrane fusion process itself. At the heart of the fusion reaction is the ectodomain of gp41; this region contains a highly hydrophobic N-terminus (the so-called “fusion peptide”) and two heptad repeat motifs, referred to as the N-helix and the C-helix (Figs. 1 and 4). Structural analyses of these two helical sequences, in the context of a gp41 ectodomain trimer, indicate that they pack in an antiparallel fashion to generate a six-helix bundle 12,13 (Fig. 4). Mutations within, or peptides derived from, these heptad repeats potently inhibit membrane fusion. It appears that the N- and C-helices undergo

Lentiviral Vector Systems for Gene Transfer

18

rearrangements following CD4/coreceptor binding that enable the N-terminal fusion peptide to insert directly into the target membrane. In many respects, this “spring-loaded” mechanism of membrane fusion closely parallels the fusion model proposed for influenza HA2.14

Post-Entry Events Following the fusion between the envelope of the incoming virus particle and the host cell plasma membrane, the viral core enters the host cytoplasm. The events that follow remain the least well understood part of the virus life cycle. In particular, the process of “uncoating”, during which the core (defined as the structure that remains after the lipid bilayer is stripped away) is converted to a complex referred to as the reverse transcription complex (RTC) and then the preintegration complex (PIC). During these steps, CA appears to be lost while at least some MA, NC, the pol-encoded enzymes RT and IN, and the accessory protein Vpr, remain associated. Because of the very high ratio of physical particles to infectious units in HIV-1 preparations (approximately 1000:1), these uncoating steps are very difficult to study; the vast majority of viral protein that enters the target cell does not lead to a productive infection. However, successful virus entry can be followed by monitoring reverse transcription and, ultimately, integration.

Reverse Transcription One of the defining features of retroviruses is their ability to convert their RNA genomes into double-stranded DNA early post-infection. This reaction is catalyzed by the RT enzyme.15 In the case of HIV-1, RT is a heterodimer of two subunits, one of which is 66 kDa (p66), the other 51 kDa (p51). These two subunits are both derived from the same region of the Pr160GagPol precursor protein; p51 is formed when the C-terminal, 15kDa RNaseH domain of p66 is removed by PR. A number of groups have crystallized the RT holoenzyme in various contexts, leaving us with a very well-defined structure. Interestingly, the p66 and p51 domains, while largely overlapping in protein sequence, adopt quite different conformations. The p66 subunit can be visualized as a right hand, with the polymerase active site within the palm, and a deep template-binding cleft formed by the palm, fingers and thumb subdomains.16 In this representation, the active site is located in the palm. Reverse transcription proceeds in a series of steps that utilize several cis-acting elements in the viral genome.15 With one interesting distinction, mentioned below, reverse transcription of the HIV-1 genome occurs by a process that is fundamentally similar to that used by other retroviruses. The steps involved in the conversion of the RNA genome to DNA are described briefly below, and are depicted in Fig. 5. 1. Reverse transcription is initiated using as a primer a molecule of tRNA that is bound to the primer binding site (pbs). DNA synthesis proceeds to the 5' end of the RNA molecule, generating a DNA/RNA hybrid. 2. The RNA portion of this hybrid is degraded by the RNaseH activity that is an inherent part of the RT holoenzyme, generating a DNA fragment known as the minus-strand strong stop DNA. 3. By using short regions of homology (the so-called “R” regions), the minus-strand strongstop DNA “jumps” from the 5' to the 3' end of the genome. This step is referred to as the first strand transfer. 4. Minus-strand synthesis occurs, using the 3' end of the minus-strand strong stop DNA as a primer. 5. Plus-strand synthesis occurs, using as primers fragments of RNA remaining from minusstrand synthesis. The primary site of priming for retroviruses takes place at a purine-rich sequence known as the polypurine tract (PPT). For HIV, priming also occurs efficiently from another site, known as the central PPT. The implications of priming at the central PPT will be discussed further in the context of nuclear import.

HIV-1 Replication

19

Fig. 5. Reverse transcription. Thin lines represent RNA; DNA is indicated as a thick line. Cis-acting elements, defined in the text, are shown. The tRNA primer is shown bound to the pbs.

6. The tRNA bound to the pbs is removed by RNaseH, thereby allowing second-strand transfer to take place. 7. Plus-strand synthesis proceeds to the end of the minus strand. For HIV, an additional termination site, referred to as the central termination signal (CTS), is located near the center of the genome. Since the CTS is 3' of the central PPT, approximately 100 nucleotides of plusstrand DNA is displaced, resulting in the formation of a DNA “flap”. It has been reported that this central flap plays a crucial role in the import of the viral PIC to the nucleus.17

20

Lentiviral Vector Systems for Gene Transfer

HIV, and other retroviruses, package two single stranded copies of their RNA genome per virion. Because reverse transcription involves “jumps” from one template to another, the RT/ template interaction is of a relatively low affinity.18 As a consequence, frequent template switches can occur; if the two RNA molecules in a particular virion are not genetically identical, this template switching will result in the generation of a novel recombinant DNA genome containing sequences derived from both parental RNAs.19 The high frequency of genetic recombination, together with the high mutation rate of HIV-1 RT (3 X 10-5 per cycle of replication20) results in HIV populations being highly heterogeneous in sequence (forming so-called “quasispecies”). As a consequence, HIV is able to rapidly evade the host immune response and develop resistance to antiviral drugs. The RT enzyme was an early target in the development of antiviral drugs. Two general classes of RT inhibitors have been developed: the nucleoside analogs [e.g., AZT (zidovudine), ddI (didanosine), ddC (zalcitabine), and 3TC (lamivudine)] and the nonnucleoside inhibitors (e.g., nevirapine and delavirdine). Because of the above-mentioned high mutation rate of RT, resistance to these compounds emerges rapidly; however, potent treatment regimens result from the combination of RT and PR inhibitors.

Nuclear Import During reverse transcription, the viral genome (RNA, then DNA) remains associated with the high molecular weight RTC. The viral DNA is eventually transported to the nucleus as part of the PIC. As mentioned above, the composition of the PIC, and the mechanism by which the PIC translocates to the nucleus, have been the subject of some debate. There appears to be widespread agreement that CA is not part of the PIC, but that at least some MA is retained. The pol-encoded enzyme IN, which of course must be present in the nucleus at the time of integration (see below), is also part of the PIC. The accessory protein Vpr and, in some studies, the Gag NC protein, are also detected in this complex. Although it was originally proposed that MA was the primary determinant of PIC nuclear import,21 certain key pieces of data supporting this model were not reproduced in subsequent studies.22 It is currently believed that Vpr plays a role in nuclear import, and some studies suggest a role for IN.23 Most recently, as alluded to above, the central DNA flap that is a product of lentiviral reverse transcription has been implicated in PIC nuclear import.17 It must be emphasized that our understanding of this process is incomplete and that further studies will be required to elucidate the role of viral, and, presumably, cellular factors in the translocation of the PIC to the nucleus. This topic will be discussed in much greater detail in Chapter 3.

Integration Following nuclear import of the viral PIC, the 32 kDa IN protein catalyzes the insertion of the linear, double-stranded viral DNA into the host cell chromosome.24 The integrated DNA, referred to as the “provirus”, behaves essentially as a cellular gene. Integration is an essential step in retrovirus replication, as IN mutants generally fail to establish spreading infections. As with reverse transcription, integration proceeds via a well-defined series of steps that are quite similar among all retroviruses. 1) The integration process begins when IN clips off several nucleotides from the 3’ termini of both strands of linear viral DNA. This reaction, known as 3'-end processing, generates a molecule of double-stranded DNA with 3’-recessed ends. 2) In the nucleus, IN makes a staggered cleavage in the cellular target DNA. The 3’ recessed ends of viral DNA formed in the 3'-end processing reaction are joined to the ends of the cleaved cellular DNA. This reaction is known as strand transfer. The integration process is completed when cellular repair enzymes fill in the gaps between the integrated viral DNA and the host target DNA.

HIV-1 Replication

21

The IN enzyme is composed of three distinct domains: an N-terminal domain that contains a zinc-finger, a central “core” sequence, and a C-terminal domain. The core contains the active site of the enzyme and bears marked structural similarity to other polynucleotidyl transferases (e.g., the bacteriophage MuA transposase and RNAse H). Three highly conserved residues, Asp64, Asp116 and Glu152 (the so-called D,D-35-E motif) are critical to IN function. It is well established that IN functions as a multimer; however, it is still not clear how many molecules of IN make up the functional holoenzyme. Recent studies suggest that not only is the viral DNA present in a large macromolecular PIC, but that the ends of the viral DNA are also organized into a multi-component complex composed of both viral and cellular proteins. These complexes at the viral DNA ends are referred to as “intasomes”.25,26 Much of what we know about retroviral integration derives from the study of in vitro integration reactions using purified IN. Despite the utility of this simplified system, the major product of such reactions using HIV-1 IN is a single end of viral DNA joined to one strand of the target, rather than the product observed in vivo in which both ends of the viral DNA are integrated. This observation highlights the importance of additional viral and perhaps cellular proteins in carrying out the integration reaction. Indeed, when PICs are purified from infected cells they are able to direct authentic and complete integration reactions in vitro. Such reactions provide a powerful tool to screen for compounds that block integration in vivo.27

Gene Expression Following integration into the host chromosome, the integrated provirus serves as the template for the synthesis of the viral RNAs that ultimately encode the full complement of structural, regulatory, and accessory proteins used to direct virus replication. In this section, we will consider the transcription of the viral RNAs, and the role of the Tat protein in RNA synthesis, and will summarize the role of Rev in the export of viral RNAs to the cytoplasm.

Transcription/Tat Function The HIV-1 LTR serves as the site of transcriptional initiation and harbors cis-acting elements required for RNA synthesis1 (Fig. 6). The LTR is composed of three regions: U3 (for unique, 3' end), R (for repeated) and U5 (for unique, 5' end). Transcription initiates at the U3/ R junction. U3 contains a variety of elements that direct the binding of RNA polymerase II (pol II) to the DNA template. A TATA element, to which transcription factor IID (TFIID) binds, is located approximately 25 nucleotides upstream of the transcription start site. Located 5' of the TATA box are three Sp1 and two NF-κB binding sites. Although mutational analyses of the HIV-1 LTR reveal that the Sp1 and NF-κB sites are variably important, depending upon the cell type, removal of all Sp1 and NF-κB sites abolishes virus replication.28 Upstream of the NF-κB sites is a domain, sometimes referred to as the “modulatory region”, which contains binding sites for several additional transcriptional factors, including LEF, Ets, and USF. The basal transcriptional activity from the HIV LTR is very low; RNA synthesis is greatly increased (by more than two logs) when the transcriptional transactivator protein Tat is present.29,30 Tat is a 101 amino acid protein encoded by a two-exon RNA; a smaller (72 amino acid) “one-exon” Tat is encoded by some isolates. Tat contains several distinct functional domains: an activation domain, which lies within the N-terminal 48 residues of the protein and which itself is comprised of an acidic domain, a Cys-rich region, and a hydrophobic core element; a highly basic RNA binding domain; and an overlapping nuclear localization signal. Much effort over the past decade has been focused on understanding the mechanism by which Tat transactivates LTR-driven gene expression. Several salient features are emphasized below: 1) Tat acts upon an RNA element, known as the transactivation response region (TAR).31 TAR, which is present at the 5' end of all viral RNAs, consists of a base-paired stem, a small

22

Lentiviral Vector Systems for Gene Transfer

Fig. 6. Schematic representation of the HIV-1 LTR. The position of binding sites for host factors (LBP-1, NFκB, LEF, Ets, USF-1, and NFAT-1) are shown at and 5' of the transcription start site. The TAR stem/ loop structure, with bulge, is represented at the 5' end of a nascent mRNA.

(trinucleotide) non-base-paired bulge, and a 6-nucleotide G-rich loop (Fig. 6). 2) Tat appears to bind to the TAR bulge. 3) It was observed a number of years ago that Tat functions poorly in rodent cells, suggesting that cellular factors play an important role in Tat activity. This prediction was borne out recently with the identification of a cellular protein that interacts, via its activation domain, with Tat.32 The protein in question is cyclin T1 (cycT1), which forms a heterodimer with a member of the cyclin-dependent kinase family (CDK9). The cycT1/CDK9 heterodimer is itself part of a large protein complex related to the Drosophila positive-transcriptional elongation factor b (P-TEF-b). Tat recruits the human P-TEFb complex to TAR, resulting in the phosphorylation of the C-terminal domain (CTD) of RNA Pol II and a dramatic stimulation of transcriptional processivity.

RNA Export/Rev Function Transcription from the HIV-1 LTR leads to the generation of a large number (more than 30) of viral RNAs.33 These fall into three major classes: 1) unspliced RNAs, which function as the mRNAs for the Gag and GagPol polyprotein precursors, and are packaged into progeny virions as genomic RNA, 2) partially spliced mRNAs, which are around 5 kb in size and encode the Env, Vif, Vpu, and Vpr proteins, and 3) small (1.7 to 2.0 kb), multiply spliced mRNAs, which are translated into Rev, Tat, and Nef. Since most cellular mRNAs are fully spliced before their transport out of the nucleus, the need for unspliced and partially spliced RNAs in the cytoplasm poses a problem for HIV. This problem has been overcome through the evolution of a novel viral protein, Rev (for “regulator of expression of viral proteins”), and a cis-acting RNA element, the Rev responsive element (RRE).34 Rev is 19 kDa (116 amino acid) phosphoprotein encoded by two exons, both of which are essential for protein function.34 Rev contains two functional domains: an Arg-rich sequence that is required for RNA binding and nuclear localization, and a hydrophobic, Leu-rich motif that mediates nuclear export. The RRE is a large (250 nucleotide), highly structured RNA element that is located in the env gene and is present in all unspliced and partially spliced HIV1 RNAs. It folds into a series of stem-loop structures emanating from a central bubble. A Rev monomer initially binds one of the stem-loops (known as stem-loop 2) and then, through cooperative protein-protein and protein-RNA interactions, multimerizes with other Rev molecules. This Rev multimer eventually coats the RRE at a stochiometry of approximately eight Rev molecules per RRE. Rev binding to the RRE results in the formation of a complex capable of interacting with the cellular nuclear export machinery. As a consequence, the RRE-containing RNA is transported, in an unspliced or partially spliced form, to the cytoplasm. Rev then shuttles back to the nucleus, using its nuclear localization signal. An additional level of

HIV-1 Replication

23

complexity is provided by the presence in HIV-1 RNAs of cis-acting elements, located within gag, pol, and env, whose presence inhibits RNA utilization in the absence of Rev.35,36 These elements likely bind cellular proteins that retain RNAs containing the elements in the nucleus. Rev counteracts this effect, allowing efficient RNA export to proceed. It is probable that the more “simple” retroviruses contain in their RNAs RRE-like sequences that interact with cellular Rev-like proteins to facilitate their transport to the cytoplasm. Indeed, such an element, referred to as the constitutive transport element (CTE) has been identified in the genome of Mason-Pfizer Monkey virus.37 By introducing the CTE in HIV-1 RNAs, the requirement for Rev and the RRE can be bypassed.

Virus Particle Production Following the synthesis of the full complement of viral proteins, the assembly process begins. The major player in virus assembly is the Gag precursor polyprotein, Pr55Gag.22,38 This protein contains determinants that target it to the plasma membrane, bind the membrane itself, promote Gag-Gag interactions, encapsidate the viral RNA genome, associate with the viral Env glycoproteins, and stimulate budding from the cell. Our understanding of virus assembly has been greatly assisted by detailed mutational analyses of all major domains of the Gag precursor, and by NMR and X-ray crystallography studies that have solved the structures of MA, the N-terminal and C-terminal domains of CA, and NC.39 It is important to note, however, that at this time we have little structural information comparing the folding of the mature Gag proteins with the corresponding regions of the precursor protein from which they are derived. Pr160GagPol is synthesized as the result of a frameshifting event during Gag translation. The use of this frameshifting strategy ensures that the Pol proteins are expressed at 5-10% the level of the Gag proteins. Pr160GagPol is packaged into virions via its Gag domain, largely using the same Gag-Gag interactions that drive Gag assembly.

Gag Membrane Binding and Targeting The assembly of lentiviruses, including HIV-1, takes place at the plasma membrane of the infected cell (Fig. 3). The MA domain of Gag, which occupies the N-terminal 131 residues of Pr55Gag, is largely responsible for targeting and binding the plasma membrane. The MA domain is cotranslationally modified at its N-terminus by myristic acid (Fig. 7); this fatty acid modification is essential for Gag membrane binding. Structural studies reveal that HIV-1 MA folds to expose a patch of highly basic residues on the face of MA predicted to juxtapose the plasma membrane. Together with the finding that mutations in these residues can impair membrane binding, this observation suggests that the positively charged face of MA interacts with the negatively charged acidic phospholipids on the inner leaflet of the plasma membrane thereby stabilizing membrane binding. Although MA contains the primary determinants of Gag membrane binding, the mature MA protein binds membrane weakly relative to Pr55Gag. By analogy with cellular proteins (such as recoverin), it has been proposed that MA undergoes a conformational transition following its cleavage from the Gag precursor.40 This conformation change, according to the model, causes a sequestration of the N-terminal myristic acid moiety and a partial loss in membrane binding ability. This so-called “myristyl switch” model could explain how MA might detach from the lipid bilayer and associate with the PIC following infection. HIV-1 Gag associates with membrane within minutes of its synthesis, and this membrane binding appears to be largely specific for the host cell plasma membrane. Although it is not fully understood how Gag specifically targets the plasma membrane in preference to more abundant intracellular membranes, MA, in particular the highly basic domain (Fig. 7), clearly plays a role in this process. Mutations within MA can re-route assembly to intracellular compartments (e.g., the ER or Golgi),41,42 and substitution of MA with a heterologous membrane

24

Lentiviral Vector Systems for Gene Transfer

Fig. 7. Functional domains of the HIV-1 Gag domains MA (A), CA (B), NC (C), and p6 (D). Details are provided in the text. Adapted from Freed, 1998 (ref. 22).

binding signal results in promiscuous targeting of virus assembly to both plasma and intracellular membranes.

RNA Encapsidation As discussed above, each retrovirus particle contains two single-stranded copies of genomic RNA. The cis-acting sequence that directs RNA encapsidation, known variously as the packaging signal, encapsidation element, or ψ-site, is typically located in the region of the RNA 5' of the gag initiation codon.43 In the case of HIV-1, the packaging signal appears to be more dispersed than in the murine and avian retroviruses and a larger region of the genome is required for efficient RNA packaging. The HIV-1 packaging signal is composed primarily of four stem-loop structures, referred to as SL1-SL4, although sequences outside this region contribute to efficient RNA encapsidation. The secondary structure of this packaging signal, rather than the actual nucleotide sequence itself, seems to be important in conferring RNA encapsidation specificity. RNAs that lack the packaging signal are not efficiently encapsidated

HIV-1 Replication

25

into virus particles. Retroviral RNAs are linked together at a sequence near the 5' end of the genome known as the dimer initiation signal (DIS). It is currently unclear to what extent RNA dimerization is required for efficient encapsidation. The specific encapsidation of retroviral RNAs into virus particles is mediated by interactions between the packaging signal and the NC domain of Gag. Nearly all retroviral NC proteins contain one or two zinc finger motifs, each of which coordinates a zinc ion. HIV-1 NC contains two zinc-finger motifs (Fig. 7) of the “CCHC” type (Cys-X2-Cys-X4-His-X4-Cys; where X denotes a variable amino acid). The zinc fingers of HIV-1 NC are flanked by highly basic sequences. NC displays both non-specific nucleic acid binding properties and specific genomic RNA binding; in general, the non-specific binding properties are conferred by the basic residues, whereas the zinc finger motifs, in conjunction with the basic residues, contribute to the specificity of the NC/RNA interaction. It has been proposed that NC first interacts, in a sequence-specific fashion, with the packaging signal (SL3 appears to be particularly important in this regard) then additional NC domains coat the viral RNA in a sequence-independent manner. In addition to its function in RNA encapsidation, NC plays a role in a variety of additional steps in the viral life cycle. In many cases these activities can be attributed to the ability of NC to function as a nucleic acid chaperone.44 This property enables NC to refold nucleic acid molecules to the most energetically favorable conformation. In general, retroviruses are able to package their own RNAs genomes, but not those of other retroviruses. The evaluation of chimeric viruses indicates, as mentioned above, that encapsidation specificity is determined by NC. In some cases, non-reciprocal packaging can be observed; for example, HIV-1 packages both HIV-1 and HIV-2 genomic RNAs, whereas HIV2 reportedly does not efficiently encapsidate HIV-1 RNA.45

Assembly Once Gag has arrived at the plasma membrane, it must engage in Gag-Gag (as well as Gag-lipid and Gag-RNA) interactions to enable the assembly of progeny virions to take place. A number of approaches, including in vitro assembly reactions, mutational analyses, virion incorporation assays, and yeast two-hybrid screens have suggested that several domains in Gag are important in mediating Gag-Gag interactions.22 These domains primarily encompass a region spanning the C-terminus of CA, the p2 spacer peptide, and the N-terminal portion of NC. The HIV-1 CA folds into two distinct structural and functional domains: an N-terminal “core” domain, and a C-terminal “dimerization” domain (Fig. 7). The C-terminal domain harbors the so-called “major homology region” which contains residues conserved among many retrovirus genera. Mutations within the C-terminal dimerization domain often cause defects in virus assembly, suggesting the involvement of this region in Gag-Gag interactions. HIV-1 CA, when expressed in vitro, can direct the assembly of tubular or spherical particles, again highlighting the role of CA in virus assembly. Mutations in NC, specifically within the highly basic residues near the N-terminus of the protein, also induce defects in virus particle production. This finding, together with the observation that CA-NC fusion proteins assemble in vitro more efficiently than does CA alone, have suggested that NC also promotes Gag-Gag interactions. Interestingly, in vitro assembly of CA-NC fusion proteins requires the presence of nucleic acid.46 This latter result suggests a model in which interactions between the Gag NC domain and RNA allow Gag molecules to align and pack. This model is consistent with mutational studies demonstrating that mutation in the basic residues within NC disrupt both Gag assembly and RNA binding.

Env Transport and Incorporation The HIV-1 Env glycoprotein is synthesized in the rough ER to generate the Env precursor protein, gp160.47 The protein is cotranslationally inserted into the lumen of the ER through

26

Lentiviral Vector Systems for Gene Transfer

the use of an N-terminal signal peptide; a stop-transfer signal is located in the gp41 portion of gp160. The gp120 domain is very heavily glycosylated, initially with high-mannose side chains which are converted to complex side chains during transport though the Golgi. In the ER, gp160 forms intramolecular disulfide bonds and undergoes oligomerization. Currently, the functional multimeric form appears to be a trimer, although dimers and tetramers can also be detected. gp160 is transported to the cell surface via the secretory pathway; during its trafficking through the Golgi it is cleaved by a host protease (furin or a furin-like enzyme) to generate the mature SU glycoprotein, gp120, and TM glycoprotein, gp41. Cleavage takes place at the C-terminus of a highly conserved Lys/Arg-X-Lys/Arg-Arg motif. gp160 processing is absolutely required to activate the fusogenic potential of the gp120/gp41 complex, presumably by allowing the exposure of the hydrophobic fusion peptide located at the N-terminus of gp41 (Fig. 1). After gp160 cleavage, gp41 anchors the Env complex in the membrane and associates non-covalently with gp120. The relatively weak nature of the gp120-gp41 interaction results in significant amounts of gp120 shedding from the surface of Env-expressing cells and virions. Env glycoprotein complexes that reach the cell surface are either rapidly internalized (through recognition by host cell machinery of an endocytosis motif in the gp41 cytoplasmic tail) or are incorporated into virus particles. Although the process by which the Env glycoproteins are incorporated into virions remains incompletely understood, a number of lines of evidence suggest that an interaction between the gp41 cytoplasmic tail and the MA domain of Gag recruits Env into virions.22 Observations that provide support for a gp41-MA interaction include: 1) mutations in both MA and the gp41 cytoplasmic tail can disrupt Env incorporation, 2) the effects of MA mutations on Env incorporation can be reversed by truncating the gp41 cytoplasmic tail, and the effects on Env incorporation of a small deletion in the gp41 cytoplasmic tail can be reversed by a single amino acid change in MA.48 3) HIV-1 Env is able to target virus budding to the basolateral surface of polarized epithelial cells, and 4) a direct gp41MA interaction has reportedly been detected in vitro. Despite these findings, the incorporation of heterologous or mutant Env glycoproteins containing short cytoplasmic tails can occur efficiently. It therefore appears that the incorporation of full-length Env, which contains an unusually long (150 amino acid) cytoplasmic tail, requires an interaction with, or at least must be sterically accommodated by MA. In contrast, heterologous Env glycoproteins containing short cytoplasmic tails are not subject to steric restrictions and can be efficiently incorporated. Interestingly, it was recently demonstrated that an Env mutant lacking the gp41 cytoplasmic tail was incorporated in a limited number of cell lines (e.g., HeLa and MT-4) but was largely excluded from virions in most other cell types.49 The cell type-dependent requirement for the gp41 cytoplasmic tail in Env incorporation suggests the involvement of host factors in the Env incorporation process. A practical ramification of HIV’s ability to incorporate heterologous Env glycoproteins is evident in the frequent use of the A-MuLV Env and VSV-G in pseudotyping studies. These glycoproteins, in particular VSV-G, are quite stable and confer markedly higher levels of infectivity than does the HIV-1 Env glycoprotein. In addition, the ability of VSV-G to confer virus infectivity in a wide range of cell types makes it particularly useful in potential gene therapy applications.

Budding The final step in the process of virus assembly and release involves the pinching off, or budding, of the virus particle from the host cell plasma membrane. While it was felt initially that budding was likely to be a spontaneous event, it has become clear that a wide range of retroviruses (and at least several other enveloped viruses) encode specific sequences that promote particle release. These sequences are collectively referred to as “late” or “L” domains to reflect their role late in virus assembly. Retroviruses encode their L domains at a variety of

HIV-1 Replication

27

positions in Gag; in the case of HIV-1, the L domain is present in p6 (Fig. 7). Deletion of p6, or mutations within a highly conserved Pro-Thr/Ser-Ala-Pro (P-T/S-A-P) motif located near the N-terminus of p6, markedly impair particle release.50,51 Examination of cells expressing p6-mutant HIV-1 clones reveals the presence of large numbers of particles attached to the plasma membrane by a thin tether, apparently unable to pinch off from the cell. Although the mechanism by which L domains stimulate virus release remains to be elucidated, evidence is increasing that these domains function by interacting with host factors. Most retroviral L domains contain the motif Pro-Pro-Pro-Tyr (P-P-P-Y), which is the consensus binding sequence for the so-called “WW” family of proteins. Indeed, direct interactions between P-P-P-Y-type L domains and WW proteins have been detected in vitro.52 One of these proteins, Nedd4, is a ubiquitin ligase. This latter observation, together with the finding that proteosome inhibitors, which disrupt ubiquitination, impair both HIV and Rous sarcoma virus release, suggests that L domains may function through the host ubiquitination pathway.53 How this would promote virus release remains an interesting question for future investigation.

Maturation During or shortly after virus release from the plasma membrane, the viral PR cleaves the Gag and GagPol polyprotein precursors to generate the mature Gag and Pol proteins (Figs. 1 and 3). Retroviral PRs are members of the family of aspartyl proteases. X-ray crystallography indicates that retroviral PRs function as dimers, with the substrate-binding site located in a cleft formed by two (identical) monomers.54 PR-mediated Gag and GagPol processing sets in motion a series of structural rearrangements that ultimately leads to virion maturation. PR cleaves each site with a differing efficiency; as a result, PR-mediated Gag and GagPol processing takes place as an ordered, stepwise cascade of cleavage reactions. The most visible outcome of HIV-1 maturation is that virion morphology is converted from doughnut-shaped (containing an electron-lucent center) to containing an electron-dense, conical core. Because of the degree and magnitude of protein rearrangements that are presumed to occur during maturation, this process can be considered as a second assembly (or re-assembly) reaction. Studies performed using cryo-EM techniques have visualized unprocessed retroviral Gag monomers in immature virions as being aligned likes spokes on a wheel, projecting inward from the membrane-associated MA domain at the N-terminus to NC at the C-terminus.55 Following cleavage, CA forms a conical shell around the RNA/protein complex within the core (Fig. 2). Numerous mutations have been reported, many of which are located within the N-terminal domain of CA, that prevent the formation of the normal conical core. Invariably, the failure of the virion to mature properly is associated with a complete loss of infectivity; core condensation thus appears to be essential during an early post-entry step of the replication cycle. Interestingly, core-like structures can assemble in vitro from a CA-NC fusion protein in the presence of RNA. These cones closely resemble the fullerenes formed by elemental carbon.56 During virus assembly, the N-terminal domain of CA binds, and ultimately packages into virions, the host protein cyclophilin A.57,58 Mutations that disrupt CA-cyclophilin A binding, or treatment of virus-infected cells with cyclosporin (which prevents cyclophilin A incorporation) result in markedly impaired virus infectivity. The incorporation of cyclophilin A into virions is HIV-1-specific, as neither HIV-2 nor SIV CA proteins bind this host factor. Although uncertainty remains regarding the role that cyclophilin A plays in stimulating infectivity, it has been suggested that this protein, which functions in the cell as a peptidyl-prolyl cis-trans isomerase, functions as a chaperone during maturation to prevent unfavorable CA aggregation.59 The absolute requirement for PR-mediated virion maturation has been applied to the treatment of HIV-infected individuals using inhibitors of PR. Although in infected patients,

28

Lentiviral Vector Systems for Gene Transfer

the use of PR inhibitors alone rapidly leads to the generation of drug-resistant variants, when combined with anti-RT inhibitors [(in so-called “triple” therapy or “highly active antiretroviral therapy” (HAART)], long-lasting, clinically significant reductions in virus loads can be achieved.

Role of the HIV Accessory Proteins in Virus Replication In addition to the proteins encoded by other replication-competent retroviruses (i.e., the products of the gag, pol, and env genes), and the regulatory proteins (Rev and Tat), lentiviruses encode several additional proteins with a variety of interesting, and in many cases poorly understood, functions.1,60 These proteins are commonly referred to as being “accessory” or “auxiliary” to reflect the observation that at least in culture, they are not essential for virus replication. However, in vivo, these proteins contribute, to varying degrees, to efficient virus spread and disease induction.

Vpu Vpu (for viral protein u) is an 81 amino acid integral membrane phosphoprotein that is unique to HIV-1; with the exception of the highly HIV-1-related SIVcpz, it is not encoded by the genomes of SIV or HIV-2 isolates. Vpu performs two major functions during HIV-1 replication: 1) it enhances the release of virus particles, and 2) promotes the degradation of CD4. Mutational inactivation of Vpu reduces by several-fold the efficiency with which virions are released from virus-expressing cells.61 This release function is independent of the p6 Gag L domain (described above). Interestingly, although divergent retroviruses, such as MuLV and visna, do not encode Vpu proteins, HIV-1 Vpu can stimulate their release as well.62 This observation suggests that the enhancement of release occurs through a general pathway that does not depend on the interaction of Vpu with specific viral factors. HIV-1 has evolved several mechanisms to downregulate cell-surface expression of the major receptor CD4. One mechanism involves intracellular trapping of CD4 by Env glycoproteins, a second operates through Nef (see below) and the third is promoted by Vpu. CD4 degradation by Vpu is mediated through the host ubiquitin/proteasome pathway.63 One outcome of Vpuinduced CD4 degradation is to liberate gp160 from Env/CD4 complexes in the ER, thereby increasing the amount of Env glycoprotein available for transport to the cell surface.

Vpr The vpr gene (for viral protein r) encodes a 14-kDa, 96 amino acid protein that is incorporated efficiently into virions. The incorporation of Vpr depends upon a specific interaction with a Leu-rich motif located near the C-terminus of p6 (Fig. 7). In addition to weakly stimulating gene expression from the HIV LTR, Vpr also induces the arrest of Vpr-expressing cells in the G2 phase of the cell cycle, and reportedly facilitates transport of the viral PIC to the nucleus of infected cells. Although it is not entirely clear why HIV would evolve a cell-cycle arrest function, a number of groups have reported that Vpr rapidly and efficiently arrests cells in G2. Indeed, it has been suggested that de novo Vpr expression is not required and that the Vpr present on incoming virions is sufficient to induce cell-cycle arrest. G2 arrest appears to result from inhibition of the p34cdc2-cyclin B kinase complex.64 It has been proposed that HIV LTR expression is increased during G2, thus perhaps providing a rationale for the evolution of this cell-cycle arrest function. Several observations have suggested that Vpr might play a role in nuclear import of the viral PIC: 1) Vpr mutation reduces virus infectivity in fully differentiated monocyte-derived macrophages, 2) Vpr is detected in PICs, and 3) Vpr, when expressed in cells, localizes to the nucleus. A variety of models have been proposed to account for the ability of Vpr to stimulate nuclear import; these will be discussed in more detail in the next chapter.

HIV-1 Replication

29

Vif Expression of Vif (for viral infectivity factor) is highly conserved among lentiviruses; it is encoded by all lentiviruses except equine infectious anemia virus. Vif mutation can cause profound defects in virus infectivity. Interestingly, the defective phenotype is cell-type dependent and is determined not by the target cell but by the virus-producing cell. Thus, certain cell lines (for example HeLa, COS, 293T, SupT1, CEM-SS and Jurkat) are “permissive” for Vif mutants; virus produced from these lines is fully infectious regardless of the target cell used. In contrast, other cell types (most notably, primary lymphocytes and macrophages) are “nonpermissive”. This cell-type specificity argues that host factors play a role in Vif function, and that the defect observed with vif defective mutants is imposed during virus assembly. Analysis of transient heterokaryons formed between permissive and non-permissive cells has indicated that the non-permissive phenotype is dominant, perhaps suggesting that Vif counteracts the effect of a cellular factor that inhibits the formation of infectious virions.65 Although the Vif-defective phenotype may be imposed during assembly, it is manifested early post-entry as a failure to efficiently reverse transcribe the viral genome. Reminiscent of observations made with certain post-entry Gag mutants, Vif(-) virions have been reported to display defects in proper core condensation. Interestingly, Vif has been detected at low levels in virus particles; however, the implications of virion incorporation for Vif function remain unclear.

Nef Although Nef was originally reported to suppress gene expression from the HIV LTR (hence the name “negative factor”) it is now clear that Nef plays an important positive role in lentiviral pathogenesis. Nef is a 27 kDa, membrane-associated phosphoprotein; like Gag, its membrane binding is dependent upon a myristic acid moiety covalently attached to the N terminus. As is the case for the other HIV accessory proteins, several primary Nef functions have been reported: 1) downregulation of CD4 and major histocompatibility class I (MHC I) molecules from the cell surface, 2) stimulation of virus infectivity in single-round assays, and 3) modulation of cellular activation pathways. As mentioned above, Nef is one of the three viral proteins (along with Env and Vpu) whose expression reduces cell-surface expression of CD4. Nef-induced CD4 downregulation is achieved by increasing the rate at which CD4 is internalized from the plasma membrane66 reportedly via Nef acting as a bridge between CD4 and adapter protein (AP) complexes in clathrin-coated pits. Nef also downregulates cell-surface expression of MHC I molecules, perhaps impairing the ability of cytotoxic T lymphocytes (CTLs) to detect and eliminate virus-expressing cells.67 Although in general the effects of Nef deletion on virus replication kinetics in culture are quite limited, it has been reported that in single-cycle assays, the presence of Nef modestly stimulates virus infectivity. Again, the Nef(-) defect is manifested by a reduction in the amount of viral DNA synthesized post-infection. It is currently unclear to what extent this enhancement of virus infectivity contributes to the requirement for Nef expression in vivo. It has been unambiguously demonstrated that Nef(-) clones of SIV display profound defects in virus replication and disease induction in infected macaques.68 The presence of nef deletions in virus isolates obtained from at least some infected individuals who progress to disease very slowly (the so-called long-term non-progressors) implies that Nef may play a similar role in maintaining high virus loads in HIV-1 infected humans. Although this requirement for Nef in vivo remains unexplained, Nef contains a highly conserved consensus binding site for Src homology region 3 (SH3) domains. Indeed, Nef has been reported to interact with a variety of Src-like kinases and affect their activities.69 The effect of such interactions on signal transduction pathways could stimulate virus replication in vivo. Nef has been detected at low levels in virus particles, where it localizes to the virion core. As is the case for Vif, the implications of these findings remain to be determined.

30

Lentiviral Vector Systems for Gene Transfer

Vpx Although the focus of this chapter is on HIV-1, it is worth noting briefly the functions of an additional protein, Vpx, that is encoded by the genomes of the HIV-2/SIVsm/SIVmac lineage of primate lentiviruses (but not by HIV-1). Vpx bears considerable sequence homology with Vpr, and, like the latter protein, is incorporated at relatively high levels into virions via an interaction with the C-terminus of Gag. Vpx appears to play a role in infection of non-dividing cells but does not induce cell-cycle arrest.70 Thus, the proposed nuclear import/cell cycle arrest functions of HIV-1 Vpr are segregated into two proteins (Vpr and Vpx) in the HIV-2/SIVsm/ SIVmac lentiviruses.

Concluding Remarks During the past 15 years, a wealth of knowledge has been acquired concerning the replication cycle of HIV-1. However, it should be clear from the above discussions that much remains to be learned before our understanding is complete. From the perspective of lentiviral vector development, the mechanism by which HIV-1 infects non-dividing cells is of particular interest. This topic will be the focus of Chapter 3.

References 1. Freed EO, Martin MA. HIVs and their replication. In: Knipe DM, Howley PM, Griffin D, Lamb RA, Roizman B, Martin MA, Straus SE, eds. Fields Virology, 4th ed. Philadelphia: Lippincott, Williams, and Wilkins, 2001:1971-2041. 2. Bour S, Geleziunas R, Wainberg MA. The human immunodeficiency virus type 1 (HIV-1) CD4 receptor and its central role in promotion of HIV-1 infection. Microbiol Rev 1995; 59:63-93. 3. Kwong PD, Wyatt R, Robinson J et al. Structure of an HIV gp120 envelope glycoprotein in complex with the CD4 receptor and a neutralizing human antibody. Nature 1998; 393:648-659. 4. Hoxie JA, Alpers JD, Rackowski JL et al. Alterations in T4 (CD4) protein and mRNA synthesis in cells infected with HIV. Science 1986; 234:1123-1127. 5. Berger EA, Murphy PM, Farber JM. Chemokine receptors as HIV-1 coreceptors: Roles in viral entry, tropism, and disease. Annu Rev Immunol 1999; 17:657-700. 6. Doms RW, Peiper SC. Unwelcomed guests with master keys: How HIV uses chemokine receptors for cellular entry. Virology 1997; 235:179-90. 7. O’Brien WA, Koyanagi Y, Namazie A et al. HIV-1 tropism for mononuclear phagocytes can be determined by regions of gp120 outside the CD4-binding domain. Nature 1990; 348:69-73. 8. Endres MJ, Clapham PR, Marsh M et al. CD4-independent infection by HIV-2 is mediated by fusin/CXCR4. Cell 1996; 87:745-756. 9. Dean M, Carrington M, Winkler C et al. Genetic restriction of HIV-1 infection and progression to AIDS by a deletion allele of the CKR5 structural gene. Hemophilia Growth and Development Study, Multicenter AIDS Cohort Study, Multicenter Hemophilia Cohort Study, San Francisco City Cohort, ALIVE Study. Science 1996; 273:1856-1862. 10. Liu R, Paxton WA, Choe S et al. Homozygous defect in HIV-1 coreceptor accounts for resistance of some multiply-exposed individuals to HIV-1 infection. Cell 1996; 86:367-377. 11. Samson M, Libert F, Doranz BJ et al. Resistance to HIV-1 infection in Caucasian individuals bearing mutant alleles of the CCR-5 chemokine receptor gene. Nature 1996; 382:722-725. 12. Chan DC, Fass D, Berger JM et al. Core structure of gp41 from the HIV envelope glycoprotein. Cell 1997; 89:263-273. 13. Weissenhorn W, Dessen A, Harrison SC et al. Atomic structure of the ectodomain from HIV-1 gp41. Nature 1997; 387:426-430. 14. Bullough PA, Hughson FM, Skehel JJ et al. Structure of influenza haemagglutinin at the pH of membrane fusion. Nature 1994; 371:37-43. 15. Telesnitsky A, Goff SP. Reverse transcription and the generation of retroviral DNA. In: Coffin JM, Hughes SH, Varmus HE, eds. Retroviruses. Cold Spring Harbor: Cold Spring Harbor Laboratory Press, 1997:121-160.

HIV-1 Replication

31

16. Arnold E, Jacobo-Molina A, Nanni RG et al. Structure of HIV-1 reverse transcriptase/DNA complex at 7 A resolution showing active site locations. Nature 1992; 357:85-89. 17. Zennou V, Petit C, Guetard D et al. HIV-1 genome nuclear import is mediated by a central DNA flap. Cell 2000; 101:173-185. 18. Temin HM. Retrovirus variation and reverse transcription: abnormal strand transfers result in retrovirus genetic variation. Proc Natl Acad Sci USA 1993; 90:6900-6903. 19. Pathak VK, Hu W-S. “Might as well jump!” Template switching by retroviral reverse transcriptase, defective genome formation, and recombination. Semin Virol 1997; 8:141-150. 20. Mansky LM, Temin HM. Lower in vivo mutation rate of human immunodeficiency virus type 1 than that predicted from the fidelity of purified reverse transcriptase. J Virol 1995; 69:5087-5094. 21. Bukrinsky MI, Haggerty S, Dempsey MP et al. A nuclear localization signal within HIV-1 matrix protein that governs infection of non-dividing cells. Nature 1993; 365:666-669. 22. Freed EO. HIV-1 Gag proteins: diverse functions in the virus life cycle. Virology 1998; 251:1-15. 23. Gallay P, Hope T, Chin D, Trono D. HIV-1 infection of nondividing cells through the recognition of integrase by the importin/karyopherin pathway. Proc Natl Acad Sci USA 1997; 94:98259830. 24. Brown PO. Integration. In: J. M. Coffin, Varmus HE, eds. Retroviruses. Cold Spring Harbor: Cold Spring Harbor Laboratory Press, 1997:161-203. 25. Chen H, Wei SQ, Engelman A. Multiple integrase functions are required to form the native structure of the human immunodeficiency virus type I intasome. J Biol Chem 1999; 274:17358-17364. 26. Wei SQ, Mizuuchi K, Craigie R. A large nucleoprotein assembly at the ends of the viral DNA mediates retroviral DNA integration. EMBO J 1997; 16:7511-7520. 27. Farnet CM, Bushman FD. HIV cDNA integration: molecular biology and inhibitor development. Aids 1996; 10:S3-11. 28. Ross EK, Buckler-White AJ, Rabson AB et al. Contribution of NF-kB and Sp1 binding motifs to the replicative capacity of human immunodeficiency virus type 1: Distinct patterns of viral growth are determined by T-cell types. J Virol 1991; 65:4350-4358. 29. Dayton AI, Sodroski JG, Rosen CA et al. The trans-activator gene of the human T cell lymphotropic virus type III is required for replication. Cell 1986; 44:941-947. 30. Fisher AG, Feinberg MB, Josephs SF et al. The trans-activator gene of HTLV-III is essential for virus replication. Nature 1986; 320:367-371. 31. Berkhout B, Silverman RH, Jeang K-T. Tat trans-activates the human immunodeficiency virus through a nascent RNA target. Cell 1989; 59:273-282. 32. Wei P, Garber ME, Fang SM et al. A novel CDK9-associated C-type cyclin interacts directly with HIV-1 Tat and mediates its high-affinity, loop-specific binding to TAR RNA. Cell 1998; 92:451-462. 33. Purcell DF, Martin MA. Alternative splicing of human immunodeficiency virus type 1 mRNA modulates viral protein expression, replication, and infectivity. J Virol 1993; 67:6365-6378. 34. Pollard VW, Malim MH. The HIV-1 Rev protein. Annu Rev Microbiol 1998; 52:491-532. 35. Maldarelli F, Martin MA, Strebel K. Identification of posttranscriptionally active inhibitory sequences in human immunodeficiency virus type 1 RNA: Novel level of gene regulation. J Virol 1991; 65:5732-5743. 36. Rosen CA, Terwilliger E, Dayton A, Sodroski JG, Haseltine WA. Intragenic cis-acting art generesponsive sequences of the human immunodeficiency virus. Proc Natl Acad Sci USA 1988; 85:2071-2075. 37. Bray M, Prasad S, Dubay JW et al. A small element from the Mason-Pfizer monkey virus genome makes human immunodeficiency virus type 1 expression and replication Rev-independent. Proc Natl Acad Sci USA 1994; 91:1256-1260. 38. Swanstrom R, Wills JW. Synthesis, assembly, and processing of viral proteins. In: Coffin H, Varmus HE, eds. Retroviruses. Cold Spring Harbor: Cold Spring Harbor Laboratory Press, 1997:263-334. 39. Turner BG, Summers MF. Structural biology of HIV. J Mol Biol 1999; 285:1-32. 40. Zhou W, Resh MD. Differential membrane binding of the human immunodeficiency virus type 1 matrix protein. J Virol 1996; 70:8540-8548.

32

Lentiviral Vector Systems for Gene Transfer

41. Facke M, Janetzko A, Shoeman RL et al. A large deletion in the matrix domain of the human immunodeficiency virus gag gene redirects virus particle assembly from the plasma membrane to the endoplasmic reticulum. J Virol 1993; 67:4972-4980. 42. Freed EO, Orenstein JM, Buckler-White AJ et al. Single amino acid changes in the human immunodeficiency virus type 1 matrix protein block virus particle production. J Virol 1994; 68:5311-5320. 43. Berkowitz R, Fisher J, Goff, SP. RNA packaging. Curr Top Microbiol Immunol 1996; 214:177-218 44. Rein A, Henderson LE, Levin JG. Nucleic-acid-chaperone activity of retroviral nucleocapsid proteins: Significance for viral replication. Trends Biochem Sci 1998; 23:297-301. 45. Kaye JF, Lever AM. Nonreciprocal packaging of human immunodeficiency virus type 1 and type 2 RNA: A possible role for the p2 domain of Gag in RNA encapsidation. J Virol 1998; 72:5877-5885. 46. Campbell S, Vogt VM. Self-assembly in vitro of purified CA-NC proteins from Rous sarcoma virus and human immunodeficiency virus type 1. J Virol 1995; 69:6487-6497. 47. Freed EO, Martin MA. The role of human immunodeficiency virus type 1 envelope glycoproteins in virus infection. J Biol Chem 1995; 270:23883-23886. 48. Murakami T, Freed EO. Genetic evidence for an interaction between human immunodeficiency virus type 1 matrix and alpha-helix 2 of the gp41 cytoplasmic tail. J Virol 2000; 74:3548-3554. 49. Murakami T, Freed EO. The long cytoplasmic tail of gp41 is required in a cell type-dependent manner for HIV-1 envelope glycoprotein incorporation into virions. Proc Natl Acad Sci USA 2000; 97:343-348. 50. Gottlinger HG, Dorfman T, Sodroski JG et al. Effect of mutations affecting the p6 gag protein on human immunodeficiency virus particle release. Proc Natl Acad Sci USA 1991; 88:3195-3199. 51. Huang M, Orenstein JM, Martin MA et al. p6Gag is required for particle production from fulllength human immunodeficiency virus type 1 molecular clones expressing protease. J Virol 1995; 69:6810-6818. 52. Garnier L, Wills JW, Verderame MF et al. WW domains and retrovirus budding. Nature 1996; 381:744-745. 53. Vogt VM. Ubiquitin in retrovirus assembly: Actor or bystander? Proc Natl Acad Sci USA 2000; 97:12945-12947. 54. Wlodawer A, Erickson JW. Structure-based inhibitors of HIV-1 protease. Annu Rev Biochem 1993; 62:543-585. 55. Yeager M, Wilson-Kubalek EM, Weiner SG et al. Supramolecular organization of immature and mature murine leukemia virus revealed by electron cryo-microscopy: Implications for retroviral assembly mechanisms. Proc Natl Acad Sci U S A 1998; 95:7299-7304. 56. Ganser BK, Li S, Klishko VY, Finch JT, Sundquist WI. Assembly and analysis of conical models for the HIV-1 core. Science 1999; 283:80-83. 57. Franke EK, Yuan HE, Luban J. Specific incorporation of cyclophilin A into HIV-1 virions. Nature 1994; 372:359-362. 58. Luban J, Bossolt KL, Franke EK et al. Human immunodeficiency virus type 1 Gag protein binds to cyclophilins A and B. Cell 1993; 73:1067-1078. 59. Grattinger M, Hohenberg H, Thomas D et al. In vitro assembly properties of wild-type and cyclophilin-binding defective human immunodeficiency virus capsid proteins in the presence and absence of cyclophilin A. Virology 1999; 257:247-260. 60. Cullen BR. HIV-1 auxiliary proteins: Making connections in a dying cell. Cell 1998; 93:685-692. 61. Strebel K, Klimkait T, Martin MA. A novel gene of HIV-1, vpu, and its 16-kilodalton product. Science 1988; 241:1221-1223. 62. Gottlinger HG, Dorfman T, Cohen EA et al. Vpu protein of human immunodeficiency virus type 1 enhances the release of capsids produced by gag gene constructs of widely divergent retroviruses. Proc Natl Acad Sci USA 1993; 90:7381-7385. 63. Margottin F, Bour SP, Durand H et al. A novel human WD protein, h-beta TrCp, that interacts with HIV-1 Vpu connects CD4 to the ER degradation pathway through an F-box motif. Mol Cell 1998; 1:565-574. 64. He J, Choe S, Walker R, Di Marzio P, Morgan DO, Landau NR. Human immunodeficiency virus type 1 viral protein R (Vpr) arrests cells in the G2 phase of the cell cycle by inhibiting p34cdc2 activity. J Virol 1995; 69:6705-6711.

HIV-1 Replication

33

65. Simon JH, Gaddis NC, Fouchier RA et al. Evidence for a newly discovered cellular anti-HIV-1 phenotype. Nat Med 1998; 4:1397-1400. 66. Aiken C, Konner J, Landau NR et al. Nef induces CD4 endocytosis: Requirement for a critical dileucine motif in the membrane-proximal CD4 cytoplasmic domain. Cell 1994; 76:853-864. 67. Collins KL, Chen BK, Kalams SA et al. HIV-1 Nef protein protects infected primary cells against killing by cytotoxic T lymphocytes. Nature 1998; 391:397-401. 68. Kestler HW, III, Ringler DJ, Mori K et al. Importance of the nef gene for maintenance of high virus loads and for development of AIDS. Cell 1991; 65:651-62. 69. Marsh JW. The numerous effector functions of Nef. Arch Biochem Biophys 1999; 365:192-198. 70. Fletcher TM, 3rd, Brichacek B, Sharova N et al. Nuclear import and cell cycle arrest functions of the HIV-1 Vpr protein are encoded by two separate genes in HIV-2/SIV(SM). Embo J 1996; 15:6155-6165.

CHAPTER 3

Determinants for Lentiviral Infection of Non-Dividing Cells Marie A. Vodicka

Abstract

L

entiviruses share the common characteristic of infecting non-dividing target cells, distinguishing them from the oncogenic retroviruses which only productively infect dividing cells. The search for determinants for infection of non-dividing cells has produced a number of candidates. From HIV-1, the viral proteins matrix, integrase and Vpr have all been implicated. A structural determinant, the central DNA flap, has also been implicated. The supporting evidence for each of these proposed determinants will be examined and compared to how other viruses, non-retroviruses, transport their genomes to the nucleus. With currently available data, integrase and the central DNA flap appear to be the key players, and yet the mechanism for infection of non-dividing cells remains undefined.

Introduction Because retroviruses require integration into the host cell genome to complete the viral life cycle, the virus must have a mechanism for getting its genome in proximity to the host cell chromosomes. The oncogenic retroviruses require mitosis for integration, presumably because this then allows them access to host cell chromosomes.1,2 However, one of the distinguishing features of lentiviruses is the ability to infect non-dividing cells.3 Why lentiviruses share this characteristic and how they are able to infect non-dividing cells is not well understood. Macrophages, or cells of the monocyte lineage, are major in vivo targets of all lentiviruses. Macrophages are terminally differentiated cells that are in a G0, or non-proliferating, state. From a teleological perspective, lentiviruses either have the capacity to infect non-dividing cells, and so macrophages became a major target; or because the lentiviruses need to infect macrophages to complete their life cycles, they evolved mechanisms to infect non-dividing cells. Interestingly, the primary targets for the primate lentiviruses (HIV and SIV) and FIV are activated T-cells, a rapidly dividing cell type. Yet infection of non-dividing cells is important for the life cycle and pathogenesis of these viruses. For HIV, non-dividing target cells include macrophages which have been implicated in transmission of virus between hosts, and resting T cells which have been implicated in transmission and in persistence of a viral reservoir.4 Whether proliferating or not, cells must be metabolically active; completely quiescent cells cannot complete reverse transcription and are not productively infected by HIV.5 Thus even the lentiviruses with expanded cell tropism depend on infection of non-dividing cells. The proposed determinants for HIV infection of non-dividing cells, examining the evidence for each determinant and pathway and referring to relevance of HIV results for other Lentiviral Vector Systems for Gene Transfer, edited by Gary L. Buchschacher, Jr. ©2003 Eurekah.com.

36

Lentiviral Vector Systems for Gene Transfer

lentiviruses will be presented. This list includes the viral structural protein—matrix (MA), a viral enzymatic protein—integrase (IN), a viral accessory protein—Vpr, a viral cis-acting DNA sequence—the central polypurine track (cPPT), and the host cell cytoskeleton.

Nuclear Transport Before discussing how viruses reach the nucleus, a brief overview of cellular nuclear transport will be given (Fig. 1). Bi-directional transport of molecules between the nucleus and cytoplasm of eukaryotic cells is controlled by large (approximately 125 MD), multi-protein complexes called nuclear pore complexes (NPC) that are found throughout the double layered nuclear envelope which maintain the separation of these compartments.6 The aqueous channels permit the diffusion of small molecules into and out of the nucleus, but larger molecules, usually anything greater than 50 KD or larger in diameter than 9 nm, require signal-mediated transport.7,8 However, nuclear molecules smaller than the theoretical exclusion size usually contain nuclear import signals. Nuclear import and export of proteins and RNAs are mediated by a family of nuclear transport proteins, generally termed importins (or karyopherins) and exportins (reviewed in ref. 8).9 In general, nuclear proteins contain a nuclear localization sequence (NLS) that is recognized, either directly by a member of the importin β/transportin family of transporters, or indirectly through an adapter protein which subsequently binds to the importin β for cargo transport. The most studied of these adapters is importin α, which mediates protein import via the “classical pathway”. Importin α binds to a single or bipartite stretch of basic amino acids that functions as a NLS. However, other adapter proteins have been identified which recognize different NLSs. Nuclear export works in a similar way, but with different transporter and adapter proteins.8,10 These pathways often involve the transport of RNA or RNPs from the nucleus to the cytoplasm. As for nuclear import, different nuclear export substrates use different nuclear export signals (NES) and transportins, but all pathways converge at the nuclear pore. The best characterized nuclear exporter is CRM1, which mediates nuclear export of leucine-rich NES cargo proteins by direct binding to the NES.11,12 Finally, there are some nucleocytoplasmic shuttling proteins, best exemplified by the M9 signal of hnRNP A1,13 containing overlapping import and export signals which often cannot be separated, and so are termed nuclear shuttling (NS) signals.14 Directionality of nucleocytoplasmic transport is thought to be controlled by a gradient of GTP and GDP bound forms of the GTPase, Ran.15-17 The Ran GTPase-activating protein (RanGAP) and its activators are in high concentrations in the cytoplasm while the Ran guanine nucleotide exchange factor (RanGEF) is at high concentrations in the nucleus, leading to the prediction that cytoplasmic Ran is mostly GDP loaded and nuclear Ran is mostly GTP loaded. For nuclear import, RanGTP promotes dissociation of importin β and its cargo on the nuclear side while nuclear export factors bind cargo at higher affinity in the presence of RanGTP (nuclear) and release cargo in the presence of RanGDP (cytoplasmic). This model fits currently available data to explain directionality of transport, but the mechanism for actual transit through the pore is still not understood.18

Transit of Non-Retroviral Particles One of the issues confounding analysis of retroviral infection and integration is our paucity of knowledge about viral particle composition and structure after it enters the cytoplasm of the target cell. Although the molecular details of reverse transcription and integration are well characterized (see Chapter 2), the transit of the virus core and viral genome from the site of entry at the plasma membrane to the site of integration in the host chromosomes is poorly understood for retroviruses. However, a number of diverse viruses must deliver their genomes to the nucleus to replicate, and some of these pathways are better understood (Table 1).

37

Determinants for Lentiviral Infection of Non-Dividing Cells

Table 1. Delivery of viral genomes to nucleus Virus

Genome

Structure

Transport (envelope/capsid size)

Nuclear Entry to Nucleus

Adenovirus

dsDNA

non-enveloped 90 nm

capsid in endosome & direct contact with microtubule motors; capsid disassembly during transport

capsid binds NPC, injects DNA through pore

SV40

dsDNA

non-enveloped 50 nm

capsid in endosome & soluble import factors

through NPC after capsid conformational changes

Herpes Virus dsDNA

enveloped 100 nm

capsid direct transport on microtubules by dynein

through NPC; unknown mechanism

Influenza

minusstrand RNA, segmented

enveloped 20X100 nm long rods

RNA-RNP complexes with soluble import

classical import via NPC factors

Oncogenic Retovirus

RNA, enveloped capsid 50X with DNA 100 nm cones; intermediate PIC 28 nm

PIC in cytoplasm, unknown mechanism

at mitosis; unknown mechanism

Lentivirus (HIV)

RNA, enveloped capsid 50X with DNA 100 nm cones; intermediate PIC 28 nm

PIC in cytoplasm, possible actin & microtubule cytoskeleton; soluble import factors

through NPC; mechanism unknown

Many DNA viruses transport their genomes to the intact nucleus to take advantage of host cell DNA replication factors. Some RNA viruses also transport their genomes to the nucleus, not for RNA synthesis, but to make use of host cell splicing factors. The mechanisms for transport are as varied as the viruses, but all pathways converge at the nuclear pore for entry.19 Transport of DNA is not a normal cellular function, and so viruses must adapt the normal cellular nuclear import pathway to deliver their genomes. Viral genome complexes and capsids are too large to diffuse through the cytoplasm,20 and so many of them enlist host cell cytoskeletal transport systems, particularly microtubules,21 in addition to the soluble nuclear import factors.22 Two DNA tumor viruses, SV40 and adenovirus, use different mechanisms for delivering their genomes to the nucleus. However, both remain as intact capsids throughout much of their cytoplasmic transit. The non-enveloped SV40 virion enters the cell by endocytosis and is transported through the cytoplasm in the endosome. Near the nucleus, the viral particle, having adopted a modified conformation, exits the endosome. In this partially dissociated form composed of viral structural proteins, the still recognizable viral particle enters the nucleus through the NPC, facilitated by its NLS containing proteins.23 In contrast, adenovirus capsids travel on microtubules, initially within an endocytic vesicle, but later, by direct interaction with host cell microtubule motors.24,25 The adenovirus capsid docks at the NPC and then

38

Lentiviral Vector Systems for Gene Transfer

Fig. 1. Nuclear import in the cell. The major nuclear import factors and pathways are represented schematically, beginning with NLS-substrates binding to soluble nuclear import factors (the adapter, importin α and the transporter importin β or transportin) in the cytoplasm. These complexes are docked at the nuclear pore and then translocated through the pore into the nucleus where they dissociate when RanGDP is converted to RanGTP. Import factors are recycled to the cytoplasm via nuclear export pathways and NLS substrates remain in the nucleus. The directionality is likely controlled by higher concentrations of RanGDP in the cytoplasm and higher concentrations of RanGTP in the nucleus, maintained by the Ran GTPase activating protein (RanGAP) and the Ran guanine nucleotide exchange factor (RanGEF), respectively.

dissociates as its genome is injected through the pore into the nucleoplasm.22,26 Interestingly, the dissociation of the adenovirus genome from its capsid occurs in a stepwise fashion, beginning immediately upon endocytic uptake of the virion as it enters the cell, and continuing until the genome is completely dissociated from the capsid to enter the nucleus.24 Herpes virus capsid enters the cytoplasm via direct fusion between the viral envelope and the cell plasma membrane. The released capsid is transported by dynein along microtubules to the NPC where the capsid is left and the genome with associated viral proteins enters the nucleus.27 The RNA genome of influenza virus enters the nucleus as part of ribonucleoprotein (RNP) complexes, making use of the cellular nuclear import pathways by NLSs in the protein component of the RNPs.28, 29 Thus delivery of a viral payload to the nucleus is not unique to retroviruses, but this part of the life cycle has been uniquely difficult to study for retroviruses. The capsids of the viruses described above remain largely intact, or at least as recognizable structures, throughout much of this process. Retroviral particles disassemble, and the resulting complexes for reverse transcription and integration are difficult to visualize and remain poorly defined throughout the cytoplasmic transport and nuclear entry.

Determinants for Lentiviral Infection of Non-Dividing Cells

39

Fig.2. HIV and MLV preintegration complexes and nuclear entry. Structural (black circles) and enzymatic proteins (white circles) are part of the viral core, and some remain associated with the PIC after viral entry, uncoating and reverse transcription. The structure of the PIC is not well defined, but is known to differ in composition between the lentivirus HIV (which has a partially triple stranded genome and carries the accessory protein Vpr) and the oncogenic retrovirus, MLV. HIV infects in the absence of mitosis and contains several potential determinants for nuclear entry, listed right, but MLV requires mitosis for integration and productive infection. Note, DNA flap drawn larger, out of scale, for emphasis. RT, reverse transcriptase; NC, nucleocapsid; CA, capsid; MA, matrix; IN, integrase.

Retroviral Preintegration Complexes Post-entry retroviral particles first form a reverse transcription complex (RTC), and upon completion of viral DNA synthesis from the RNA template (see Chapter 2 for details), the viral genome complex is termed a preintegration complex (PIC). Studied mostly from MLV and HIV, the strict functional definition of the PIC is to be integration competent in any of a number of in vitro integration assays. The structure and components of the retroviral PIC remain elusive. However, analysis of integration competent complexes from infected cells has provided some information on the size (estimated Stokes radius of 28 nm)30 and composition of the PIC (Fig. 2). In contrast to the viruses described above, neither capsid or nucleocapsid were found associated with the viral genome. But the matrix protein (MA), Vpr, integrase (IN), and reverse transcriptase (RT) have all been found in theses complexes,31 in addition to several cellular proteins, most notably HMGI/Y and BAF.32,33 Failure to detect a particular protein in the complex is not proof that it is absent, however, because it may be due to sensitivity of detection or loss of components during isolation. HMG and BAF proteins appear to facilitate integration itself and have not been posited for importance in transport of the complex or for infection of non-dividing cells.33-35

40

Lentiviral Vector Systems for Gene Transfer

Assays for Infection of Non-Dividing Cells HIV infection of non-dividing cells is assayed both directly and indirectly. Productive, spreading infection in a non-dividing cell type, particularly macrophages, and other terminally differentiated cells, such as neurons, is the best demonstration of infection of non-dividing cells. Failure to replicate in macrophages, however, must be contrasted with replication in a dividing cell type, usually T-cells, to demonstrate that the virus being assayed is not completely replication defective. Alternatively, single cycle assays are also used, often comparing infection of artificially growth-arrested cell cultures to their unmanipulated, proliferating counterparts. Indicator cell lines that activate a reporter gene upon expression of newly synthesized viral proteins (MAGI, GHOST)36,37 are most commonly used for these experiments, but HIV or other lentiviral derived vectors carrying reporter genes can be assayed in diverse cell types, limited only by envelope tropism. Nuclear entry of viral genomes, although not a measure of virus infectivity, is often used as an assay read-out because it is a prerequisite for viral integration. Detection of 2 LTR circles, circularized viral genomic DNA found only in the nucleus,38 by a variety of methods (PCR, Southern blot), is the most widely used of these assays. Thus, although the 2 LTR circle is a dead-end product in the viral life cycle, it is an indication that reverse transcription is complete and that the viral genome has entered the nucleus. More recently, assays to detect bona fide integration have also been developed, using Alu-PCR, a nested PCR method with primers to the LTR and to repetitive Alu-elements found throughout the human genome.39

Viral Protein Determinants for HIV-1 Infection of Non-Dividing Cells HIV MA was the first protein examined for its effects on infection of non-dividing cells. A stretch of basic residues in the amino terminus of MA (25GKKKYKLKH) was tested in a classic assay for NLS function;40 microinjection of BSA coupled to a synthetic peptide containing (25GKKKYKLKH) resulted in nuclear uptake of the conjugate.41 Substitution of the first two arginines with threonines (25GTTKYKLKH) to disrupt NLS function inhibited infection and nuclear accumulation of viral genomes (2 LTR circles) in growth arrested T-cells while leaving infectivity and nuclear entry unperturbed in proliferating T-cells.41 Further experiments demonstrated that mutations in the MA-NLS which knock-out the nuclear import function dramatically decreased HIV infection of macrophage but not dividing T-cell cultures.42-44 The hypothesis that MA promotes nuclear entry of the HIV genome complex by using the classical nuclear import pathway of the cell was supported by additional experiments. MA-GST fusions enter the nucleus in microinjection experiments.45 MA fails to enter the nucleus when the importin α/β pathway is disrupted, and manipulations interfering with MA nuclear entry also block HIV infection of macrophages.46 MA is myristoylated at its N-terminus to target the Gag polyprotein precursor to the plasma membrane during virion assembly (see Chapter 2). This would seem to contradict its NLS, and thus it was proposed that a Cterminal phosphorylation of MA regulates MA localization.44 However the role of phosphorylation in regulation of MA localization and nuclear import of the PIC remains controversial.4749 In fact, the role of N-terminal basic region of MA for promoting nuclear entry of the PIC has been called into question.50-52 Subsequent researchers were unable to demonstrate an autonomous NLS function for the proposed MA-NLS, nor were they able to demonstrate nuclear import of MA as part of a fusion protein, and the “NLS mutant” phenotype of MA was attributed to gag processing defects. Yet some researchers, having demonstrated a second MA NLS, contend that the karyophilic properties of MA are necessary and sufficient for the nuclear import of the HIV PIC and infection of non-dividing cells.53 The importance of MA is further

Determinants for Lentiviral Infection of Non-Dividing Cells

41

Fig. 3. Lentivirus genomes. Note that the accessory genes, vpr/vpx are found only in primate lentiviruses and that vif is found in all lentiviruses except EIAV (Modified from ref. 3).

42

Lentiviral Vector Systems for Gene Transfer

confounded by experiments demonstrating that MA is dispensable for HIV infection in some conditions.54 Vpr was the next HIV protein examined for its effect on infection of non-dividing cells. It had been noted that Vpr mutations decreased HIV-1 infection of macrophages.55-57 Subsequent experiments attributed this to Vpr nuclear targeting of the PIC in a partially redundant fashion with MA.43 Deletion of Vpr decreased transport of the viral genome to the nucleus (2 LTR circles) and decreased infection of macrophages.43 Again, the attenuation in infection was specific to non-proliferating cells without affecting infection of proliferating cells. Significantly, viruses with mutations deleting Vpr and disrupting the NLS of MA had a more severe phenotype than the single mutants, with a decrease in infection of macrophages and growth arrested cells. However no conventional NLS is detectable in Vpr, and Vpr mediated nuclear import was not disrupted by interference with the importin pathway.46 Vpr binds to importin α, not as an NLS substrate but through a different site on importin α from its NLS binding site.58 Additionally Vpr binds to nucleoporins, and it has been proposed that this facilitates docking of the PIC to the nuclear pore for viral genome entry.59-61 Nuclear pore targeting was shown to be necessary for Vpr to positively affect HIV infection of macrophages. Two independent signals within Vpr, one in the amino half and the other in the carboxy half of the protein, have been implicated in Vpr nuclear import,62-64 and these seem to function in the absence of additional soluble factors, usually required for nuclear import.65 Vpr has also been shown to weakly enhance nuclear uptake of NLS substrates.66 For HIV-2, the closely related protein, Vpx, is posited to perform a similar function to HIV-1 Vpr in terms of promoting macrophage infection.67,68 Vpr (or Vpx in HIV-2) is an attractive candidate as a determinant for infection of non-dividing cells on the one hand because it is part of the PIC and has clearly demonstrated nuclear import or NPC docking ability. And yet, it is a poor candidate on the other hand because it is only present in primate lentiviruses (Fig. 3). The non-primate lentiviruses infect non-dividing cells without a Vpr. Viruses lacking Vpr and MA-NLS still replicate, at reduced levels, in non-dividing cells, and so another determinant was sought. The third HIV protein implicated in nuclear entry of the PIC is the viral integrase (IN), whose enzymatic function is essential for integration of the double-stranded DNA viral genome into the chromosomes of the infected cell to form the provirus. IN also displays nuclear import function when assayed by microinjection, but unlike Vpr and like MA, IN nuclear import is blocked when the importin α/β pathway is disrupted.45 However mutations in IN to inactivate the putative NLS are replication defective—apparently integration defective, and therefore it was impossible to separate the nuclear import properties of IN from its integration function using this mutant. Recently this IN NLS (aa 211-219) has been disputed for its importance in infection of non-dividing cells and even for its NLS function. However the nuclear import abilities of IN first assayed by microinjection of GST-IN45 fusion proteins have been confirmed using green fluorescent protein-IN fusion proteins.69,70 By using truncations of these fusion proteins, another region of IN with NLS function has been identified by Malim and colleagues.71 This sequence (aa 161-173), IIGQVRDQAEHLK, does not resemble a classical NLS, but addition of this sequence confers nuclear import to a heterologous substrate. Mutagenesis of the IN NLS prevents nuclear accumulation of INfusion protein. In a viral context, the IN NLS mutant is replication defective both for dividing and non-dividing cells. However, the authors demonstrate the primary defect is nuclear entry because the enzyme is integration competent in vitro and viral infection can be complemented by addition of IN with a mutation in the catalytic site and a wild-type NLS. From this the authors infer that since IN forms dimers, one subunit provides the catalytic activity and the other provides the NLS function, thus demonstrating that IN NLS is required for all HIV infection. Based on results primarily from HIV-1 studies, IN appears the most likely candidate for a viral protein determinant for infection of non-dividing cells. However, the transfer of any of

Determinants for Lentiviral Infection of Non-Dividing Cells

43

Fig. 4. Involvement of cytoskeleton during early HIV infection. Many viruses (Table 1) depend upon host cell cytoskeletal transport systems for delivery of the viral genome to the nucleus. This possibility has only recently been investigated for HIV, and this diagram remains somewhat speculative.

these three proteins (MA, Vpr, IN) to MLV does not enable MLV to infect non-dividing cells (A. Perez, unpublished data).72 The role of MA is in dispute. Vpr has nuclear import properties, but it is not part of non-primate lentiviruses. Although Vpr may have a facilitory function, it is unlikely to be essential for HIV infection of non-dividing cells. Nuclear import functions of Vpr and MA may be important for a different stage of the life cycle because both proteins have been identified as nucleocytoplasmic shuttling proteins.63,73 IN has a demonstrable NLS, the function of which appears important for HIV infection. However, it appears essential for infection of dividing as well as non-dividing cells. How this might change our thinking about determinants for infection of non-dividing cells is discussed further in the next sections.

Viral Genome Structural Determinants The newest development in understanding HIV infection of non-dividing cells is not a protein at all, but a structural component of the HIV genome.74 All lentiviruses, including HIV, contain a central polypurine track (cPPT) that enables initiation of second plus-strand synthesis (see Chapter 2). 75,76 This results in what has been termed a central DNA flap (see Fig. 5). Recent experiments demonstrated that mutations in the HIV genome which prevent formation of this flap are replication defective. Even minor alterations to the cPPT affecting flap formation result in severe attenuation of infectivity. The authors attribute the defect to a failure of nuclear entry of the genome. This is because virus production is not affected, reverse transcription products form in the appropriate time and amount, and isolated PICs can perform in vitro integration. Thus, the defect appears to be between completion of reverse transcription and integration and, therefore, at the point of nuclear entry. Further supporting the importance of

44

Lentiviral Vector Systems for Gene Transfer

Fig. 5. Models for lentivirus (HIV) nuclear transport and entry. These models are conjectural based upon currently available data. A) Lentiviruses may require transport of the PIC through the cytoplasm to the nucleus, making use of host cell microtubule transport. Microtubule motors would perform the actual translocation, but it is unclear if motors would interact directly with viral components of the PIC or through a cellular protein. B) IN is the best viral protein candidate for HIV nuclear entry. It may facilitate viral genome entry through proposed classical NLSs or through more recently proposed functional nuclear import signal that likely operates by binding to another protein with a NLS. This other protein, designated “X”, would interact with the cellular nuclear import factors, such as importin β. Protein X could be a viral protein, such as MA, or more likely, an unidentified cellular protein. C) The DNA flap produced from the cPPT is a structural element required for HIV infection, appearing to be important for nuclear entry. D) All of the above proposed determinants may work together to promote nuclear entry of the PIC, enabling infection of non-dividing cells. In this view, the complex may be transported to the nucleus on microtubules, where soluble nuclear import factors and viral proteins, such as Vpr, dock it to the NPC. Nuclear import signals in viral and cellular proteins associated with the complex and the genome structure promote translocation of the complex through the pore, into the nucleus where it is able to integrate into the host cell genome. Note, DNA flap drawn larger, out of scale, for emphasis. PIC, preintegration complex; IN, integrase; β, importin β; NPC, nuclear pore complex.

Determinants for Lentiviral Infection of Non-Dividing Cells

45

this structure for infection of non-dividing cells is that when transferred to a HIV vector, it enhances transduction in non-dividing cells by about 10 fold.77,78 However there are problems with interpreting this central DNA flap as the determinant for nuclear entry of the genome and infection of non-dividing cells. Mutations in the cPPT affect infection of dividing as well as non-dividing cells. Therefore the defect is not specific to non-dividing cells. The authors point out that there is no direct evidence for a mitosis-dependent mechanism for lentiviral genome nuclear entry and that a failure at nuclear entry may affect dividing as well as non-dividing cells;74 this is similar to the phenotype of the newly described IN-NLS mutant above.71 Also, it is unclear if this cDNA flap is a stable structure that is preserved up to integration.79 Additionally, the cPPT is not absolutely required for infection of non-dividing cells because HIV vectors that lack the cPPT still transduce non-dividing cells.80 Perhaps the most striking feature of these studies is that they raise the possibility of a structural determinant for infection of non-dividing cells—a structure found in lentiviruses, but not in oncogenic retroviruses.

Cellular Transport Pathways and the Cytoskeleton The previous sections have focused on lentiviral nuclear entry, but not on transport through the cytoplasm. However, above it was discussed how other viruses use nuclear transport factors, and especially microtubule transport, to reach the nucleus. Unless retroviruses are significantly different from other viruses, it seems likely that the cytoskeleton plays a role in transport of retroviral particles through the cytoplasm to the nucleus (Table 1; Fig. 4). Actin is present in the HIV virion, and the integrity of the actin cytoskeleton is thought to be important for virion assembly.81,82 In analogy with entry of other viruses, integrity of the cortical actin cytoskeleton is likely important for HIV entry at the plasma membrane.21 An intact actin cytoskeleton may also be necessary for completion of reverse transcription,83 but microfilament disruption in this experiment cannot distinguish between effects at entry and early post-entry. There is also recent evidence for involvement of microtubules in HIV transport to the nucleus (Vodicka, Hope, McDonald, unpublished data). Thus HIV cytoplasmic transport may rely on microtubules as do adenovirus, herpes virus and SV40. It is unclear if an HIV protein in the PIC would bind directly to microtubule motors or if this interaction would be mediated by a cellular protein. However, considering that the lentiviral PIC is too big to diffuse through the cytoplasm and that diverse viruses use the host cell cytoskeleton, especially microtubules, to deliver their genomes to the nucleus, it is likely that lentiviral infection will also rely on cellular cytoskeletal transport (Fig. 4).

Summary and Conclusions Unfortunately, the current data leave us with no clear answer for a single, or multiple, determinant for lentiviral infection of non-dividing cells (Fig. 5). In this discussion, and in the field, it has been assumed that the mechanism is shared between all lentiviruses, but it should at least be considered that different lentiviruses may infect non-dividing cells by different mechanisms. However, embracing that assumption, the most conserved characteristics between lentiviruses that have been implicated as determinants for infecting non-dividing cells are the IN and cPPT. The recent demonstration of the importance of HIV-1 IN NLS and DNA flap may lead us to approach the question a little differently. Perhaps, as suggested by Charneau’s and Malim’s groups, lentiviruses always go through the NPC to get inside the nucleus even in dividing cells, and the previous experimental distinction between infection of dividing and non-dividing cells is misleading when searching for a determinant for infection of non-dividing cells. If so, what makes lentiviruses different from oncogenic retroviruses? There is evidence that avian sarcoma virus IN contains a functional NLS,84 and yet this oncogenic retrovirus does not infect nondividing cells. Perhaps we should look to a structural determinant, such as the DNA flap. Yet

46

Lentiviral Vector Systems for Gene Transfer

even this is not absolutely required to infect non-dividing cells. It is striking that neither HIV1 IN or the cPPT transferred to MLV confers ability to infect non-dividing cells. Most studies on lentiviral infection of non-dividing cells have focused on nuclear entry, but perhaps nuclear entry per se is not the determining factor. This suggests that nuclear entry may be necessary for infection of non-dividing cells, but some other determinant(s) may govern this process. Perhaps in analogy with infection of other viruses, the infection needs to be viewed as a whole—viral entry, uncoating and reverse transcription, transport of the PIC, nuclear entry of the genome, and integration—with each step a necessary precursor to the next. In this whole process, a hierarchy of necessary steps, lies the difference between lentiviruses and oncogenic retroviruses, and while disruption of a single step will block lentiviral infection, transfer of a single determinant will not enable oncogenic retroviral infection of non-dividing cells. The gaps between entry at the plasma membrane and nuclear entry, and between nuclear entry and integration may hold the keys to our understanding lentiviral infection of nondividing cells.

References 1. Lewis PF, Emerman M. Passage through mitosis is required for oncoretroviruses but not for the human immunodeficiency virus. J Virol 1994; 68:510-516. 2. Roe T, Reynolds TC, Yu G et al. Integration of murine leukemia virus DNA depends on mitosis. EMBO J 1993; 12:2099-2108. 3. Joag SV, Stephens EB, Narayan O. Lentiviruses. 3rd ed. Vol. 2. Philadelphia: Lippincott-Raven, 1996. 4. Zhang Z, Schuler T, Zupancic M et al. Sexual transmision and propagation of SIV and HIV inresting and activated CD4+ T cells. Science 1999; 286:1353-1357. 5. Korin YD. Nonproductive human immunodeficiency virus type 1 infection in nucleoside-treated G0 lymphocytes. J Virol 1999; 73:6526-6532. 6. Talcott B, Moore MS. Getting across the nuclear pore complex. Trends Cell Biol 1999; 9:312-318. 7. Mataj IW, Englmeier L. Nucleocytoplsamic transport: The soluble phase. Ann Rev Biochem 1998; 67:265-306. 8. Nakielny S, Dreyfuss G. Transport of proteins and RNAs in and out of the nucleus. Cell 1999; 99:677-690. 9. Adam SA, Gerace L. Cytosolic proteins that specifically bind nuclear location signals are receptors for nuclear import. Cell 1991; 66:837-847. 10. Cullen BR. Nuclear RNA export pathways. Mol Cell Biol 2000; 20:4181-4187. 11. Wen W, Meinkoth JL, Tsien RY et al. Identificaion of a signal for rapid export of proteins from the nucleus. Cell 1995; 82:463-473. 12. Fukuda M, Asano S, Nakamura T et al. CRM1 is repsonsible for intracellular transport mediated by the nuclear export signal. Nature 1997; 390:308-311. 13. Michael WM, Choi M, Dreyfuss G. A nuclear export signal in hnRNP A1: A signal-mediated, temperature-dependent nuclear protein export pathway. Cell 1995; 83:415-422. 14. Michael WM. Nucleocytoplasmic shuttling signals: Two for the price of one. Trends Cell Biol 2000; 10:46-50. 15. Cole CN, Hammell CM. Nucleocytoplasmic transport: driving and directing transport. Curr Biol 1998; 8:R368-372. 16. Moore MS, Blobel G. The GTP-binding protein Ran/TC4 is required for protein import in to the nucleus. Nature 1993; 365:661-663. 17. Görlich D, Dabrowski M, Bischoff FR et al. A novel class of RanGTP binding proteins. J Cell Biol 1997; 138:65-80. 18. Ohno M, Fornerod M, Mattaj IW. Nucleocytoplasmic transport: the last 200 nanometers. Cell 1998; 92:327-336. 19. Izaurralde E, Kann M, Pante N et al. Viruses, mircoorganisms and scientists meet the nuclear pore. EMBO J 1999; 18:289-296.

Determinants for Lentiviral Infection of Non-Dividing Cells

47

20. Luby-Phelps K. Ctyoarchitechture and physical properits of cytoplasm: volume, visocity, diffucion, intracellullar surface area. Int Rev Cytol 2000; 192:198-221. 21. Sodeik B. Mechanisms of viral transport in the cytoplasm. Trends Microbio 2000; 8:465-472. 22. Whittaker GR, Helenius A. Nuclear import and export of viruses and virus genomes. Virology 1998; 246:1-23. 23. Greber UF, Kasamatsu H. Nuclear targeting of SV40 and adenovirus. Trends Cell Biol 1996; 6:189-195. 24. Greber UF, Willetts M, Webster P et al. Stepwise dismantling of adenovirus 2 during entry into cells. Cell 1993; 75:477-486. 25. Soumalainen M, Nakano MY, Keller S et al. Microtubule-dependent plus-and minus end-directed motilities are competing for nuclear targeting of adenovirus. J Cell Biol 1999; 144:657-672. 26. Greber UF, Soumaliainen M, Stidwill RP et al. The role of the nuclear pore complex in adenovirus DNA entry. EMBO J 1997; 16:5998-6007. 27. Sodeik B, Ebersold MW, Helenius A. Microtubule-mediated transport of incoming herpes implex virus 1 capsids to the nucleus. J Cell Biol 1997; 136:1007-1021. 28. O’Neill RE, Jaskunas R, Blobel G et al. Nuclear import of influenza virus RNA can be mediated by viralnucleoproteina nd transport factors requred for protein import. J Biol Chem 1995; 270:22701-22704. 29. Whittaker G, Bui M, Helenius A. The role of nuclear import and export in influenza virus infection. Trends Cell Biol 1996; 6:67-71. 30. Miller MD, Farnet CM, Bushman FD. Human immunodeficiency virus type 1 preintegration complexes: Studies of organization and composition. J Virol 1997; 71:5382-5390. 31. Farnet CM, Haseltine WA. Determination of viral proteins present in the human immunodeficiency virus type 1 preintegration complex. J Virol 1991; 65:1910-1915. 32. Farnet CM, Bushman FD. HIV-1 cDNA integration: requirement of HMGI(Y) protein for function of preintegration complexes in vitro. Cell 1997; 88:483-492. 33. Li L, Farnet CM, Anderson WF et al. Modulation of activity of Moloney Murine Leukemia virus preintegration complexes by host factors in vitro. 1998; 72:2125-2131. 34. Chen H, Engelman A. The barrier-to-autointegration protein is a host factor for HIV type 1 integration. Proc Natl Acad Sci USA 1998; 95:15270-15274. 35. Hindmarsh P, Ridky T, Reeves R et al. HMG protein family members stimulate human immunodeficiency virus type 1 and avian sarcoma virs concerted DNA integration in vitro. J Virol 1999; 73:2994-3003. 36. Kimpton J, Emerman M. Detection of replication-competent and pseudotyped human imunodeficiency virus with a sensitive cell line on the basis of activation of an integrated betagalatosidase gene. J Virol 1992; 66:2232-2239. 37. Cecilia D, KewalRamani VN, O’Leary J et al. Neutralization profiles of primary human immunodeficiency virus type 1 isolates in the context of coreceptor usage. J Virol 1998; 72:6988-6996. 38. Brown PO, Bowerman B, Varmus HE et al. Correct integration of retroviral DNA in vitro. Cell 1987; 49:349-365. 39. Courcoul M, Patience C, Rey F et al. Peripheral blood mononuclear cells produce normal amounts of defective Vif-human immunodeficiency virus type 1 particles which are restricted for the preretrotranscription steps. J Virol 1995; 69:2068-2074. 40. Kalderon D, Roberts BL, Richardson WD et al. A short amino acid sequence able to specify nuclear localization. Cell 1984; 39:499-509. 41. Bukrinsky MI, Haggerty S, Dempsey MP et al. A nuclear localization signal within HIV-1 matrix protein that governs infection of non-dividing cells. Nature 1993; 365:666-669. 42. von Schwedler U, Kornbluth RS, Trono D. The nuclear localization signal of matrix protein of human immunodeficiency virus type 1 allows the establishment of infection in macrophages and quiescent T lymphocytes. Proc Nat Acad Sci USA 1994; 91:6992-6996. 43. Heinzinger NK, Bukrinsky MI, Haggerty SA et al. The Vpr protein of human immunodeficiency virus type 1 influences nuclear localization of viral nucleic acids in nondividing host cells. Proc Natl Acad Sci 1994; 91:7311-7315.

48

Lentiviral Vector Systems for Gene Transfer

44. Gallay P, Swingler S, Allen C et al. HIV-1 infection of nondividing cells: C-terminal tyrosine phosophorylation of the viral matrix protein is a key regulator. Cell 1995; 80:379-388. 45. Gallay P, Hope TJ, Chin D et al. HIV-1 infection of non-dividing cells through recognition of integrase by the importin/karyopherin pathway. Proc Natl Acad Sci, USA 1997; 94:9825-9830. 46. Gallay P, Stitt V, Mundy C et al. 1996. Role of the karyopherin pathway in human immunodeficiency virus type 1 nuclear import. J Virol 70:1027-1032. 47. Paillart JC, Gottlinger HG. Opposing effects of human immunodeficiency virus type 1 matrix mutations support a myristyl switch model of gag memrane targeting. J Virol 1999; 73:2604-2602. 48. Fouchier RAM, Malim MH. Nuclear import of human immunodeficiency virus type-1 preintegration complexes. Adv Virus Res 1999; 52:275-299. 49. Freed EO. Phosphorylation of residue 131 of HIV-1 matrix is not required for macrophage infection. Cell 1997; 88:171-173. 50. Freed EO, Englund G, Martin, MA. Role of the basic domain of human immunodeficiency virus type 1 matrix in macrophage infection. J Virol 1995; 69:3949-3954. 51. Fouchier RAM, Meyer BE, Simon JHM et al. HIV-1 infection of non-dividing cells: Evidence that the amino-terminal basic region of the viral matrix protein is important for Gag processing but not for post-entry nuclear import. EMBO J 1997; 16:4531-4539. 52. Freed, EO. HIV-1 gag proteins: Diverse functions in the virus life cycle. Virology 1998; 251:1-15. 53. Haffar OK, Popov S, Dubrosvsky L et al. Two nuclear localization signals in the HIV-1 matrix protein regulate nuclear import of the HIV-1 pre-integration complex. J Mol Biol 2000; 299:359-368. 54. Reil H, Bukovsky AA, Gelderblom HR et al. Efficient HIV-1 replication can occur in the absence of the viral matrix protein. EMBO J 1998; 17:2699-2708. 55. Westervelt P, Henkel T, Trowbridge DB et al. Dual regulation of silent and productive infection in monocytes by distinct human immunodeficiency virus type 1 determinants. J Virol 1992; 66:3925-3931. 56. Conno RI, Chen BK, Choe S et al. Vpr is required for efficient replication of human immunodeficiency virus type 1 in mononuclear phagocytes. Virology 1995; 206:935-944. 57. Ballliet JW, Kolson DL, Eiger G et al. Distinct effects in primary macrophages and lymphocytes of the human immunodeficiency virus type 1 accessory genes, vpr, vpu, and nef: Mutational analysis of a primary HIV-1 isolate. Virology 1995; 206:935-934. 58. Popov S, Dubrovsky L, Lee MA et al. Critical role of reverse transcriptase in the inhibitory mechanism of CNI-H0294 on HIV-1 nuclear translocation. Proc Nat Acad Sci USA 1996; 93:11859-11864. 59. Vodicka MA, Koepp DM, Silver PA et al. HIV-1 Vpr interacts with the nuclear transport pathway to promote macrophage infection. Genes & Dev 1998; 12:175-185. 60. Fouchier RA, Meyer BE, Simon JH et al. Interaction of the human immunodeficiency virus type 1 Vpr protein with the nuclear pore complex. J Virol 1998; 72:6004-6013. 61. Popov S, Rexach M, Blobel G et al. Viral protein R regulates docking of the HIV-1 preintegration complex to the nuclear pore complex. J Biol Chem 1998; 273:13347-13352. 62. Jenkins Y, McEntee M, Weis K et al. Characterization of HIV-1 vpr nuclear import: Analysis of signals and pathways. J Cell Biol 1998; 143:878-885. 63. Sherman MP, Noronha DE, Heusch MI et al. Nucleocytoplasmic shuttling by human immunodeficiency virus type 1 Vpr. J Virol 2001; 75:1522-1532. 64. Kamata M, Aida Y. Two putative alpha-helical domains of human immunodeficiency virus type 1 Vpr mediate nuclear localization. J Virol 2000; 74:7179-7186. 65. Adam SA, Sterne Marr R, Gerace L. Nuclear protein import in permeabilized mammalian cells requires soluble cytoplasmic factors. J Cell Biol 1990; 111:807-816. 66. Popov S, Rexach M, Zybarth G et al. Viral protein R regulates nuclear import of the HIV-1 preintegration complex. EMBO J 1998; 17:909-917. 67. Fletcher TMI, Brichacek B, Stivahtis G et al. 1996. Nuclear import and cell cycle arrest functions of the HIV-1 Vpr protein are encoded by two separate genes in HIV-2/SIVsm. EMBO J 15:6155-6165.

Determinants for Lentiviral Infection of Non-Dividing Cells

49

68. Stivahtis GL, Soares MA, Vodicka MA et al. Conservation and host-specificity of Vpr-mediated cell cycle arrest suggest a fundamental role in primate evolution and biology. J Virol 1997; 71:4331-4338. 69. Tsurutani N, Kubo M, Maeda Y et al. Identification of critical amino acid residues in human immunodeficiency virus type 1 IN required for effiecient proviral DNA formation at steps prior to integration in dividing and nondividing cells. J Virol 2000; 74:4795-4806. 70. Pluymers W, Cherepanov P, Schols D et al. Nuclear localization of human immunodeficiency virus type 1 Integrase expressed as a fusion protein with green fluorescent protein. Virology 1999; 258:327-332. 71. Bouyac-Bertoia M, Dvorin JD, Fouchier RAM et al. HIV-1 Integrase requires a functional NLS. Mol Cell 2001; 7(5):1025-1035. 72. Deminie CA, Emerman M. Functional exchange of an oncoretrovirus and a lentivirus matrix protein. J Virol 1994; 68:4442-4449. 73. Dupont S, Sharova N, DeHoratius C et al. A novel nuclear export activity in HIV-1 Matrix protein required for viral replication. Nature 1999; 402:681-685. 74. Zennou V, Petit C, Guetard D et al. HIV-1 genome nuclear import is mediated by a central DNA flap. Cell 2000; 101:173-185. 75. Charneau P, Mirambeau G, Roux P et al. HIV-1 reverse transcription a termination step at the center of the genome. J Mol Biol 1994; 241:651-662. 76. Charneau P, Alizon M, Clavel F. A second origin of DNA plus-strand synthesis is required for optimal human immunodeficiency virus replication. J Virol 1992; 66:2814-2820. 77. Follenzi A, Ailles LE, Bakovic S et al. Gene transfer by lentiviral vectors is limited by nuclear translocation and rescued by HIV-1 pol sequences. Nat Gen 2000; 25:217-222. 78. Sirven A, Plfumio F, Zennou V et al. The human immunodeficiency virus type-1 central DNA flap is a crucial determinant for lentiviral vector nuclear import and gene transduction of human hematopoietic stem cells. Blood 2000; 96:4103-4110. 79. Rumbaugh JA, Fuentes GM, Bambara RA. Processing of an HIV replication intermediate by the human DNA replication enzyme FEN1. J Biol Chem 1998; 273:28740-28745. 80. Naldini L, Blomer U, Gallay P et al. In vivo gene delivery and stable transduction of nondividing cells by a lentiviral vector. Science 1996; 272:263-267. 81. Rey O, Canon J, Krogstad P. HIV-1 Gag protein associates with F-actin present in microfilaments. Virology 1996; 220:530534. 82. Ott DE, Coren LV, Johnson DG et al. Actin-binding cellular proteins inside human immunodeficiency virus type 1. Virology 2000; 266:42-51. 83. Bukrinskaya A, Brichacek B, Mann A et al. Establishment of a functional human immunodeficiency virus typ 1 (HIV-1) reverse transcription complex involves the cytoskeleton. J Exp Med 1998; 188:2113-2125. 84. Kukolj G, Jones KS, Skalka AM. Subcellular localization of avian sarcoma virus and human immunodeficiency virus type 1 integrases. J Virol 1997; 71:843-847.

CHAPTER 4

HIV-1 Vector Systems Narasimhachar Srinivasakumar

Abstract

H

uman immunodeficiency virus type 1 (HIV-1) based gene transfer systems are gaining in popularity due to their ability to transduce terminally differentiated and non-dividing cells. Oncoretroviral vectors based on Moloney murine leukemia virus (MoMLV), on the other hand, can only transduce dividing cells. The reasons for increased ability of lentivirus vectors to transduce such cells has been attributed to several of the viral proteins (integrase, matrix and Vpr) that are purported to be involved in the nuclear import of the pre-integration complex (PIC). Nuclear import is also augmented by a unique triple stranded DNA region created during reverse transcription of the incoming viral RNA in the target cell (discussed in chapter 3). This chapter deals with the rationale behind the design of human immunodeficiency virus type 1 (HIV-1) based packaging systems with an emphasis on some recent advances in the field for the creation of safe and efficient HIV-1 based vectors. The review covers trans-acting proteins and cis-sequences required for the deployment of HIV-1 vectors for gene transfer. This is a rapidly advancing field that with further refinements may soon allow the utilization of HIV-1 based and/or other lentivirus vectors in a clinical setting.

Introduction Designing a gene transfer system using HIV-1 requires an understanding of key features of the replicative cycle of the virus (see chapter 2 for details). A brief overview of HIV-1 replication is presented below emphasizing those aspects critical for the creation of an HIV-1 based packaging system. The overview is followed by a detailed description of the various components of an HIV-1 based packaging systems including the various flavors it comes in.

Regulation of Gene Expression The proviral form of HIV-1 has the prototypical structure of retroviruses with the coding regions sandwiched between two long terminal repeats (LTR). Each LTR consists of unique 3 (U3), repeat (R) and unique 5 (U5) regions. The viral enhancer and promoter elements are present in the U3 region. The HIV-1 genome encodes for at least nine proteins (Fig. 1). Transcription of vector RNA begins at the first nucleotide of R in the 5' LTR and polyadenylation (pA) occurs at the last nucleotide of R in the 3' LTR.1 Thus the genomic RNA is bounded by the r and u5 sequences at the 5' end and u3 and r sequences at the 3' end. Transcription from the 5' LTR requires the viral transcriptional activator protein Tat. Tat, unlike most transcriptional activators that contain specific DNA binding motifs, binds to a cis-element, the transacting response (or TAR) element at the 5’ end of the nascent RNA within r and enhances processivity of RNA polymerase II.2-5 This results in an abundant amount of genome-length Lentiviral Vector Systems for Gene Transfer, edited by Gary L. Buchschacher, Jr. ©2003 Eurekah.com.

52

Lentiviral Vector Systems for Gene Transfer

Fig. 1. Schematic representation of the coding regions in HIV-1 provirus (A). The coding region of the viral regulatory proteins, Tat and Rev, are each derived from two exons. The major 5' splice site is marked but other splice donor and acceptor sites are not shown. Transcription from the 5' LTR of the provirus results in single genome-length RNA (B) from which all other RNA species are derived. The full-length or unspliced mRNA is used for encapsidation into the virus particle and is also the substrate for translation into Pr55Gag and Pr160Gag-Pol precursor polyproteins (C). The viral proteins Vif, Vpr, Vpu and Env are translated from singly spliced mRNAs while Rev, Tat and Nef are translated from multiply spliced RNAs. The various subunits derived from the two precursor polyproteins are shown. Further details are provided in the text.

viral RNA to be produced in the infected cell. In the absence of Tat most transcripts suffer premature termination. The full-length message derived from the provirus is alternatively spliced to generate multiple species of mRNA. At least 30 species of mRNA have been described in HIV-1 with several slightly different mRNAs coding for the same protein.6-13 The fully spliced mRNAs encode for Rev, Tat or Nef. The unspliced message codes for Gag and Gag-Pol proteins while the singly and partially spliced messages code for envelope (Env) and the accessory proteins Vif, Vpr or Vpu. Since incompletely spliced messages are usually retained in the nucleus of eukaryotic cells, the expression of proteins derived from these messages requires that the corresponding RNAs exit the nucleus without undergoing splicing.14-17 Moreover, sequences have been identified in gag that retain the unspliced message in the nucleus.18-20 The inhibitory sequences have been referred to as cis-acting repressor sequence (CRS) elements or cis-acting inhibitory sequences (INS). The Rev protein binds to a target sequence within the env coding region called Rev-response element (RRE) to bring about nucleo-cytoplasmic transport of such messages. Rev may also play a role in the stability of the RNA and its polysomal association or translation.21,22 Studies have indicated that for Rev to bring about nucleo-cytoplasmic transport, the RNA must form a complex with the splicing machinery. Thus, an intact 5’ splice

HIV-1 Vector Systems

53

donor site in the RNA is essential for Rev to produce its effect.23 The 5’ splice donor is also used for the generation of partially spliced messages that are translated into Vif, Vpr and Vpu proteins.

Virion Assembly HIV-1 is an enveloped virus. The surface of the virus particle is decorated with the envelope (Env) glycoprotein consisting of a surface glycoprotein (gp120) and a transmembrane protein (gp41). Lining the inner aspect of the lipid envelope is the viral matrix (MA/p17) protein. The most characteristic structure of the virus is the cone shaped core composed predominantly of the capsid (p24) protein. Also present within the virion are the nucleocapsid (NC/p7) protein , and the viral enzymes consisting of the reverse transcriptase (RT), integrase (IN) and protease (PR). The internal proteins of the virion are derived for the most part from the Gag (Pr55gag) and GagPol (Pr160 gagpol) precursor polyproteins (Fig. 1). Both precursors are translated from the unspliced mRNA molecule containing the gag and pol coding sequences. The Pr160Gag-Pol is synthesized by a ribosomal frame-shifting mechanism that occurs during translation of Gag once every 10 or 20 translation events.1,24 The Gag precursor can assemble into virus-like particles in the absence of other internal proteins of the virion.25-35 The Gag-Pol precursor, while it cannot form virus particles on its own, is drawn into the assembling particle through its interaction with the Gag precursor.36-39 The enzymes present as part of the Gag-Pol precursor, PR, RT and IN, are thus incorporated into the virus particle. Vpr and Vif have also been demonstrated within the virion. Vpr is incorporated into assembling virus particle through an interaction with the p6 protein present in the C-terminus of the Gag polyprotein.40-42 Vif is incorporated into the virion via an interaction with the genomic RNA.43 The precursor polyproteins, together with two molecules of genomic RNA, accumulate beneath the portion of the plasma membrane containing the viral Env glycoprotein. A viruslike particle then pinches off from the plasma membrane. The genomic RNA is incorporated into the assembling virion via interactions between the nucleocapsid (NC) protein within Gag/ Gag-Pol proteins and the encapsidation or packaging (E/ψ) signal present in the vector RNA molecule. Following release of the virus particle, the Pr160Gag-Pol precursor is cleaved to yield the viral PR, RT and IN proteins. Proteolytic cleavage of the Gag precursor yields the MA (p17), CA (p24), NC (p7) and p6 proteins. The mature virus particle is then ready to infect target cells.

Virus Entry Following binding of the virus to the target cells via the Env glycoprotein and corresponding receptor/coreceptors on the target cell, fusion of the viral and plasma membrane of the cell ensues. This results in the deposition of the viral core into the cytoplasm. The core is stripped of the capsid protein and the genomic RNA is reverse transcribed into a double stranded cDNA. During reverse transcription, due to presence of a central polypurine tract (cPPT) and termination sequence (CTS), a unique triple-stranded DNA region consisting of the double stranded DNA and a DNA flap is created in the middle of the newly synthesized cDNA. This unique structure, by an unknown mechanism, enhances the import of the viral pre-integration complex (PIC) into nucleus of the target cell.44-46 The PIC also contains the viral IN protein, MA and Vpr proteins. These proteins may also be involved in nuclear import of the PIC although the role of the MA protein in nuclear import is controversial. The cDNA is then inserted into the chromosome of the target cell, apparently at a random location, by the viral IN protein.

HIV-1 Vector System Development The design of lentivirus packaging systems has benefited greatly from detailed studies conducted with oncoretroviral based packaging systems. However, creating a packaging or

54

Lentiviral Vector Systems for Gene Transfer

Fig. 2. A schematic representation of the components of an HIV-1 based vector system consisting of a packaging or helper construct (A), an Env expression construct (B) and a gene transfer vector (C). The various RNA species derived from each expression construct are shown. Rev is required for expression of helper and gene transfer vector RNA, both of which contain the RRE. Tat is required for expression of the gene transfer vector RNA from the viral LTR promoter. The various components (proteins and two copies of the vector RNA) assemble into a virus particle at the plasma membrane. Once budded off from the membrane, the virus is then ready to infect and transduce the gene transfer vector together with its encoded transgene into target cells.

gene transfer system using HIV-1 poses several unique challenges. This is because of the complexity of gene expression in HIV-1 (see above and chapter 2). For instance, expression of both helper and gene transfer vector RNAs require sequences that ensure the transport of the RNA molecules into the cytoplasm. To create a packaging system with HIV-1, one needs a) packaging or helper constructs that provide all necessary virion proteins to form a virus-like particle and b) a gene transfer vector that provides the RNA for encapsidation by the assembling virus particle (Fig. 2). The gene transfer vector ideally should expresses only the transgene of interest but none of the viral proteins. The helper proteins consist of HIV-1 proteins that form the internal proteins of the virion and an Env protein, either from HIV-1 or, more commonly, from a different virus, which allows the virus particle to infect target cells bearing the appropriate receptor. The basic components of an HIV-1 based packaging system are described below.

HIV-1 Vector Systems

55

Fig. 3. Cis-elements required for creation of an HIV-1 based gene transfer vector. The putative secondary structure of the E/ψ signal that encompasses the entire 5' untranslated region is depicted. Elements involved in reverse transcription and integration: primer binding site (PBS), polypurine tract (PPT),central polypurine tract and central termination sequences (cPPT/CTS) and integration (att sites) are shown. The cPPT/CTS, while not absolutely essential for the reverse transcription step in the context of lentivirus vectors, improves nuclear translocaton of the pre-integration complex (PIC) in both dividing and non-dividing cells. The left att site is derived from the u3 region present in the RNA genome while the right att site is derived from the u5 region in the RNA. The RRE is required for nucleo-cytoplasmic of the vector RNA. Several different lengths of RRE can be used. The nucleotide positions shown correspond to the extended RRE described by Mann et al.208 ψ: packaging signal. The transgene expression cassette is usually positioned between the 3' Tat/Rev splice acceptor site and the 3' LTR. Details of the cis-elements are discussed further in the text. The sequences and numbers shown are with reference to the molecular clone pNL4-3 (GenBank Accession number M19921).

Packaging or Helper Constructs Packaging constructs are designed to ensure high levels of particle production. A typical packaging plasmid (Fig. 2) contains a heterologous promoter driving the expression of the HIV-1 gag/pol coding region. The HIV-1 polyadenylation signal present in the 3' LTR is replaced with a heterologous polyadenylation signal. Also required are signals for transport of gag/pol RNA from the nucleus to the cytoplasm for efficient translation. Without a transport mechanism, no Gag and Gag-Pol expression can occur. The Rev and RRE of HIV-1 are most commonly used to provide transport functions for the gag/pol message. The RRE sequence is positioned between the 3' end of the Gag-Pol coding sequence and the polyadenylation signal. The major 5' splice donor site present upstream of gag is retained for Rev function. Expression of HIV-1 Gag and Gag-Pol from this construct requires the coexpression of Rev in trans during virus stock production. Rev can be expressed by including rev coding sequences together with the gag/pol sequences in the same packaging construct (Fig. 2) or can be provided using a separate plasmid encoding the cDNA of Rev (Fig. 7).

56

Lentiviral Vector Systems for Gene Transfer

Gene Transfer Vectors The gene transfer vector should contain, in addition to the transgene expression cassette, all cis-acting sequences required for transcription, polyadenylation, efficient nucleo-cytoplasmic transport, encapsidation, reverse transcription and integration of the vector RNA. The ideal vector lacks all viral protein coding sequences and has a capacity for large transgenes (Figs. 2 and 3). It is believed that HIV-1 vectors can accommodate a transgene expression cassette of approximately 8-11 kb in length.47,48 Recent studies have also revealed that providing cissignals that enhance nuclear import of PICs would be beneficial. The cis-elements that are required for constructing gene transfer vectors (Fig. 3) are described, as they occur from 5' to 3' in the vector, in greater detail below.

Transcription Signals Transcription of the vector RNA requires the enhancer and promoter elements in the U3 region and the R region in the 5’LTR. The R region contains the TAR element that mediates the transactivation by viral Tat protein to generate abundant amounts of genome-length RNA.

Encapsidation Signals In order for the vector RNA to be encapsidated into the assembling virus particle, the vector must contain an E/ψ signal. The entire 5' untranslated region constitutes the E/ψ region and exhibits a complex secondary structure (Fig. 3). This region consists of seven stem-loop structures. Four of these stem-loop (SL) structures encompassing the 5’ splice donor site and the 5' end of gag have been termed SL1, SL2, SL3 and SL4.49 SL1, SL3 and SL4 are involved in Gag binding while SL2 contains the 5' splice site and is not essential for encapsidation. SL1 also contains the dimer-linkage signal (DLS) in its loop. Due to its location, SL1/DLS is present in all HIV-1 RNAs, i.e., it is present in both spliced and unspliced RNAs. Deletion or mutations in SL1 or SL3 reduces encapsidation of the RNA. Combined deletion of SL1 and SL3 has a greater effect then deletion of each SL by itself. Studies have revealed that sequences upstream of the 5’ splice donor site including the lower stem of TAR and the poly A stem-loop are also essential for packaging.50-53 It should be emphasized that mutations to the E/ψ region or structures do not completely eliminate packaging but instead allow spliced RNAs to be packaged in direct relation to their intracellular concentration.51,54,55 This is of concern because most packaging systems designed to date contain a deletion within the so-called E/ψ region present in the packaging or helper plasmid with the hope that the RNA derived from the helper construct will not be copackaged with the vector RNA. The good news is that in the presence of RNA with an intact E/ψ region, this RNA is likely to be preferred over the RNA with deletions in the E/ψ region. However, it also appears that sequences outside of the E/ψ region can influence encapsidation and or dimerization. For instance, presence of foreign sequences can negatively affect encapsidation of vector RNA. Foreign sequences have also been shown to enhance dimerization of RNA. Although the encapsidation and dimerization sequences overlap, they can be functionally delinked from each other. Thus mutations to the dimer linkage site can affect infectivity without affecting encapsidation.54 Thus more careful analysis of encapsidation and dimerization of RNA present within vector particles is required to evaluate the effect of the deletions or mutations within the E/ψ region of the helper construct on its encapsidation and delivery to target cells.

gag Sequences Studies in MoMLV suggested that the encapsidation signal extends into the 5' portion of the gag.56 This appears to be true in the case of HIV-1 also. The SL4 structure of the HIV-1 encapsidation signal is present in the extreme 5' end of the gag coding region.49 While at least 40 nucleotides of gag are required for optimal packaging of the vector RNA, inclusion of

HIV-1 Vector Systems

57

additional sequence of up to 653 or even up to 1053 nucleotides improves encapsidation and/ or gene transfer.55,57,58 It is not clear how exactly these additional gag sequences influence encapsidation. Most investigators use about 350 nucleotides of gag sequence in gene transfer vectors. A frameshift or stop codon is inserted near the 5’end taking care not to disrupt the structure of SL4 packaging element to ensure that no significant length of Gag peptide is synthesized in the transduced target cell.59

Nucleo-Cytoplasmic Transport Signals Just as for the expression of helper proteins, the transport of full-length vector RNA also requires Rev and RRE. Likewise, an intact 5’ splice donor site is required for Rev to function with the RRE.23 The gag coding region harbors so-called inhibitory sequences (CIS/CRS) that retain the RNA in the nucleus.19,20,60 Since gene transfer vectors lack most of the coding regions of HIV-1, the requirement for transport signals is a bit puzzling. Consistent with this, it appears that if one eliminates almost all of the gag region with the exception of the 5’ 40 nucleotides, one can then create a vector lacking the RRE and containing a mutation in the 5’ major splice donor site. Such a vector has nearly 50% of the titer of wild-type vector.61

Transgene Expression Cassette All gene transfer vectors contain, by definition, a transgene expression cassette. This cassette is usually positioned between the RRE and the 3' LTR. The transgene expression cassette usually contains promoter-enhancer elements from a heterologous virus or from a cellular promoter. Some of the more popular promoters evaluated in HIV-1 vectors are human and simian cytomegalovirus immediate early promoters, simian virus 40 early promoter , phosphoglycerate kinase promoter and the elongation factor, EF1α promoter.59,62-65 HIV-1 vectors that express the transgene under control of the viral LTR promoter have also been described.63,64 In this case, it is necessary to express the viral Tat protein together with the transgene of interest. The choice of the promoter is dictated by the target cell for gene transfer since the same promoter and enhancer elements may work with differing efficiencies in different cell types.63,65,66

Polyadenylation Signal The polyadenylation (pA) signal for the vector RNA is derived from the 3' LTR and the core AATAAA sequence in HIV-1 is present within the R region. Several ‘enhancers’ of pA have been identified in the upstream U3 region using artificial constructs.67-69 It appears that most of the U3 sequence can be deleted without compromising titer which in turn suggests that the effect of these pA enhancers are not profound in the context of HIV-1 vectors.70,71 This aspect is discussed in greater detail under self-inactivating (SIN) vectors

Sequences Involved in Reverse-Transcription and Integration The key elements required for reverse-transcription of viral RNA into a double-stranded DNA molecule are the primer binding site, the repeat (R) regions and the polypurine tract (Chapter 2). In addition to these signals, sequences recognized by the integrase present at the 5' end of U3 and 3' end of U5 (att sites) are required to effect integration of the reverse-transcribed product into the target cell chromosome. Unique to lentiviruses is the presence of a central polypurine tract (cPPT) and central termination sequence (CTS) within pol.44 During reverse transcription, these sequences, allow for the formation of a unique triple-stranded DNA molecule in this portion of the HIV-1 reverse-transcribed product. This triple-stranded DNA region or DNA-flap appears to enhance the entry of HIV PIC into the nucleus of both dividing and non-dividing cells.45,46 Thus, most recent modifications to HIV-1 vectors include the cPPT and CTS.

58

Lentiviral Vector Systems for Gene Transfer

Envelope Expression Construct The final component of an HIV-1 based packaging system is the Env expression construct. The HIV-1 virus particle is quite promiscuous in that it can be pseudotyped with Env proteins derived from many different viruses. One study compared several Env proteins from rhabodviruses (G glycoprotein of rabies, Mokola and vesicular stomatitis virus), MoMLV (4070A Env), HIV-1 and human foamy virus.72 All the different Env proteins could be used to pseudotype HIV to varying degrees with the Rhabdovirus Envs and MoMLV Env pseudotyped vectors exhibiting the highest titers. The VSV-G pseudotyped vectors are the most widely used because such vectors are quite stable under high centrifugation forces and can be readily concentrated by ultracentrifugation.73 In contrast to VSV-G, the amphotropic MoMLV Env pseudotyped vectors cannot be concentrated by ultracentrifugation without loss of titer. Some Env proteins cannot be used to pseudotype HIV-1 vectors. One example of this is the gibbon ape leukemia virus (GALV) Env. This Env is of interest because previous studies showed that the GALV Env allows enhanced gene transfer of MoMLV vectors into hematopoietic stem cells.74 The incompatibility in GALV ENV that prevents pseudotyping of HIV-1 appears to reside in the cytoplasmic domain of the protein because partial or complete substitution of the cytoplasmic tail of GALV Env with that of MoMLV Env allows more efficient pseudotyping of HIV-1 vectors.75 Another Env protein of interest is derived from the spleen necrosis virus (SNV). Vectors pseudotyped with this Env protein cannot infect human cells but can infect canine cells. In a series of elegant experiments 76-80 Dornburg’s group has shown that modification of SNV Env with an appropriate single-chain antibody of known specificity allows one to target SNV based vectors to specific cell types bearing unique markers, such as the CD34 antigen on hematopoietic stem cells, transferrin receptor present on liver cells or Her2neu antigen present on many human cells. The SNV Env can be used to pseudotype HIV-1 vectors but suffers from the drawback that the resultant vector titers are not very impressive and concentration of vector particles by ultracentrifugation is not feasible without a significant drop in titer (Srinivasakumar et al., unpublished observations). Ultrafiltration of vector particles is an alternative approach for concentration of vector particles bearing Envs that are sensitive to high g-forces. Unfortunately, ultrafiltration of vectors pseudotyped with amphotropic Env or SNV Env appears to enrich for inhibitors of virus binding and/or entry.81,82 This results in a paradoxical decrease in virus titer at lower dilutions that nullifies the effect of concentration. The inhibitors are suspected to be high molecular weight sulfated proteoglycans such as chondroitin sulfate.83-86 In contrast to these observations, other investigators have reported that ultrafiltration actually removes inhibitors.77,87 If methods can be devised for efficient removal of these inhibitors, these other Env proteins can then be used for pseudotyping lentivirus vectors to provide a different host-range for gene transfer ex vivo or in vivo.

Accessory and Regulatory Proteins and Gene Transfer Rev The requirement of Rev and RRE for expression of packaging and gene transfer vector RNAs has been discussed above.

Tat Tat is not required for expression of viral helper proteins but is required for the expression of the gene transfer vector RNA (see above). Tat is, therefore, coexpressed along with the other helper plasmids for production of virus stocks. The Tat protein is encoded in two exons. However recent experiments have shown that only the first coding exon of Tat is sufficient for production of vector stocks for gene transfer.63,88 Tat has also been implicated in causation of

HIV-1 Vector Systems

59

immune modulation, aberrant cytokine expression and induction of Kaposi’s sarcoma.89-104 It is therefore felt that Tat should be excluded from packaging systems if possible. The requirement for Tat can be obviated by making gene transfer vectors Tat-independent (see below).

Vif, Vpr and Vpu Early versions of packaging plasmids encoded most viral structural and regulatory proteins with the exception of the Env glycoprotein.59,72,105,106 This was probably because the viral accessory proteins Vif, Vpr and Vpu appeared to play important roles in the replicative cycle of HIV-1 and it was not clear if deletion of these proteins would affect gene transfer in any way. Vif has been shown to increase infectivity of virions produced in certain human T-cell lines.107,108 Vif appears to affect events after binding and fusion, either during uncoating or during RT.109 Vif exerts its effect in the producer cell. Producer cells can be classified as permissive or non permissive. In permissive cells (e.g., 293), both Vif(-) and Vif(+) constructs can be used to produce virus stocks that are equally infectious for T cells. In contrast, in non permissive cells (e.g., H9 and peripheral blood mononuclear cells) the virions produced with Vif(-) constructs are relatively less infectious than those produced with Vif(+) constructs. Vpu is a 16-kilodalton phosphoprotein that has been shown to increase virion export in human cells such as HeLa or Jurkat but not in cell lines of simian origin (e.g., Cos).110,111 In this function, it resembles the p6 protein present at the C-terminus of the Gag precursor that is also involved in virion export112 but differs from Vpu in that it demonstrates its activity in simian cells and not in human cells.111 Other functions of Vpu include down modulation of cell surface expression of CD4 and major histocompatibility complex class I molecules.113,114 Vpr is another protein with pleiotropic effects. It can cause cells to arrest in G2 phase of the cell cycle.115-117 This effect has been linked to a modest increase in transcriptional output from the viral LTR promoter.118 Vpr has also been shown to be essential for importation of PIC into the nucleus of macrophages.119-122 Vpr is recruited into the virus particle via its interaction with the p6 region of the Gag molecule.40-42 One possible disadvantage of Vpr is that when concentrated stocks of HIV-1 vector stocks are used in gene transfer experiments, Vpr delivered by the virus particles can result in apoptosis of target cells.123-125 So the advantages of using Vpr for nuclear import of PICs in some types of cells must be weighed against its propensity to cause apoptosis. In spite of these known functions of Vif, Vpr and Vpu, it appears that one can safely eliminate these proteins for preparation of high-titer virus stocks in 293T cells for transduction of many cell types. Recent reports indicate, however, that accessory proteins may be necessary for getting optimal levels of gene transfer into certain types of cells such as lymphocytes.126 A more careful evaluation for the requirement of accessory proteins for transduction of different cell types appears to be therefore warranted.

Nef The Nef protein is synthesized early after infection together with Tat and Rev and has pleiotropic effects. Nef has been shown to down modulate cell-surface expression of CD4 by inducing endocytosis, resulting in the degradation of this molecule in lysosomes.127-130 Coexpression of Nef during virus stock production increases efficiency of gene transfer into target cells.131-134 This has been attributed in part to its positive effect on increasing the efficiency of reverse transcription during viral entry.134,135 Fortunately, this effect of Nef is restricted to certain types of Env proteins used for pseudotyping virus. For instance, Nef increases infectivity of virions pseudotyped with HIV-1 or amphotropic MoMLV Env proteins but not those pseudotyped with VSV-G or Ebola virus Env proteinss.132,136,137 Thus Nef is not required for vectors pseudotyped with Env proteins that effect low pH mediated fusion and entry of viruses. Nef can, therefore, be eliminated from packaging systems that employ VSV-G

60

Lentiviral Vector Systems for Gene Transfer

Fig. 4. Schematic representation of simple gene transfer vector systems based on HIV-1. In A, a transgene expression cassette consisting of a heterologous promoter driving a marker or therapeutic gene is inserted into the env coding region and thus inactivates it. To produce infectious virus one needs to complement the above construct with a separate expression construct for Env (B). The transduced target cell not only expresses the transgene but also all other viral proteins with the exception of Env. A replication competent gene transfer vector is shown in C. The transgene in this vector is positioned within Nef and is expressed from a spliced mRNA. The expression of the transgene is, therefore, under control of the viral LTR promoter. The vector produces all viral proteins with the exception of Nef.

to pseudotype vectors. This is a fortunate occurrence, because experiments in transgenic mice have revealed that expression of HIV-1 Nef in lymphocytes and macrophages leads to a profound immune defect.138 Due to these negative connotations, it may be prudent to remove Nef from HIV-based packaging systems unless it is required for increasing the efficiency of gene transfer with some Env proteins.

Evolution of HIV-1 Vector and Packaging Systems Early HIV-1 Vector Systems The simplest HIV-1-based gene transfer systems contained not only the transgene expression cassette but also encoded for most of the internal proteins of HIV-1. The transgene expression cassette was usually located within the env coding region. To produce infectious virus with such constructs, one only needs to complement with an Env glycoprotein expression construct during virus stock production (Fig. 4 A).139,140 In a variation of the above theme (Fig. 4 B), HIV-1 vectors that contain the transgene within the nef coding region have been described.141,142 In this configuration, the virus encodes for all transacting viral protein including Env (with the exception of Nef ) as well as the transgene. Thus, this vector is replication competent. Clearly, the coexpression of viral proteins together with the transgene make these packaging systems less than ideal for use in gene therapy applications. However, such vectors are useful for studying the biology of the virus.141,142

HIV-1 Vector Systems

61

Fig. 5. Schematic representation of an HIV-1 vector system 59 consisting of a first generation packaging plasmid (A), gene transfer vector (B) and an Env expression construct (C). The packaging plasmid encodes for all HIV-1 proteins with the exception of the Env protein. The 5' LTR is replaced with a heterologous promoter and the 3’LTR is replaced with a heterologous poly A (pA) signal. A deletion is engineered into the ψ region (∆ψ). The gene transfer vector does not express any of the viral proteins and contains all cissequences for packaging, reverse transcription and integration (Fig. 2). It contains a transgene expression cassette containing a therapeutic or marker gene under control of a heterologous internal promoter. An Env expression construct (C) is provided for pseudotyping the vector to allow infection of target cells containing the appropriate receptor to which the Env protein can bind.

First Generation Packaging System Subsequent vector systems contained three components: a first generation packaging plasmid, an Env protein expressing plasmid and a gene transfer vector59 (Fig. 5). The packaging plasmids were simple modifications of the HIV-1 proviral genome. The 5’LTR was replaced with heterologous viral promoter/enhancer elements. A deletion within the E/ψ region present between the 5’ major splice donor site and the beginning of Gag coding sequence was engineered to prevent encapsidation of the RNA derived from the helper plasmid. The downstream promoter was replaced with a heterologous poly A signal. The env coding sequence was interrupted by a mutation but still retained the RRE. Such a helper construct expresses all viral proteins, including Tat and Rev with the exception of the viral Env. A separate Env expression construct was provided by cotransfection during virus stock production. Typically a VSV-G expressing envelope construct was provided to allow concentration of vector stocks by ultracentrifugation. The gene transfer vector contained the 5' LTR and the entire 5' untranslated region including about 350 nucleotides of Gag coding sequences. The vector also contained the RRE and the 3' LTR. The transgene expression cassette was positioned between the RRE and the 3' LTR. The transgene expression cassette typically used the cytomegalovirus immediate early promoter and enhancer elements to drive a green fluorescent protein gene or firefly luciferase gene or β-galactosidase gene.

62

Lentiviral Vector Systems for Gene Transfer

Fig. 6. Schematic representation of a second generation HIV-1 packaging system,72,106,143,144 along with a vector and Env-expressing construct. This is similar to the first generation packaging system (Fig. 5) but for the elimination of the viral accessory proteins Vif, Vpr , Vpu and Nef.

Second Generation Packaging System Many of the HIV-1 accessory and regulatory proteins have been implicated in the pathogenesis of acquired immune deficiency syndrome (AIDS) or have other untoward effects (see above). The second generation packaging systems were, therefore, designed not to express Vif, Vpr or Vpu but in other respects are identical to the first generation packaging system described above72,106,143,144 (Fig. 6). Virus stocks produced with such a minimal packaging system have been found to be effective in transducing most target cells in vitro and in vivo. It is not clear if this observation will hold true for transduction of all cell types.126

Third Generation Packaging System The major concern in the use of lentivirus based packaging systems is the possibility of generation of a replication competent HIV-1 or novel retrovirus during virus stock production or following gene transfer and transplantation of the tissue back in the host. The recombination between helper and vector constructs can occur at the level of input plasmid DNA within the virus producer cell or by recombination of helper and vector RNA molecules copackaged within the same virus particle. One approach to reduce this possibility is to decrease the homology between the helper and gene transfer vector. Another approach is to segregate gag/pol and rev coding sequences in separate plasmids.145 Such a packaging system uses four plasmids for the creation of vector stocks instead of the usual three (Fig. 7). The first plasmid contains gag/pol coding region together with the RRE. The second plasmid encodes for Rev cDNA to allow expression of gag/pol from the first plasmid. The third plasmid is the gene transfer vector while the fourth is the Env expression construct. Use of this type of packaging system necessitates co-evolution of the gene transfer vector, since otherwise one would also require a fifth construct to express Tat. The requirement for Tat by the gene transfer vector can be eliminated by using a Tat-independent vector in which the HIV-1 U3 in the 5' LTR is replaced with enhancer and promoter elements derived from either

HIV-1 Vector Systems

63

Fig. 7. Schematic representation of a third generation HIV-1 packaging system.145 This is a minimal HIV1 packaging system that consists of three helper plasmids: a Gag/Gag-Pol expression construct (A), a Rev expression construct (D) and Env construct (C) and a gene transfer vector. The expression of Gag/Gag-Pol and the gene transfer vector RNAs requires the coexpression of Rev. Rev is produced using a separate expression construct (D). The requirement for Tat for production of large amounts of gene-transfer vector RNA is obviated by using a Tat-independent vector (B) in which the 5’LTR is replaced with a chimeric promoter consisting of heterologus promoter/enhancer elements substituting the corresponding viral elements (see also Fig. 10).

Rous sarcoma virus or from cytomegalovirus immediate early promoter (see below). The use of four plasmids instead of the three and the elimination of Tat increases the safety of the system by further decreasing the probability of recombination between the various helper plasmids and the gene transfer vector to recreate a replication competent virus.

Recent Modifications to the Packaging System Packaging Systems Using Alternative RNA Transport Elements The constitutive transport element (CTE) is a small structured RNA element in MasonPfizer monkey virus (MPMV) that performs a similar function as Rev and RRE in HIV-1 i.e., the transport of unspliced message from the nucleus to the cytoplasm.146,147 Experiments have shown that the CTE can substitute for the function of Rev and RRE in HIV-1 provirus as well in packaging and gene transfer vectors, albeit to varying efficiencies.63,88,106,133,148-152 The use

64

Lentiviral Vector Systems for Gene Transfer

Fig. 8. Design of packaging systems based on the type of RNA-transport elements used in packaging and gene transfer vectors.88 In the RRE/Rev-based packaging system (A) expression of RNA from both helper plasmid and gene transfer vector is regulated by Rev and RRE. Rev coexpression is required for expression of the helper and vector RNAs. In the CTE-based system (B), helper and vector RNA expression is regulated by the MPMV-CTE. This is a Rev-independent packaging system i.e., no Rev coexpression is required for production of vector stocks. In the combination packaging systems (C), the helper plasmid is regulated by CTE while the gene transfer vector is controlled by RRE and Rev. An alternative scenario in which the helper plasmid is regulated by RRE and Rev while the gene transfer vector is controlled by CTE is not shown. Rev coexpression is required for production of vector stocks in this system. Possible sites of recombination between helper and vector constructs in the three types of packaging systems are shown using bi-directional arrows. Note that in C, the two constructs share homology only in the gag region. This design may be safer than other packaging systems for production of vector stocks. For clarity, constructs expressing Env, accessory and regulatory proteins are not depicted. Adapted from: Srinivasakumar N, Schuening FG. J Virology 1999; 73: 9589-9598.

of CTE gives reasonable titers in some studies63,88,133,150,151 while other studies have reported lower titers.106,149 The reason for these differences between the different studies is not yet clear. There are two regions of homology between the helper and gene transfer vector constructs. One is at the 5’ end of gag and the other is the shared RRE or CTE towards the 3’ end of the helper construct and the same element present within the gene transfer vector. Depending on the RNA transport element being used, one can design three types of packaging systems for production of HIV-1 vector stocks88 (Fig. 8) . The traditional or classical HIV-1 packaging system uses the RRE and Rev for the expression of both helper and gene transfer vector RNAs. The second packaging system uses the MPMV-CTE for expression of helper and gene transfer vector RNAs. A possible utility of a CTE-based Rev-independent HIV-1 packaging system is for the delivery of dominant negative forms of Rev into HIV-1 susceptible cells.133,150,151 In a

HIV-1 Vector Systems

65

Fig. 9. Trans-lentiviral packaging system.155 In this packaging system, the proteins encoded in the gag and pol coding regions are segregated in two different expression plasmids. One plasmid encodes for the Gag and Gag-Protease (A) while a separate construct encodes for a Vpr-RT-IN fusion protein (B). The Vpr-RT-IN is expressed using the LTR and RRE from HIV-2. This is based on the observation that Tat and Rev proteins of HIV-1, encoded in the Gag-Pro expression construct, can also function in the context of HIV-2 LTR and RRE-2. The gene transfer vector (C) and an Env expression construct (D) are the other components of the packaging system required for the production of vector stocks.

Rev-RRE based system, the coexpression of a dominant negative Rev as part of the gene transfer vector would be inhibitory for vector stock production. The third kind of packaging system is called the combination or reciprocal packaging system and utilizes the Rev and RRE for expression of one component of the packaging system and the CTE for the other component.88,151 Using dissimilar transport elements for the expression of the helper plasmid and the gene transfer vector RNAs in the combination packaging system may render the packaging system safer by reducing the chance of replication competent retrovirus (RCR) formation. Packaging System Using Codon-Optimized Helper Construct As an alternative approach to decreasing the homology at the gag end of the helper and gene transfer vector constructs, some investigators have constructed a helper plasmid containing a ‘humanized’ gag coding sequence.153 Interestingly, this not only reduces the homology in the gag region due to the silent mutations introduced, it also renders gag expression Revindependent. Elimination of homology at the gag end and at the 3’ end by removal of RRE in the helper plasmid in most probability makes this packaging system one of the safer systems

66

Lentiviral Vector Systems for Gene Transfer

described to date. On the other hand, a non-homologous recombination between the helper and gene transfer vector involving the gag region can result in a recombinant that may be able to express the Gag protein in target cells even in the absence of Rev. One can envisage the use of such codon optimized packaging system for the delivery of antisense RNA expression cassettes targeted to the gag and/or env coding regions of HIV-1. It may be possible to use this system to deliver dominant negative Rev into target cells providing that the inhibitory effect of dominant negative Rev on vector RNA transport in the producer cell can be overcome using the CTE or other modifications to vector backbone. Vector System Using Combination of Two Different Lentiviruses Another approach for decreasing homology between the components of a lentivirus packaging system is to use a helper construct derived from SIV to package an HIV-1 gene transfer vector.154 Currently, such packaging systems are still in their infancy and need to be developed further. For instance, the titer of this system is about an order of magnitude less than that obtainable with an Rev-RRE based HIV-1 packaging system. However such novel systems have great potential because of other benefits. For example, the Vpr of HIV-1 can increase efficiency of infection of macrophages. But Vpr of HIV-1 has the disadvantage of producing cell-cycle arrest and/or apoptosis of target cells when large amounts of protein are delivered via virus particles. In SIV, these functions of HIV-1 Vpr are segregated between two different proteins, the SIV Vpr and Vpx.1 The SIV Vpx, like HIV-1 Vpr, can augment nuclear import of PICS but without the apoptotic or cell-cycle arrest phenotype. This latter function is relegated to the SIV Vpr protein. Thus one can create an SIV based helper plasmid that encodes for Gag, Gag-Pol and Vpx but not Vpr and use this for generating HIV-1 vector stocks. The alternative system containing HIV-1 packaging construct and SIV based gene transfer vector has also been described.62 The latter packaging system does not have some of the advantages of the former, namely the use of SIV Vpx instead of HIV-1 Vpr. Packaging System Using Separate Helper Plasmids for Expression of Gag and Pol Coding Regions A recent novel modification to the packaging system involves a clever approach to separate gag and pol coding regions.155 In this novel packaging system (Fig. 9), gag/protease is expressed using one construct. The other enzymes of the Gag-Pol precursor, namely RT and IN are expressed as a Vpr-RT-IN fusion protein using a separate plasmid construct. The Vpr-RT-IN fusion protein is drawn into the assembling virus particle through its interaction with the p6 domain of the Gag precursor (see Virus Assembly above). Thus the internal proteins of virus particle are derived from two separate plasmid expression constructs. The use of separate plasmids for gag and pol proteins provide a higher margin of safety and has been shown to predictably reduce the frequency of recombination between the packaging plasmids and gene transfer vector.

Modifications to Gene Transfer Vector to Improve Safety and Efficacy Modifications have been engineered into gene transfer vectors not only to improve safety but also to enhance gene expression. Figure 10 shows various modifications to lentivirus vectors to improve efficacy and safety. Self-Inactivating HIV-1 Vectors Due to the nature of reverse-transcription, the promoter/enhancer elements present within the 3’ u3 of the viral genomic RNA is duplicated and positioned in both LTRs of the provirus. Thus promoter disabling mutations engineered into the U3 region of the 3’ LTR will be transferred to the 5’ LTR during reverse-transcription. Such vectors, after integration into the target cells, will not be able to generate full-length vector RNA and are called self-inactivating (SIN)

HIV-1 Vector Systems

67

Fig. 10. Modifications to HIV-1 gene transfer vector to improve safety and efficacy. Details are provided in the text. Adapted from Srinivasakumar N. Packaging cell system for lentivirus vectors. In: Morgan J, ed. Gene Therapy Protocols, 2nd ed. © 2001, Humana Press.

vectors. SIN vectors were first described for MoMLV and SNV vectors.156,157 These studies revealed that mutations in U3 frequently resulted in lower vector titers; this observation was explained by the requirement of some of the sequences in this region for efficient polyadenylation of vector RNA. In HIV-1, the core pA signal (AAUAAA) is present within the ‘R’ region. Sequences in the U3 region have been identified in HIV-1 that also modulate the efficiency of poly adenylation.67,69,158 The most important of these sequences is present between the TATA box and the R region and is about 20 bp in length. In contrast to what has been observed with MoMLV vectors, it appears that most of the U3 region of the 3’ LTR ,with the exception of the poly A ‘enhancer’ described above and the attachment (att) sequence at the 5’end of U3 recognized by viral IN, can be safely deleted without compromising vector titer.70,71 Such a deletion ensures that transcription from the 5’ LTR promoter is efficiently suppressed following reverse transcription and integration into the target cell chromosome. Another advantage of using vectors with deletions in the U3 region is the enhanced transgene expression noticed from some internal promoters in such vectors.65,71,159 This is probably due to a decrease in promoter competition between the viral LTR and the internal promoter.160 Tat-Independent HIV-1 Vectors Tat is not required for expression of helper proteins but is essential for obtaining high levels of transcription from the viral LTR promoter of the gene transfer vector. Several studies have shown that titers are lower if Tat is not provided during virus stock production.133,145,149 To overcome this, hybrid promoters that use enhancer elements from other viruses such as cytomegalovirus immediate early promoter and Rous sarcoma virus LTR, instead of those present in the U3 of HIV-1, have been created.145, 149 These vectors appear to be nearly as efficient in terms of vector titer as the original Tat-dependent vectors that contained the wild type HIV-1 LTR. Although there are reports showing that Tat, in addition to its effect on transcription from the viral LTR, can also effect the efficiency of reverse transcription,161 studies using HIV1 vectors have not revealed this requirement.145,149 Despite the observation that the requirement for Tat can be overcome by using hybrid promoters, it may not be possible to entirely remove the TAR element due to the presence of sequences essential to ensure optimal

68

Lentiviral Vector Systems for Gene Transfer

packaging of vector RNA.53,162 Moreover, the R region, including TAR, is essential for reversetranscription. Rev-Independent HIV-1 Vectors The CTE from MPMV can substitute for Rev and RRE function in proviral clones, packaging plasmids and in gene transfer vectors (see above). To create a Rev-independent gene transfer vector, this small RNA element of about ~200 bp is usually inserted between the transgene expression cassette and the 3' LTR63,88,133,150,151,163-165 (Fig. 10). If the CTE is placed further upstream, say in place of the RRE, its function seems to be compromised. Some investigators have hypothesized that the CTE needs to be as close as possible to the pA signal (< 200 nucleotides) to exert its effect.165 A Rev-independent vector is ideal for transducing transdominant Rev-encoding transgenes into HIV-1 susceptible cells.133,150 The drawback of this vector is that recombination with the helper construct involving the gag/pol region could allow Rev-RRE independent expression of Gag-Pol proteins in the transduced target cells. Minimal Gene Transfer Vectors The two important reasons for attempting to create a minimal gene transfer vector are: 1) to decrease regions of homology between the gene transfer vector and packaging plasmid and 2) to increase the payload carrying capacity of the vector. It is believed that it may be possible to accommodate between 8 and 11 kb size of foreign sequence within an HIV-1 vector.47,48 Therefore, the smaller the size of the vector backbone, the larger the size of the insert that it can accommodate. One of the smallest vectors described to date 61 has about 550 bp of the 9.7 kb of the HIV-1 provirus genome. This vector contains the 5’LTR and the entire 5' untranslated region upstream of gag and about 40 nucleotides (nt) of gag. The 5' major splice donor site in the 5' untranslated region is mutated. The vector lacks the RRE and contains deletions in U3 and U5 regions of the 3’LTR. The pA functions in the U5 region is restored using the bovine growth hormone pA signal. This vector has approximately 50% of the titer of the wild-type vector. Insertion of the cPPT and CTS sequence in this vector (~170 nt) will probably improve the titer obtained with this minimal vector without significantly affecting the payload capacity. Most vectors used to date, in contrast, contain the 5’ splice donor, an extended gag (350 to 500 nt), the cPPT and the RRE since the addition of these features usually provides higher vector titers. Vectors with Woodchuck Post-Transcriptional Regulatory Element (WPRE) for Enhanced Transgene Expression An element with Rev and RRE like properties is the woodchuck post-transcriptional regulatory element (WPRE).166 While the WPRE can substitute for Rev and RRE function in a reporter construct that is widely used for assessing transport of intron containing messages, it is not known if this sequence can substitute for Rev and RRE in the context of gene transfer vectors or HIV-1 packaging constructs. Interestingly, experiments have revealed that addition of the 600 bp WPRE in retroviral and lentiviral vectors downstream of the transgene can increase expression by 5-8 fold 65,167(Fig. 10). This element may be of use to improve expression from weak promoters that may not function at optimal levels in some target cells.

Clinical Applications of HIV-1 Vectors The application of lentivirus vectors for the treatment of many genetic and infectious diseases is discussed in Chapter 9. Two recent publications showcase the utility of these vectors for treatment of β-thalassemia in mice168 and for prevention of neurodegeration in a primate model of parkinson’s disease.169 A possible intriguing application of HIV-1 vectors is for the treatment of AIDS caused by HIV-1. The HIV-1 vector can be considered to be a defective

HIV-1 Vector Systems

69

interfering (DI) virus. HIV-1 vectors can interfere with the replication of wild-type virus by several mechanisms. One mechanism is by competing and sequestering viral regulatory proteins170 and the other is by competing with wild-type RNA for encapsidation into the assembling virus particle.171 The presence of the vector in cells was found to interfere with the spread of the wild-type virus in culture. During the spread of the wild-type virus, the vector sequences are also likely to be mobilized to other cells where they can be expected to continue to exert their interfering phenotype. HIV-1 vectors can also be harnessed to deliver dominant negative proteins or RNA into HIV-1 susceptible cells. The trick to using HIV-1 vectors for delivering interfering proteins or antisense RNAs is to design packaging systems that will not be affected by expression of these anti-HIV therapeutics in the vector producing cells. Even with packaging systems currently in use, it should be possible to deliver antisense RNAs directed against the HIV-1 env region since most packaging systems use Env from a heterologous virus for production of vector stocks. Likewise one may be able to use gag antisense molecules in packaging systems using codonoptimized helper constructs as long as the antisense sequence does not overlap the gag portion still present in the gene transfer vector. Finally, one can use a Rev-independent packaging system where nucleo-cytoplasmic transport and expression of both helper and gene transfer vector RNAs is regulated by MPMV-CTE to deliver interfering Rev into HIV-1 susceptible cells.63,88,133,150,151 It is debatable if the use of HIV-1 based vectors for delivery of some of these therapeutic genes has any advantages over using an unrelated lentivirus (discussed elsewhere in this book) to achieve some of the same goals. A novel approach to exploit interactions between the wild-type virus and vector sequence has been suggested.58 This approach consists of expressing dominant negative proteins or antigenic epitopes of common infectious agents such as the hemagglutinin epitope in the HIV-1 vectors under control of the viral LTR promoter. When a wild-type virus infects the cell, Tat produced from the incoming virus, transactivates the vector leading to the synthesis of dominant-negative protein or antigen. Cytotoxic T- cells directed against expressed antigen on HIV1 infected cells, if present, would then be expected to clear the antigen bearing cells. Alternatively, the dominant negative protein would be expected to interfere with replication of HIV-1 in that cell.

Anticipated Developments in HIV-1 Based Packaging Systems and Possible Confounding Factors Long-Term and Cell-Type Specific Gene Expression The gene-expression in target cells transduced with MoMLV is subject to silencing as a result of methylation of vector DNA and/or deacetylation of histones in the promoter region.172-175 These modifications to DNA or histones can alter the chromatin structure resulting in gradual obtundation of gene expression. Studies in the case of HIV-1 vectors, on the other hand, seem to indicate that gene expression can persist for at least for six months or more in certain tissues such as neurons, muscle and liver cells.59,143,176,177 A recent study indicates that in embryonal stem cells, differentiation leads to rapid loss of gene expression in the context of MoMLV vectors but is only partially obtunded in the case of lentiviral vectors.178 Clearly, modifications to gene transfer vectors that can provide long-term gene expression is necessary to effect a permanent cure for many diseases. Such modifications to HIV-1 vectors could involve the introduction of boundary elements or insulators to improve long-term gene expression in target cells and thereby prevent gradual shut-down of transgene expression.179-181 Currently, most HIV-1 vectors express transgenes under control of strong viral promoters or constitutively active cellular promoters such as the phosphoglycerate kinase promoter or the

70

Lentiviral Vector Systems for Gene Transfer

human elongation factor EF1 α promoter. But in many situations it may be preferable to express physiological amounts of the transgene and/or express the transgene in only the tissue of interest i.e., expression that is regulated in a tissue/cell specific manner. One excellent example of this is a recent report using a lentivirus vector for erythroid-specific expression of human β-globin.168 In this vector the β-globin gene was expressed using the β-globin promoter together with the upstream regulatory regions consisting of DNAse I hypersensitive sites that are known to confer erythroid-specificity of gene expression. The entire transgene expression cassette was inserted in an inverse orientation to that of the vector. Such an orientation allowed the retention of splicing signals within the expression cassette leading to significantly higher levels of transgene expression than that observed with other vectors. Because the transgene expression was restricted to erythroid cells, there was no negative impact on vector titer due to an antisense effect in the producer cell. This vector provided therapeutic levels of β-globin expression in thallasemic mice. Reports such as this auger well for the development of lentivirus vectors containing cell-type specific promoters for expression of transgenes. Other modifications to HIV-1 vectors will be in the use of regulatable promoters that can be turned on or off using small molecules such as tetracycline,182 Rapamycin, estradiol or RU486.183-191 The drawback of the latter approach is that one will have to coexpress the protein that is required for gene regulation (e.g., tetracycline regulated transactivator) along with the transgene. It is not clear how these regulators will affect functioning of the cell or whether expression of these proteins will lead to immune-mediated elimination of the transduced cells. Such vectors will eventually allow precise control of transgene expression in vivo. Precise control of transgene expression may be of particular importance in the nervous system where overexpression of neurotransmitters could lead to neurological effects.169

Site-Specific Integration Due to the random nature of integration, lentiviruses have the some of the same disadvantages as other retroviruses, such as the possibility of either inactivating cellular genes (e.g., tumor suppressor genes) or activating or overexpression of other genes (e.g., oncogenes). Inadvertent activation of cellular genes, including oncogenes, can be avoided by using later generation of self-inactivating vectors which are devoid of promoter and enhancer elements in the viral 3’ LTR (see above). Alternatively, it may be possible to overcome the drawback of random integration by redirecting PICs to specific regions of the host chromosome by fusing sequence specific DNA-binding domains to the viral integrase.192,193

Packaging Cell Lines Most investigators currently use a transient transfection approach to produce vector stocks. A drawback of this procedure is that there may be a significant degree of rearrangement of the input DNAs. This can lead to contamination of vector stocks with defective vector genomes. To produce ‘clean’ vector stocks it may be preferable to derive well-characterized packaging cell lines. The major hurdle for the creation of packaging cell lines appears to be the toxicity of certain viral proteins. For instance, Vpr causes cells to arrest in G2 phase of the cell cycle115,116,194 and, therefore, would not be conducive for producing cell lines that express high levels of this protein. Likewise, the viral protease has been shown to also cleave cellular proteins, which could interfere with the establishment of cell lines that stably express viral Gag-Pol proteins.195-197 VSV-G protein expression is also toxic to cells. Although cell lines that constitutively express viral packaging proteins have been described,133,198,199 more recent approaches to create lentivirus packaging cell lines resort to the use of inducible or regulatable promoters to overcome potential toxic effects of viral proteins.105,200 Availability of second generation packaging cell lines using advanced packaging and gene transfer vectors201will enable the eventual testing of HIV1 vectors in a clinical setting.

HIV-1 Vector Systems

71

Immune Response to Transgene Encoded Products One of the biggest hurdles to gene therapy may not be vector related issues, most of which will likely be solved, but the possible elimination of transduced cells if the expressed transgene is perceived as foreign by the host. Thus successful gene therapy will also involve use of therapeutic strategies that can overcome this hurdle. This could involve the coexpression of viral protein derived from HIV-1 or other viruses that are known to interfere with antigen presentation. Examples of proteins with such activities include Nef141 and/or Tat202,203 of HIV-1.

Safety Issues Safety issues for the deployment of HIV-1 vectors for gene therapy in humans are discussed further in Chapter 8. Some aspects of vector safety are elaborated upon here. Recombination between helper and gene transfer vector molecules can occur at the level of input DNAs in the producer cells or can occur during reverse transcription as consequence of copackaging of vector and helper RNAs or vector and cellular RNAs in the same virus particle. One favored site of recombination between RNA molecules during reverse-transcription appears to be in the poly A tract of the helper RNA and the sequences upstream of the 3' U3 and PPT in the gene transfer vector.155 The upstream recombination appears to occur, as anticipated, in the gag region, which results in the joining of the 5’LTR and packaging sequence of the gene transfer vector with the gag ORF of the helper plasmid. Whatever the mechanism of recombination, the consequence is the transduction of molecules into the target cells that may have an intact open reading frames for one or more viral proteins such as GagPol and/or Tat. Can these cells, harboring gag/pol sequences, produce viral proteins? Will this lead to an immune response in the transplant recipient against the proteins expressed by the recombinant vector? In other words, will the person be rendered HIV positive as deduced from antibody assays? Another concern is that the reverse-transcription of the vector RNA is error prone. It is likely that some of the vector genomes introduced into target cells will contain deleterious mutations. These mutations (e.g., deletions, insertions or duplications) can lead to inactivation of the transgene or synthesis of aberrant or nonfunctional proteins. What is the effect of introducing these altered DNA sequences into target cells? Finally, vector genomes can interact with wild-type HIV and possibly with endogenous retroviral sequences or elements. What are the consequences of the interactions between vector and wild-type HIV-1 or endogenous retroviruses? Will this lead to the mobilization of vector or host-sequences from one cell to another? Can novel viruses emerge as a result of recombination between vector and wildtype sequences or endogenous retroviral sequences? The answers to many of these questions can only be ascertained by vigorous and careful evaluation of the vectors in large animals. Preliminary studies from some investigators appear to indicate that HIV-1 based vectors are safe in primates.204 But more careful studies are clearly needed. How does one evaluate the safety of lentivirus vector stocks? The Center for Biologics Evaluation and Research (CBER) of the Food and Drug Administration (FDA) recommends that 5% of each vector supernatant lot and 108 or 1% of the end of production cells (which ever is less) be tested for RCR for MoMLV vectors.205 There is, at present, no agreement on what standards to apply for lentivirus vectors, particularly those based on HIV-1. It would need to equal, if not exceed, those standards designed to detect RCRs in MoMLV vector stocks. Also, there are no agreed upon assays to detect RCRs in HIV-1 vector stocks. The methods commonly used are: 1. Infection of target cells (including phytohemagglutinin-stimulated human peripheral blood mononuclear cells) with vector stocks and assaying supernatant periodically for HIV-1 p24 over a two to four week culture period.204,206 2. Infection of indicator cell lines harboring lacz gene (Magi-cell assay)207 or puromycin resistance gene under control of HIV-1 LTR.155

Lentiviral Vector Systems for Gene Transfer

72

3. Marker mobilization assays.155 The first method (assay for p24) can only detect replication competent HIV while the latter two (infection of indicator cell lines and marker mobilization) can detect partial recombinants. The Magi-cell assay requires the recombinant to synthesize the HIV-1 Tat protein and thereby turn on lacz expression from the HIV-1 LTR. Marker mobilization detects recombinants that can synthesize Gag-Pol , Tat, Rev and Env. The last method can be modified to detect partial recombinants that can synthesize Gag-Pol after appropriate modifications to the assay.155 Once the safety concerns are addressed, then the first steps towards deploying HIV-1 vectors for clinical use can be taken.

Conclusions There has been tremendous progress in the development of HIV-1 based vectors in the last five years in terms of their safety and efficacy to deliver genes to a wide variety of tissues both in vitro and in vivo. It is a vector system that shows great promise. However, the stigma associated with HIV-1 based vectors, the lack of standardized assays for detection of recombinant replication competent HIV-1 emerging during virus stock production and the risk of immune-mediated rejection of cells bearing a foreign transgene by the host are significant hurdles that need to be addressed before clinical trials using HIV-1 vectors can be undertaken.

Acknowledgements I wish to offer my apologies to those authors whose work was not cited due to an oversight or because of space considerations. I wish to thank Gary Buchschacher, Jr. for his immense patience, support and constructive criticisms during the writing of this chapter. I was supported by grants from the American Foundation for AIDS Research and the National Institutes of Health for my research studies cited in this chapter.

References 1. Coffin J.M., Stephen HH, Varmus HE. Retroviruses: CSHL; 1997. 2. Feinberg MB, Baltimore D, Frankel AD. The role of Tat in the human immunodeficiency virus life cycle indicates a primary effect on transcriptional elongation. Proc Natl Acad Sci USA 1991; 88(9):4045-4049. 3. Gaynor RB. Regulation of HIV-1 gene expression by the transactivator protein Tat. Curr Top Microbiol Immunol 1995; 193:51-77. 4. Jeang KT, Xiao H, Rich EA. Multifaceted Activities of the HIV-1 Transactivator of Transcription, Tat. J Biol Chem 1999; 274(41):28837-28840. 5. Jones KA. Taking a new TAK on tat transactivation [comment]. Genes Dev 1997; 11(20):2593-2599. 6. Purcell DF, Martin MA. Alternative splicing of human immunodeficiency virus type 1 mRNA modulates viral protein expression, replication, and infectivity. J Virol 1993; 67(11):6365-6378. 7. Arrigo SJ, Weitsman S, Zack JA et al. Characterization and expression of novel singly spliced RNA species of human immunodeficiency virus type 1. J Virol 1990; 64(9):4585-4588. 8. Guatelli JC, Gingeras TR, Richman DD. Alternative splice acceptor utilization during human immunodeficiency virus type 1 infection of cultured cells. J Virol 1990; 64(9):4093-4098. 9. Muesing MA, Smith DH, Cabradilla CD et al. Nucleic acid structure and expression of the human AIDS/lymphadenopathy retrovirus. Nature 1985; 313(6002):450-458. 10. Robert-Guroff M, Popovic M, Gartner S et al. Structure and expression of tat-, rev-, and nefspecific transcripts of human immunodeficiency virus type 1 in infected lymphocytes and macrophages. J Virol 1990; 64(7):3391-3398. 11. Schwartz S, Felber BK, Benko DM et al. Cloning and functional analysis of multiply spliced mRNA species of human immunodeficiency virus type 1. J Virol 1990; 64(6):2519-2529. 12. Schwartz S, Felber BK, Fenyo EM et al. Env and Vpu proteins of human immunodeficiency virus type 1 are produced from multiple bicistronic mRNAs. J Virol 1990; 64(11):5448-5456. 13. Schwartz S, Felber BK, Pavlakis GN. Expression of human immunodeficiency virus type 1 vif and vpr mRNAs is Rev-dependent and regulated by splicing. Virology 1991; 183(2):677-686.

HIV-1 Vector Systems

73

14. Fornerod M, Ohno M, Yoshida M et al. CRM1 is an export receptor for leucine-rich nuclear export signals [see comments]. Cell 1997; 90(6):1051-1060. 15. Hammarskjold ML, Heimer J, Hammarskjold B et al. Regulation of human immunodeficiency virus env expression by the rev gene product. J Virol 1989; 63(5):1959-1966. 16. Malim MH, Hauber J, Le SY et al. The HIV-1 rev trans-activator acts through a structured target sequence to activate nuclear export of unspliced viral mRNA. Nature 1989; 338(6212):254-257. 17. Malim MH, Tiley LS, McCarn DF et al. HIV-1 structural gene expression requires binding of the Rev trans- activator to its RNA target sequence. Cell 1990; 60(4):675-683. 18. Salahuddin SZ, Nakamura S, Biberfeld P et al. Angiogenic properties of Kaposi’s sarcoma-derived cells after long-term culture in vitro. Science 1988; 242(4877):430-433. 19. Schwartz S, Felber BK, Pavlakis GN. Distinct RNA sequences in the gag region of human immunodeficiency virus type 1 decrease RNA stability and inhibit expression in the absence of Rev protein. J Virol 1992; 66(1):150-159. 20. Schwartz S, Campbell M, Nasioulas G et al. Mutational inactivation of an inhibitory sequence in human immunodeficiency virus type 1 results in Rev-independent gag expression. J Virol 1992; 66(12):7176-7182. 21. Arrigo SJ, Chen IS. Rev is necessary for translation but not cytoplasmic accumulation of HIV-1 vif, vpr, and env/vpu 2 RNAs. Genes Dev 1991; 5(5):808-819. 22. Butsch M, Hull S, Wang Y et al. The 5' RNA terminus of spleen necrosis virus contains a novel posttranscriptional control element that facilitates human immunodeficiency virus Rev/RRE-independent Gag production. J Virol 1999; 73(6):4847-4855. 23. Lu XB, Heimer J, Rekosh D et al. U1 small nuclear RNA plays a direct role in the formation of a rev- regulated human immunodeficiency virus env mRNA that remains unspliced. Proc Natl Acad Sci USA 1990; 87(19):7598-7602. 24. Wills JW, Craven RC. Form, function, and use of retroviral gag proteins [editorial]. Aids 1991; 5(6):639-654. 25. Smith AJ, Cho MI, Hammarskjold ML et al. Human immunodeficiency virus type 1 Pr55gag and Pr160gag-pol expressed from a simian virus 40 late replacement vector are efficiently processed and assembled into viruslike particles. J Virol 1990; 64(6):2743-2750. 26. Gheysen D, Jacobs E, de Foresta F et al. Assembly and release of HIV-1 precursor Pr55gag viruslike particles from recombinant baculovirus-infected insect cells. Cell 1989; 59(1):103-112. 27. Delchambre M, Gheysen D, Thines D et al. The GAG precursor of simian immunodeficiency virus assembles into virus- like particles. Embo J 1989; 8(9):2653-2660. 28. Jacobs E, Gheysen D, Thines D et al. The HIV-1 Gag precursor Pr55gag synthesized in yeast is myristoylated and targeted to the plasma membrane. Gene 1989; 79(1):71-81. 29. Karacostas V, Wolffe EJ, Nagashima K et al. Overexpression of the HIV-1 gag-pol polyprotein results in intracellular activation of HIV-1 protease and inhibition of assembly and budding of virus-like particles. Virology 1993; 193(2):661-671. 30. Karacostas V, Nagashima K, Gonda MA et al. Human immunodeficiency virus-like particles produced by a vaccinia virus expression vector. Proc Natl Acad Sci USA 1989; 86(22):8964-8967. 31. Hu SL, Travis BM, Garrigues J et al. Processing, assembly, and immunogenicity of human immunodeficiency virus core antigens expressed by recombinant vaccinia virus. Virology 1990; 179(1):321-329. 32. Shioda T, Shibuta H. Production of human immunodeficiency virus (HIV)-like particles from cells infected with recombinant vaccinia viruses carrying the gag gene of HIV. Virology 1990; 175(1):139-148. 33. Haffar O, Garrigues J, Travis B et al. Human immunodeficiency virus-like, nonreplicating, gag-env particles assemble in a recombinant vaccinia virus expression system. J Virol 1990; 64(6):2653-2659. 34. Mergener K, Facke M, Welker R et al. Analysis of HIV particle formation using transient expression of subviral constructs in mammaliancells. Virology 1992; 186(1):25-39. 35. Bryant M, Ratner L. Myristoylation-dependent replication and assembly of human immunodeficiency virus 1. Proc Natl Acad Sci USA 1990; 87(2):523-527. 36. Smith AJ, Srinivasakumar N, Hammarskjold ML et al. Requirements for incorporation of Pr160gagpol rom human immunodeficiency virus type 1 into virus-like particles. J Virol 1993; 67(4):2266-2275.

74

Lentiviral Vector Systems for Gene Transfer

37. Srinivasakumar N, Hammarskjold ML, Rekosh D. Characterization of deletion mutations in the capsid region of human immunodeficiency virus type 1 that affect particle formation and Gag- Pol precursor incorporation. J Virol 1995; 69(10):6106-6114. 38. Wang CT, Stegeman-Olsen J, Zhang Y et al. Assembly of HIV GAG-B-galactosidase fusion proteins into virus particles. Virology 1994; 200(2):524-534. 39. Park J, Morrow CD. The nonmyristylated Pr160gag-pol polyprotein of human immunodeficiency virus type 1 interacts with Pr55gag and is incorporated into viruslike particles. J Virol 1992; 66(11):6304-6313. 40. Paxton W, Connor RI, Landau NR. Incorporation of Vpr into human immunodeficiency virus type 1 virions: requirement for the p6 region of gag and mutational analysis. J Virol 1993; 67(12):7229-7237. 41. Selig L, Pages J, Tanchou V et al. Interaction with the p6 domain of the gag precursor mediates incorporation into virions of vpr and vpx proteins from primate lentiviruses [In Process Citation]. J Virol 1999; 73(1):592-600. 42. Muller B, Tessmer U, Schubert U et al. Human immunodeficiency virus type 1 vpr protein is incorporated into the virion in significantly smaller amounts than gag and Is phosphorylated in infected cells [In Process Citation]. J Virol 2000; 74(20):9727-9731. 43. Khan MA, Aberham C, Kao S et al. Human immunodeficiency virus type 1 Vif protein is packaged into the nucleoprotein complex through an interaction with viral genomic RNA. J Virol 2001; 75(16):7252-7265. 44. Charneau P, Mirambeau G, Roux P et al. HIV-1 reverse transcription. A termination step at the center of the genome. J Mol Biol 1994; 241(5):651-662. 45. Follenzi A, Ailles LE, Bakovic S et al. Gene transfer by lentiviral vectors is limited by nuclear translocation and rescued by HIV-1 pol sequences. Nat Genet 2000; 25(2):217-222. 46. Zennou V, Petit C, Guetard D et al. HIV-1 genome nuclear import is mediated by a centralDNA flap. Cell 2000; 101(2):173-185. 47. Trono D, Baltimore D. A human cell factor is essential for HIV-1 Rev action. Embo J 1990; 9(12):4155-4160. 48. Verma IM, Somia N. Gene therapy—promises, problems and prospects. Nature 1997; 389(6648):239242. 49. Clever J, Sassetti C, Parslow TG. RNA secondary structure and binding sites for gaggene products in the 5' packaging signal of human immunodeficiency virus type 1. J Virol 1995; 69(4):2101-2109. 50. Das AT, Klaver B, Klasens BI et al. Aconserved hairpin motif in the R-U5 region of the human immunodeficiency virus type 1 RNA genome is essential for replication. J Virol 1997; 71(3):2346-2356. 51. McBride MS, Panganiban AT. The human immunodeficiency virus type 1 encapsidation site is a multipartite RNA element composed of functional hairpin structures [published erratum appears in J Virol 1997 Jan;71(1):858]. J Virol 1996; 70(5):2963-2973. 52. McBride MS, Panganiban AT. Position dependence of functional hairpins important for human immunodeficiency virus type 1 RNA encapsidation in vivo. J Virol 1997; 71(3):2050-2058. 53. Helga-Maria C, Hammarskjold ML, Rekosh D. An intact TAR element and cytoplasmic localization are necessary for efficient packaging of human immunodeficiency virus type 1 genomic RNA. J Virol 1999; 73(5):4127-4135. 54. Clever JL, Parslow TG. Mutant human immunodeficiency virus type 1 genomes with defects in RNA dimerization or encapsidation. J Virol 1997; 71(5):3407-3414. 55. Luban J, Goff SP. Mutational analysis of cis-acting packaging signals in human immunodeficiency virus type 1 RNA. J Virol 1994; 68(6):3784-3793. 56. Bender MA, Palmer TD, Gelinas RE et al. Evidence that the packaging signal of Moloney murine leukemia virus extends into the gag region. J Virol 1987; 61(5):1639-1646. 57. Parolin C, Dorfman T, Palu G et al. Analysis in human immunodeficiency virus type 1 vectors of cis-acting sequences that affect gene transfer into human lymphocytes. J Virol 1994; 68(6):3888-3895. 58. Buchschacher GJ, Panganiban AT. Human immunodeficiency virus vectors for inducible expression of foreign genes. J Virol 1992; 66(5):2731-2739. 59. Naldini L, Blomer U, Gallay P et al. In vivo gene delivery and stable transduction of nondividing cells by a lentiviral vector. Science 1996; 272(5259):263-267.

HIV-1 Vector Systems

75

60. Nasioulas G, Zolotukhin AS, Tabernero C et al. Elements distinct from human immunodeficiency virus type 1 splice sites are responsible for the Rev dependence of env mRNA. J Virol 1994; 68(5):2986-2993. 61. Cui Y, Iwakuma T, Chang LJ. Contributions of viral splice sites and cis-regulatory elements to lentivirus vector function. J Virol 1999; 73(7):6171-6176. 62. Rizvi TA, Panganiban AT. Simian immunodeficiency virus RNA is efficiently encapsidated by human immunodeficiency virus type 1 particles. J Virol 1993; 67(5):2681-2688. 63. Srinivasakumar N, Schuening FG. Novel Tat-encoding bicistronic human immunodeficiency virus type 1-based gene-transfer vectors for high-level transgene expression. J Virol 2000; 74:6659-6668. 64. Parolin P, Taddeo B, Palu G et al. Use of cis- and trans-Acting regulatory sequences to improve expression of human immunodeficiency virus vectors in human lymphocytes. Virology 1996; 222:415-422. 65. Salmon P, Kindler V, Ducrey O et al. High-level transgene expression in human hematopoietic progenitors and differentiated blood lineages after transduction with improved lentiviral vectors [In Process Citation ] Blood 2000; 96(10):3392-3398. 66. Miyoshi H, Takahashi M, Gage FH et al. Stable and efficient gene transfer into the retina using an HIV-based lentiviral vector. Proc Natl Acad Sci USA 1997; 94(19):10319-10323. 67. DeZazzo JD, Kilpatrick JE, Imperiale MJ. Involvement of long terminal repeat U3 sequences overlapping the transcription control region in human immunodeficiency virus type 1 mRNA 3' end formation. Mol Cell Biol 1991; 11(3):1624-1630. 68. DeZazzo JD, Scott JM, Imperiale MJ. Relative roles of signals upstrea of AAUAAA and promoter proximity in regulation of human immunodeficiency virus type 1 mRNA 3' end formation. Mol Cell Biol 1992; 12(12):5555-5562. 69. Valsamakis A, Zeichner S, Carswell S et al. The human immunodeficiency virus type 1 polyadenylylation signal: a 3' long terminal repeat element upstream of the AAUAAA necessary for efficient polyadenylylation. Proc Natl Acad Sci USA 1991; 88(6):2108-2112. 70. Iwakuma T, Cui Y, Chang LJ. Self-inactivating lentiviral vectors with U3 and U5 modifications. Virology 1999; 261(1):120-132. 71. Zufferey R, Dull T, Mandel RJ et al. Self-inactivating lentivirus vector for safe and efficient in vivo gene delivery. J Virol 1998; 72(12):9873-9880. 72. Mochizuki H, Schwartz JP, Tanaka K et al. High-titer human immunodeficiency virus type 1-based vector systems for gene delivery into nondividing cells. J Virol 1998; 72(11):8873-8883. 73. Yee JK, Miyanohara A, LaPorte P et al. A general method for the generation of high-titer, pantropic retroviral vectors: highly efficient infection of primary hepatocytes. Proc Natl Acad Sci USA 1994; 91(20):9564-9568. 74. Kiem HP, McSweeney PA, Bruno B et al. Improved gene transfer into canine hematopoietic repopulating cells using CD34-enriched marrow cells in combination with a gibbon ape leukemia virus-pseudotype retroviral vector. Gene Ther 1999; 6(6):966-972. 75. Stitz J, Buchholz CJ, Engelstadter M et al. Lentiviral vectors pseudotyped with envelope glycoproteins derived from gibbon ape leukemia virus and murine leukemia virus 10A1. Virology 2000; 273(1):16-20. 76. Chu TH, Dornburg R. Retroviral vector particles displaying the antigen-binding site of an antibody enable cell-type-specific gene transfer. J Virol 1995; 69(4):2659-2663. 77. Chu TH, Dornburg R. Toward highly efficient cell-type-specific gene transfer with retroviral vectors displaying single-chain antibodies. J Virol 1997; 71(1):720-725. 78. Jiang A, Chu TH, Nocken F et al. Cell-type-specific gene transfer into human cells with retroviral vectors that display single-chain antibodies. J Virol 1998; 72(12):10148-10156. 79. Martinez I, Dornburg R. Mapping of receptor binding domains in the envelope protein of spleen necrosis virus. J Virol 1995; 69(7):4339-4346. 80. Martinez I, Dornburg R. Mutational analysis of the envelope protein of spleen necrosis virus. J Virol 1996; 70(9):6036-6043. 81. Naviaux RK, Costanzi E, Haas M et al. The pCL vector system: rapid production of helper-free, high-titer, recombinant retroviruses. J Virol 1996; 70(8):5701-5705. 82. Srinivasakumar N et al. (unpublished observations).

76

Lentiviral Vector Systems for Gene Transfer

83. Batra RK, Olsen JC, Hoganson DK et al. Retroviral gene transfer is inhibited by chondroitin sulfate proteoglycans/glycosaminoglycans in malignant pleural effusions. J Biol Chem 1997; 272(18):11736-11743. 84. Le Doux JM, Morgan JR, Snow RG et al. Proteoglycans secreted by packaging cell lines inhibit retrovirus infection. J Virol 1996; 70(9):6468-6473. 85. Le Doux JM, Morgan JR, Yarmush ML. Removal of proteoglycans increases efficiency of retroviral gene transfer. Biotechnol Bioeng 1998; 58(1):23-34. 86. Le Doux JM, Morgan JR, Yarmush ML. Differential inhibition of retrovirus transduction by proteoglycans and free glycosaminoglycans. Biotechnol Prog 1999; 15(3):397-406. 87. Reiser J. Production and concentration of pseudotyped HIV-1-based gene transfer vectors. Gene Ther 2000; 7(11):910-913. 88. Srinivasakumar N, Schuening F. A lentivirus packaging system based on alternative RNA transport mechanisms to express helper and gene transfer vector RNAs and its use to study the requirement of accessory proteins for particle formation and gene delivery. J Virol 1999; 73:9589-9598. 89. Albini A, Fontanini G, Masiello Let al. Angiogenic potential in vivo by Kaposi’s sarcoma cell-free supernatants and HIV-1 tat product: inhibition of KS-like lesions by tissue inhibitor of metalloproteinase-2. Aids 1994; 8(9):1237-1244. 90. Albini A, Barillari G, Benelli R et al. Angiogenic properties of human immunodeficiency virus type 1 Tat protein. Proc Natl Acad Sci USA 1995; 92(11):4838-4842. 91. Albini A, Benelli R, Presta M et al. HIV-tat protein is a heparin-binding angiogenic growthfactor. Oncogene 1996; 12(2):289-297. 92. Ensoli B, Barillari G, Salahuddin SZ et al. Tat protein of HIV-1 stimulates growth of cells derived from Kaposi’s sarcoma lesions of AIDS patients. Nature 1990; 345(6270):84-86. 93. Ensoli B, Barillari G, Buonaguro L et al. Molecular mechanisms in the pathogenesis of AIDSassociated aposi’s sarcoma. Adv Exp Med Biol 1991; 303:27-38. 94. Ensoli B, Buonaguro L, Barillari G et al. Release, uptake, and effects of extracellular human immunoeficiency virus type 1 Tat protein on cell growth and viral transactivation. J Virol 1993; 67(1):277287. 95. Ensoli B, Gendelman R, Markham P et al. Synergy between basic fibroblast growth factor and HIV-1 Tat protein in induction of Kaposi’s sarcoma. Nature 1994; 371(6499):674-680. 96. Chirmule N, Than S, Khan SA et al. Human immunodeficiency virus Tat induces functional unresponsiveness in T cells. J Virol 1995; 69(1):492-498. 97. Ott M, Emiliani S, Van Lint C et al. Immune hyperactivation of HIV-1-infected T cells mediated by Tat and the CD28 pathway. Science 1997; 275(505):1481-1485. 98. Ott M, Lovett JL, Mueller L et al. Superinduction of IL-8 in T cells by HIV-1 Tat protein is mediated through NF-kappaB factors. J Immunol 1998; 160(6):2872-2880. 99. Puri RK, Leland P, Aggarwal BB. Constitutive expression of human immunodeficiency virus type 1 tat gene inhibits intrleukin 2 and interleukin 2 receptor expression in a human CD4+ T lymphoid (H9) cell line. AIDS Res Hum Retroviruses 1995; 11(1):31-40. 100. Rautonen N, Rautonen J, MartinNL et al. HIV-1 Tat induces cytokine synthesis by uninfected mononuclear cells [letter]. Aids 1994; 8(10):1504-1506. 101. Reinhold D, Wrenger S, Bank U et al. CD26 mediates the action of HIV-1 Tat protein on DNA synthesis and cytokine production in U937 cells. Immunobiology 1996; 195(1):119-128. 102. Subramanyam M, Gutheil WG, Bachovchin WW et al. Mechanism of HIV-1 Tat induced inhibition f antigen-specific T cell responsiveness. J Immunol 1993; 150(6):2544-2553. 103. Vacca A, Farina M, Maroder M et al. Human immunodeficiency virus type-1 tat enhances interlekin2 promoter activity through synergism with phorbol ester and calcium-mediated activation of the NF-AT cis-regulatory motif. Biochem Biophys Res Commun 1994; 205(1):467-474. 104. Verhoef K, Bauer M, Meyerhans A et al. On the role of the second coding exon of the HIV-1 Tat protein in virus replication and MHC class I downregulation. AIDS Res Hum Retroviruses 1998; 14(17):1553-1559. 105. Kaul M, Yu H, Ron Y et al. Regulated lentiviral packaging cell line devoid of most viral cis- acting sequences. Virology 1998; 249(1):167-174. 106. Gasmi M, Glynn J, Jin MJ et al. Requirements for efficient production and transduction of human immunodeficiency vius type 1-based vectors. J Virol 1999; 73(3):1828-1834.

HIV-1 Vector Systems

77

107. Madani N, Kabat D. Cellular and viral specificities of human immunodeficiency virus type 1 vif protein. J Virol 2000; 74(13):5982-5987. 108. Simon JH, Miller DL, Fouchier RA et al. The regulation of primate immunodeficiency virus infectivity by Vif is cell species restricted: a role for Vif in determining virus host range and crossspecies transmission. Embo J 1998; 17(5):1259-1267. 109. von Schwedler U, Song J, Aiken C et al. Vif is crucial for human immunodeficiency virus type 1 proviral DNA synthesis in infected cells. J Virol 1993; 67(8):4945-4955. 110. Strebel K, Klimkait T, Maldarelli F et al. Molecular and biochemical analyses of human immunodeficiency virus type 1 vpu protein. J Virol 1989; 63(9):3784-3791. 111. Schwartz MD, Geraghty RJ, Panganiban AT. HIV-1 particle release mediated by Vpu is distinct from that mediated by p6. Virology 1996; 224(1):302-309. 112. Gottlinger HG, Dorfman T, Sodroski JG et al. Effect of mutations affecting the p6 gag protein on human immunodeficiency virus particle release. Proc Natl Acad Sci USA 1991;88(8):3195-3199. 113. Kerkau T, Bacik I, Bennink JR et al. The human immunodeficiency virustype 1 (HIV-1) Vpu protein interferes with an early step in the biosynthesis of major histocompatibility complex (MHC) class I molecules. J Exp Med 1997; 185(7):1295-1305. 114. Willey RL, Maldarelli F, Martin MA et al. Human immunodeficiency virus type 1 Vpu protein induces rapid degradation of CD4. J Virol 1992; 66(12):7193-7200. 115. Bartz SR, Rogel ME, Emerman M. Human immunodficiency virus type 1 cell cycle control: Vpr is cytostatic and mediates G2 accumulation by a mechanism which differs from DNA damage checkpoint control. J Virol 1996; 70(4):2324-2331. 116. Rogel ME, Wu LI, Emerman M. The human immunodeficiency virus type 1 vpr gene prevents cell proliferation during chronic infection. J Virol 1995; 69(2):882-888. 117. Re F, Braaten D, Franke EK et al. Human immunodeficiency virus type 1 Vpr arrests the cell cycle in G2 by inhibiting the activation of p34cdc2-cyclin B. J Virol 1995; 69(11):6859-6864. 118. Goh WC, Rogel ME, Kinsey CM et al. HIV-1 Vpr increases viral expression by manipulation of the cell cycle: a mechanism for selection of Vpr in vivo. Nat Med 1998; 4(1):65-71. 119. Heinzinger NK, Bukinsky MI, Haggerty SA et al. The Vpr protein of human immunodeficiency virus type 1 influences nuclear localization of viral nucleic acids in nondividing host cells. Proc Natl Acad Sci USA 1994; 91(15):7311-7315. 120.Popov S, Rexach M, Zybarth G et al. Viral protein R regulates nuclear import of the HIV-1 preintegration complex. Embo J 1998; 17(4):909-917. 121. Popov S, Rexach M, Ratner L et al. Viral protein R regulates docking of the HIV-1 preintegration complex to the nuclear pore complex. J Biol Chem 1998; 273(21):13347-13352. 122. Vodicka MA, Koepp DM, Silver PA et al. HIV-1 Vpr interacts with the nuclear transport pathway to promote macrophage infection. Genes Dev 1998; 2(2):175-185. 123. Stewart SA, Poon B, Jowett JB et al. Human immunodeficiency virus type 1 Vpr induces apoptosis following cell cycle rrest. J Virol 1997; 71(7):5579-5592. 124. Stewart SA, Poon B, Jowett JB et al. Lentiviral delivery of HIV-1 Vpr protein induces apoptosi in transformed cells. Proc Natl Acad Sci USA 1999; 96(21):12039-12043. 125. Stewart SA, Poon B, Song JY et al. Human immunodeficiency virus type 1 vpr induces apoptosis through caspase activation. J Virol 2000; 74(7):3105-3111. 126. Chinnasamy D, Chinnasamy N, Enriquez MJ et al. Lentiviral-mediated gene transfer into human lymphocytes: role f HIV-1 accessory proteins. Blood 2000; 96(4):1309-1316. 127. Anderson SJ, Lenburg M, Landau NR et al. The cytoplasmic domain of CD4 is sufficient for its down-regulation from the cell surface by human immunodeficiency virus type 1 Nef [published erratum appears in J Virol 1994 Jul; 68(7):4705]. J Virol 1994; 68(5):3092-3101. 128. Aiken C, Konner J, Landau NR et al. Nef induces CD4 endoctosis: requirement for a critical dileucine motif in the membrane-proximal CD4 cytoplasmic domain. Cell 1994; 76(5):853-864. 129. Garcia JV, Miller AD. Downregulation of cell surface CD4 by nef. Res Virol 1992; 143(1):52-55. 130. Garcia JV, Miller AD. Serine phosphorylation-independent downregulation of cell-surface CD4 by nef. Nature 1991; 350(6318):508-511. 131. Luo T, Livingston RA, Garcia JV. Infectivity enhancement by human immunodeficiency virus type 1 Nef is independent of its association with a cellular serine/threonine kinase. J Virol 1997; 71(12):9524-9530.

78

Lentiviral Vector Systems for Gene Transfer

132. Luo T, Douglas JL, Livingston RL et al. Infectivity enhancement by HIV-1 Nef is dependent on the pathway of virus entry: implications for HIV-based gene transfer systems. Virology 1998; 241(2):224-233. 133. Srinivasakumar N, Chazal N, Helga-Maria C et al. The effect of viral regulatory protein expression on gene delivery by human immunodeficiency virus type 1 vectors produced in stable packaging cell lines. J Virol 1997; 71(8):5841-5848. 134. Schwrtz O, Marechal V, Danos O et al. Human immunodeficiency virus tpe 1 Nef increases the efficiency of reverse transcription in the infected cell. J Virol 1995; 69(7):4053-4059. 135. Aiken C, Trono D. Nef stimulates human immunodeficiency virus type 1 proviral DNA synthesis. J Virol 1995; 69(8):5048-5056. 136. Aiken C. Pseudotyping human immunodeficiency virus type 1 (HIV-1) by the glycoprotein of vesicular stomatitis virus targets HIV-1 entry to an endocytic pathway and suppresses both the requirement for Nef and the sensitivity to cyclosporin A. J Virol 1997; 71(8):5871-5877. 137. Chazal N, Singer G, Aiken C et al. Human Immunodeficiency Virus Type 1 Particles Pseudotyped with Envelope Proteins That Fuse at Low pH No Longer Require Nef or Optimal Infectivity. J Virol 2001; 75(8):4014-4018. 138. Hanna Z Kay DG, Rebai N et al. Nef harbors a major determinant of pathogenicity for an AIDSlike disease induced by HIV-1 in transgenic mice. Cell 1998; 95(2):163-175. 139. Delwart EL, Buchschacher GL, Jr., Freed EO et al. Analysis of HIV-1 envelope mutants and pseudotyping of replication- defective HIV-1 vectors by genetic complementation. AIDS Res Hum Retroviruses 1992; 8(9):1669-1677. 140. Page KA, Landau NR, Littman DR. Construction and use of a human immunodeficiency virus vector foranalysis of virus infectivity. J Virol 1990; 64(11):5270-5276. 141. Collins KL, Chen BK, Kalams SA et al. HIV-1 Nef protin protects infected primary cells against killing by cytotoxic lymphocytes. Nature 1998; 391(6665):397-401. 142. Chen BK, Feinberg MB, Baltimore D. The kappaB sites in the human immunodeficiency virus type 1 long terminal repeat enhance virus replication yet are not absolutely required for viral growth. J Virol 1997; 71(7):5495-5504. 143. Kafri T, Blomer U, Peerson DA et al. Sustained expression of genes delivered directly into liver and muscle by lentiviral vectors. Nat Genet 1997; 17(3):314-317. 144. Zufferey R, Nagy D, Mandel RJ et al. Multiply attenuated lentiviral vector achieves efficient gene delivery in vivo. Nat Biotechnol 1997; 15(9):871-875. 145. Dull T, Zufferey R, Kelly M et al. A third-generation lentivirus vector with a conditional packaging system. J Virol 1998; 72(11):8463-8471. 146. Ernst RK, Bray M, Rekosh D et al. A structured retroviral RNA element that mediates nucleocytoplasmic export of intron-containing RNA. Mol Cell Biol 1997; 17(1):135-144. 147. Ernst RK, Bray M, Rekosh D et al. Secondary structure and mutational analysis of the MasonPfizer monkey virus RNA constitutive transport element. Rna 1997; 3(2):210-222. 148. Bray M, Prasad S, Dubay JW et al. A small element from the Mason-Pfizer monkey virus genome makes human immunodeficiency virus type 1 expression and replication Rev- independent. Proc Natl Acad Sci USA 1994; 91(4):1256-1260. 149. Kim VN, Mitrophanous K, Kingsman SM et al. Minimal requirement for a lentivirus vector based on human immunodeficiency virus type 1. J Virol 1998; 72(1):811-816. 150. Mautino MR, Ramsey WJ, Reiser J et al. Modified human immunodeficiency virus-based lentiviral vectors display decreased sensitivity to trans-dominant Rev. Hum Gene Ther 2000;11(6):895-908. 151. Mautino MR, Keiser N, Morgan RA. Improved titers of HIV-based lentiviral vectors using the SRV-1 constitutive transport element. Gene Ther 2000; 7(16):1421-1424. 152. Zolotukhin AS, Valentin A, Pavlakis GN et al. Continuous propagation of RRE(-) and Rev(-)RRE(-) human immunodeficiency virus type 1 molecular clones containing a cis-acting element of simian retrovirus type 1 in human peripheral blood lymphocytes. J Virol 1994; 68(2):7944-7952. 153. Kotsopoulou E, Kim VN, Kingsman AJ et al. A Rev-independent human immunodeficiency virus type 1 (HIV-1)-based vector that exploits a codon-optimized HIV-1 gag-pol gene. J Virol 2000; 74(10):4839-4852. 154. White SM, Renda M, Nam NY et al. Lentivirus vectors using human and simian immunodeficiency virus elements. J Virol 1999; 73(4):2832-2840.

HIV-1 Vector Systems

79

155. Wu X, Wakefield JK, Liu H et al. Development of a novel trans-lentiviral vector that affords predictable safety. Mol Ther 2000; 2(1:47-55. 156. Dougherty JP, Temin HM. A promoterlessretroviral vector indicates that there are sequences in U3 required for 3' RNA processing. Proc Natl Acad Sci USA 1987; 84(5):1197-1201. 157. Yu SF, von Ruden T, Kantoff PW et al. Self-inactivating retroviral vectors designed for transfer of whole genes into mammalian cells. Proc Natl Acad Sci USA 1986;83(10):3194-3198. 158. DeZazzo JD, Imperiale MJ. Sequences upstream of AAUAAA influence poly(A) site selection in a complex transcription unit. Mol Cell Biol 1989; 9(11):4951-4961. 159. Kung SK, An DS, Chen IS. A murine leukemia virus (MuLV) long terminal repeat derived from rhesus macaques in the context of a lentivirus vector and MuLV gag sequence results in high-level gene expression in human T lymphocytes. J Virol 2000; 74(8):3668-3681. 160. Emerman M, Temin HM. Genes with promoters in retrovirus vectors can be independently suppressed by an epigenetic mechanism. Cell 1984; 39(3 Pt 2):449-467. 161. Harrich D, Ulich C, Garcia-Martinez LF et al. Tat is required for efficient HIV-1 reverse transcription. Embo J 1997;16(6):1224-1235. 162. McBride MS, Schwartz MD, Panganiban AT. Efficient encapsidation of human immunodeficiency virus type 1 vectors and further characterization of cis elements required for encapsidation. J Virol 1997;71(6):4544-4554. 163. Rizvi TA, Schmidt RD, Lew KA et al. Rev/RRE-independent Mason-Pfizer monkey virus constitutive trasport element-dependent propagation of SIVmac239 vectors using a single round of replication assay. Virology 1996;222(2):457-463. 164. Rizvi TA, Lew KA, Murphy EC, Jr. et al. Role of Mason-Pfizer monkey virus (MPMV) constitutive transport element (CTE) in the propagation of MPMV vectors by genetic complementation using homologous/heterologous env genes. Virology 1996; 224(2):517-532. 165. Rizvi TA, Schmidt RD, Lew KA. Mason-Pfizer monkey virus (MPMV) constitutive transport element (CTE) functions in a position-dependent manner. Virology 1997; 236(1):118-129. 166. Donello JE, Loeb JE, Hope TJ. Woodchuck hepatitis virus contains a tripartite posttranscriptional regulatory element. J Virol 1998; 72(6):5085-5092. 167. Zufferey R, Donello JE, Trono D et al. Woodchuck hepatitis virus posttranscriptional regulatory element enhances expression of transgenes delivered by retroviral vectors. J Virol 1999; 73(4):2886-2892. 168. May C, Rivella S, Callegari J et al. Therapeutic haemoglobin synthesis in beta-thalassaemic mice expressing lentivirus-encoded human beta-globin. Nature 2000; 406(6791):82-86. 169. Kordower JH, Emborg ME, Bloch J et al. Neurodegeneration prevented by lentiviral vector delivery of GDNF in primate models of Parkinson’s disease. Science 2000; 290(5492):767-773. 170. Corbeau P, Wong-Staal F. Anti-HIV effects of HIV vectors. Virology 1998; 243(2):268-274. 171. Bukovsky AA, Song JP, Naldini L. Interaction of human immunodeficiency virus-derived vectors with wild- type virus in transduced cells. J Virol 1999; 73(8):7087-7092. 172. Lorincz MC, Schubeler D, Goeke SC et al. Dynamic analysis of proviral induction and De Novo methylation: implications for a histone deacetylase-independet, methylation ensity- dependent mechanism of transcriptional repression. Mol Cell Biol 2000; 20(3):842-850. 173. Cherry SR, Biniszkiewicz D, van Parijs L et al. Retroviral expression in embryonic stem cells and hematopoietic stem cells. Mol Cell Biol 2000; 20(20):7419-7426. 174. Chen WY, Townes TM. Molecular mechanism for silencing virally transduced genes involves histone deacetylation and chromatin condensation. Proc Natl Acad Sci USA 2000;97(1):377-382. 175. Chen WY, Bailey EC, McCune SL et al. Reactivation of silenced, virally transduced genes by inhibitors of histone deacetylase. Proc Natl Acad Sci USA 1997; 94(11):5798-5803. 176. Naldini L, Blomer U, Gage FH et al. Efficient transfer, integration, and sustained long-term expression of the transgene in adult rat brains injected with a lentiviral vector. Proc Natl Acad Sci USA 1996; 93(21):11382-11388. 177. Blomer U, Naldini L, Kafri T et al. Highly efficient and sustained gene transfer in adult neurons with a lentivirus vector. J Virol 1997; 71(9):6641-6649. 178. Hamaguchi I, Woods NB, Panagopoulos I et al. Lentivirus vector gene expression during ES cellderived hematopoietic development in vitro. J Virol 2000; 74(22):10778-10784.

80

Lentiviral Vector Systems for Gene Transfer

179. Agarwal M, Austin TW, Morel F et al. Scaffold attachment region-mediated enhancement of retroviral vector expression in primary T cells. J Virol 1998; 72(5):3720-3728. 180. Auten J, Agarwal M, Chen J et al. Effect of scaffold attachment region on transgene expression in retrovirus vector-transduced primary T cells and macrophages. Hum Gene Ther 1999; 10(8):1389-1399. 181. Dang Q, Auten J, Plavec I. Human beta interferon scaffold attachment region inhibits de novo methylation and confers long-term, copy number-dependent expression to a retroviral vector. J Virol 2000; 74(6):2671-2678. 182. Kafri T, van Praag H, Gage FH et al. Lentiviral vectors: regulated gene expression. Mol Ther 2000; 1(6):516-521. 183. Shockett PE, Schatz DG. Diverse strategies for tetracycline-regulated inducible gene expression. Proc Natl Acad Sci USA 1996; 93(11):5173-5176. 184. Deuschle U, Meyer WK, Thiesen HJ. Tetracycline-reversible silencing of eukaryotic promoters. Mol Cell Biol 1995; 15(4):1907-1914. 185. Hofmann A, Nolan GP, Blau HM. Rapid retroviral delivery of tetracycline-inducible genes in a single autoregulatory cassette. Proc Natl Acad Sci USA 1996; 93(11):5185-5190. 186. Paulus W, Baur I, Boyce FM et al. Self-contained, tetracycline-regulated retroviral vector system for gene delivery to mammalian cells. J Virol 1996; 70(1):62-67. 187. Rivera VM, Clackson T, Natesan S et al. A humanized system for pharmacologic control of gene expression [see comments]. Nat Med 1996; 2(9):1028-1032. 188. Spencer DM, Wandless TJ, Schreiber SL et al. Controlling signal transduction with synthetic ligands [see comments]. Science 1993; 262(5136):1019-1024. 189. Braselmann S, Graninger P, Busslinger M. A selective transcriptional induction system for mammalian cells based on Gal4-estrogen receptor fusion proteins. Proc Natl Acad Sci USA 1993; 90(5):16571661. 190. Wang Y, O’Malley BW, Jr., Tsai SY et al. A regulatory system for use in gene transfer. Proc Natl Acad Sci USA 1994; 91(17):8180-8184. 191. Whelan J, Miller N. Generation of estrogen receptor mutants with altered ligand specificity for use in establishing a regulatable gene expression system. J Steroid Biochem Mol Biol 1996; 58(1):3-12. 192. Bushman FD. Tethering human immunodeficiency virus 1 integrase to a DNA site directs integration to nearby sequences. Proc Natl Acad Sci USA 1994; 91(20):9233-9237. 193. Goulaouic H, Chow SA. Directed integration of viral DNA mediated by fusion proteins consisting of human immunodeficiency virus type 1 integrase and Escherichia coli LexA protein. J Virol 1996; 70(1):37-46. 194. Re F, Luban J. HIV-1 Vpr: G2 cell cycle arrest, macrophages and nuclear transport. Prog Cell Cycle Res 1997; 3:21-27. 195. Krausslich HG, Ochsenbauer C, Traenckner AM et al. Analysis of protein expression and virus-like particle formation in mammalian cell lines stably expressing HIV-1 gag and env gene products with or without active HIV proteinase. Virology 1993; 192(2):605-617. 196. Shoeman RL Honer B, Stoller TJ et al. Cleavage of the intermediate filament subunit protein vimentin by HIV-1 protease: utilization of a novel cleavage site and identification of higher order polymers of pepstatin A. Adv Exp Med Biol 1991; 306:533-537. 197. Shoeman RL, Mothes , Honer B et al. Effect of human immunodeficiency virustype 1 protease on the intermediate filament subunit protein vimentin: cleavage, in vitro assembly and altered distribution of filaments in vivo following micrinjection of protease. Acta Histochem Suppl 1991; 41:129141. 198. Corbeau P, Kraus G, Wong-Staal F. Efficient gene transfer by a human immunodeficiency virus type 1 (HIV- 1)-derived vector utilizing a stable HIV packaging cell line. Proc Natl Acad Sci USA 1996; 93(24):14070-14075. 199. Corbeau P, Kraus G, Wong-Staal F. Transduction of human macrophages using a stable HIV-1/ HIV-2-derived gene delivery system. Gene Ther 1998; 5(1):99-104. 200. Kafri T, van Praag H, Ouyang L et al. A packaging cell line for lentivirus vectors. J Virol 1999; 73(1):576-84. 201. Xu K, Ma H, McCown TJ et al. Geneation of a stable cell line producing high-titer self-inactivating lentiviral vectors. Mol Ther 2001; 3(1):97-104.

HIV-1 Vector Systems

81

202. Howcroft TK, Strebel K, Martin MA et a. Reprssion of MHC class I gene promoter activity by two-exon Tat of HIV. Science 1993; 260(5112):1320-1322. 203. Kanazawa S, Okamoto T, Peterlin BM. Tat competes with CIITA for the binding to P-TEFb and blocks the expression of MHC class II genes in HIV infection [In Process Citation]. Immunity 2000; 12(1):61-70. 204. An DS, Wersto RP, Agricola BA et al. Marking and gene expression by a lentivirus vector in transplanted human and nonhuman primate CD3(+) cells. J Virol 2000; 74(3):1286-1295. 205. Wilson CA, Ng TH, Miller AE. Evaluation of recommendations for replication-competent retrovirus testing associated with use of retroiral vectors. Hum Gene Ther 1997; 8(7):869-874. 206. Haas DL, Case SS, Crooks GM et al. Critical factors influencing stable transduction of human CD34(+) cells with HIV-1-derived lentiviral vectors. Mol Ther 2000; 2(1):71-80. 207. Kimpton J, Emerman M. Detection of replication-competent and pseudotyped human immunodeficiency virus with a sensitive cell line on the basis of activation of an integrated beta-galactosidase gene. J Virol 1992; 66(4):2232-2239. 208. Mann DA, Mikaelian I, Zemmel RW et al. A molecular rheostat. Co-operative rev bindng to stem I of the rev- response element modulates human immunodeficiency virus type-1 late gene expression. J Mol Biol 1994; 241(2):193-207.

CHAPTER 5

HIV-2 and SIV Vector Systems James R. Gilbert and Flossie Wong-Staal

Abstract

L

entiviral vectors have received much attention in recent years due to their ability to efficiently transduce non-dividing cells. Of the lentiviruses HIV-2 and SIV offer several unique benefits as the basis for lentiviral vector design. HIV-1, HIV-2 and SIV remain the only known primate lentiviruses, and consequently are among the most extensively studied viruses known. Substantial effort has been devoted towards identifying the pathogenic determinants of the primate lentiviruses and towards understanding their replication within primates. Of the primate lentiviruses, the pathogenicity and rates of transmission of HIV-2 and SIV fall far below that of HIV-1, potentially providing vectors based upon HIV-2/SIV with a greater degree of biosafety. Last, and perhaps most importantly, HIV-2 and SIV are viruses which may be studied within non-human primate models susceptible to AIDS-like disease, making vectors based upon these viruses accessible to substantial preclinical evaluation. We approach this Chapter presenting information regarding the basic biology of HIV-2 and SIV and conclude by pointing to how unique features of HIV-2 and SIV are well suited to vector design, hoping to leave the reader with a greater appreciation of the potential these viruses offer within the field of gene transfer applications.

Introduction Replication-defective retroviral vectors have traditionally been a preferred vehicle for gene transfer due to their ability to permanently integrate within the chromosome and establish stable gene expression within transduced populations. Further, the “simple” retroviral genome and its replication cycle have been extensively characterized enabling investigators to refine retroviral vectors and reduce the risk of an immune response against transduced cells by eliminating all non-essential viral elements. The absence of non-essential viral elements is likely to be a necessary feature within vectors in order to achieve sustained in vivo gene expression. Clinical use of simple retroviral vectors has been severely limited, however, due to their inability to efficiently transduce the quiescent and post-mitotic cells which are among the most desirable targets for gene therapy. The general basis for this limitation has been elucidated. Integration of the “simple” retroviral genome first requires mitotic dissolution of the nuclear membrane in order to provide the retroviral pre-integration complex (PIC) with access to cellular chromatin.1-3 Thus, in the absence of cell division retroviral infection is blocked prior to integration. Lentiviruses, in contrast, efficiently infect non-dividing cells during the normal course of their replication. As discussed in detail in Chapter 3, the lentiviral capacity to infect nondividing cells may result from a direct interaction between the viral PIC and cellular nuclear Lentiviral Vector Systems for Gene Transfer, edited by Gary L. Buchschacher, Jr. ©2003 Eurekah.com.

84

Lentiviral Vector Systems for Gene Transfer

import machinery, resulting in transportation of the lentiviral PIC across the nuclear envelope where integration of proviral DNA may occur.4 Viral components within the PIC which contribute to nuclear import may include the viral Gag matrix, integrase, and Vpr or Vpx, although the latter accessory gene products are reported to use non-classical import pathways.59 The extent to which each of these components contributes to nuclear import is incompletely defined, although their cumulative effect provides the primate lentiviruses with a functional redundancy that helps insure the completion of an essential step in its replication cycle. Initially, the potential benefits lentiviruses were thought to offer in gene transfer inspired many investigators to develop vectors and stable packaging lines based upon HIV-1. These initial studies confirmed that transduction of non-dividing cells by lentiviral vectors was feasible, although the titers which were obtained fell significantly below a clinically useful range.1012 This problem was in part addressed in work presented by Naldini et al describing the development of a replication-defective HIV-1 vector pseudotyped with VSV-G.13 The vector was shown to retain the lentivirus-specific ability to infect non-dividing cells, and achieved efficient gene delivery both in vitro and by direct injection into the adult rat brain in vivo. Since these findings, a flurry of effort has been devoted to the design and characterization of lentiviral vectors. Of the lentiviruses considered for vector design, two promising candidates are HIV-2 and SIV. HIV-2 and SIV, being primate lentiviruses, are among the most thoroughly studied and well characterized lentiviruses currently known. Extensive biological characterization of HIV2/SIV and long term prospective studies of associated disease provide a substantial body of information to draw upon in order to assess vector behavior within a patient. This contrasts with the non-primate lentiviruses about which far less is known in terms of basic biology or their potential behavior within primates. Like all lentiviruses, both efficiently transduce a variety of non-dividing cell types. HIV-2 and SIV, however, possess features making them uniquely well suited to vector development. Among these features are a diminished pathogenic potential when compared to HIV-1, an amenability to study in primate models, and separation of the accessory gene product-associated cell cycle arrest and nuclear import functions. Each of these characteristics offer potential advantages over other lentiviral vectors and will be discussed in detail. Our goal in this Chapter is to present current information addressing the basic biology, viral dynamics and genome organization of HIV-2/SIV as this information pertains to vector development. Because basic lentiviral replication and the design of HIV-1 vectors are discussed elsewhere within this volume, we focus in this Chapter upon areas in which the primate lentiviruses contrast with one another. Despite a variety of similarities among the primate lentiviruses, distinctions do exist. We approach this Chapter intending to address these distinctions, concluding with sections addressing the development of HIV-2/SIV vectors and potential future modifications which might be introduced to improve these vectors.

Classification and Distribution Although a human virus, HIV-2 displays greater sequence homology with SIV than HIV1. Sequence relatedness between HIV-2 and SIV is approximately 75%, whereas both viruses display less than 50% homology to HIV-1.14,15 Based upon genetic diversity, six distinct subtypes (A-F) of HIV-2 have been classified.16,17 All are found predominantly within West African countries, although infrequent cases are reported in the Americas, Europe and India. The majority of isolates characterized to date belong to subtype A, isolates of which have been obtained from diverse regions across West Africa. Subtype B predominates in Ghana and Cote d’ Ivoire and constitutes the majority of HIV-2 isolates falling outside of subtype A. Rare isolates of subtypes C-F have been described in Sierra Leone and Liberia. To date, subtypes CF have been isolated only from asymptomatic individuals, whereas the bulk of HIV-2 associated cases of AIDS which have been reported are caused by subtype A. It is tempting to speculate that this may reflect a differential virulence among the viral subtypes. Unfortunately, the

HIV-2 and SIV Vectors

85

infrequent isolation of subtypes C-F and the absence of long-term prospective studies comparing subtype-specific differences makes it impossible to evaluate the statistical significance of these observations. Nomenclature for the SIV subtypes is based upon the primate species from which prototypic viral strains representative of the subtype was isolated. Five distinct lentiviruses have been obtained from non-human primates native to Africa, including chimpanzees (SIVCPZ), sooty mangabees (SIVSM), african green monkeys (SIVAGM), mandrills (SIVMND), and sykes (SIVSYK). Each of the subtypes is endemic to the primate species for which it was named and shows no evidence of disease causation within its natural host.18 Other designations commonly encountered in the literature include SIVMAC, SIVMNE and SIVSTM, named for SIV strains isolated at primate research centers from rhesus macaques, nemestrina macaques and stump-tailed macaques, respectively. Each of these viral strains causes a fatal AIDSlike disease in macaques. Due to the absence of evidence for SIV infection of macaques in their native Asian habitat and given the close genetic relatedness of these viral strains to SIVSM, it is thought that SIV is not a virus endogenous to this species. Macaques are instead thought to have become inadvertently infected by experimental inoculation with fluids obtained from other primate species, or by co-housing of macaques with other species in research facilities. Extensive genetic diversity is found within all of the HIV-2/SIV subtypes. As with all lentiviruses, the principal driving force to diversification is the viral mutation rate, although viral strains postulated to have arisen by means of recombination between distinct viral subtypes within the same individual have been reported. Phylogenetic analysis of the HIV and SIV pol sequences indicates SIVAGM, SIVMND and SIVSYK belong to discrete lineages. HIV-1 clusters with SIVCPZ in a separate lineage, whereas HIV-2 clusters with SIVSM. Despite substantial sequence variation within the lentiviral subtypes, each of these lineages are roughly equidistant and show approximately 40% divergence from one another. These observations have led some investigators to speculate HIV-1 and HIV-2 may have originated through zoonotic transmission from non-human primates to humans.

Pathology and Viral Replication Viral Transmission The basic biology of HIV-2 and SIV, in some ways, closely resembles that of HIV-1. All three of the primate lentiviruses display common routes of transmission, cellular tropisms, long and variable incubation periods, viral replication kinetics and persistence of replication despite strong humoral and cellular immune response within the host.19 Each spreads through exchange of bodily fluids including blood, blood products, saliva, semen, vaginal secretions and milk. Trans-placental transmission of HIV/SIV can occur, although perinatal transmission of HIV-2 occurs far less efficiently than for HIV-1.20 Initial infection is mediated by binding of the viral surface glycoprotein gp120 to a cellular CD4 molecule which serves as the primary receptor for all of the primate lentiviruses.21,22 Early attempts to determine if expression of the CD4 receptor was sufficient to confer susceptibility to infection to non-permisive cell types found that, although CD4 facilitates viral binding at the cell surface, it is insufficient to permit viral fusion with the cell membrane. Further, primary viral isolates display variable tropism for CD4+ lymphocytes. Together, these findings implied the necessary existence of other cellular co-factors mediating viral binding and cell fusion. After substantial effort, the chemokine receptors CCR5 and CXCR4 were identified as the principal lentiviral co-receptors.23,24 Several studies have since shown that the CD4-glycoprotein complex, formed by viral binding to the CD4+ molecule, interacts with its chemokine coreceptor to initiate the fusion process. Conversely, viral entry

86

Lentiviral Vector Systems for Gene Transfer

may be inhibited in the presence of RANTES, MIP-1α and MIP-1β, the ligands for CCR5, or SDF-1, the ligand for CXCR4, depending upon the biological phenotype of the viral strain.

Viral Phenotype and Coreceptor Use The biological phenotype of primate lentiviruses, though, evolves during the course of infection. Forces driving this change include the viruses intrinsically rapid mutation rate and immunological pressures selecting for under-represented members of the lentiviral quasi-species found within the individual at any given moment. As discussed in Chapter 2, two general viral phenotypes have been described: macrophage tropic (M-tropic) and T lymphotropic (Ttropic). Of the two principal chemokine receptors, CCR5 functions as the predominant coreceptor for M-tropic HIV-1 strains, whereas CXCR4 functions as the predominant coreceptor for T-tropic HIV-1 strains. Unlike HIV-1 and HIV-2, it should be noted that SIV uses CCR5 but not CXCR4 at all stages of infection.25 Since the initial discovery that CCR5 and CXCR4 act as the principal lentiviral coreceptors, however, at least 14 alternate lentiviral coreceptors have been identified. In the natural progression of HIV-1 infection, a discrete series of events relating viral phenotype and chemokine receptor usage to disease prognosis has been defined. Generally, initial infection by the primate lentiviruses occurs by means of M-tropic viral strains. Supporting this claim is the fact that individuals homozygous for the CCR5∆32/∆32 allele are highly resistant to HIV-1 infection. Longitudinal studies of HIV-1 infection, however, report a shift towards principal usage of the CXCR4 coreceptor and more promiscuous use of alternate coreceptors late in the course of disease progression. Principal use of CXCR4 and/or promiscuous use of alternate coreceptors by HIV-1 directly relates to the appearance of syncytiuminducing (SI) T-tropic strains and correlates poorly with disease prognosis and survival.26 In stark contrast, many primary isolates and molecular clones of HIV-2/SIV use alternate coreceptors with efficiencies comparable to CCR5 or CXCR4.27-29 Based upon data acquired using cell-free infectivity assays, some strains of HIV-2 appear capable of using alternate chemokine receptors even in the absence of CD4. Of note, broad coreceptor usage in HIV-2, unlike HIV-1, shows no correlation what-so-ever with syncytium-inducing capabilities. For HIV-1, promiscuous coreceptor usage has been hypothesized to either contribute to or to reflect enhanced cytopathicity. This does not hold true for HIV-2, for which disease progression is substantially slower despite promiscuous coreceptor usage at all stages of infection.

Viral Pathology Although HIV-2 and SIV are far less pathogenic than HIV-1, they display common elements during the acute phase of infection.18,19 Primary infection is followed by a burst of viral replication and an acute illness characterized by mononucleosis-like symptoms. Within 2 months of infection, a significant decline in CD4+ T lymphocyte levels is detected in the peripheral blood. Typically, CD4+ lymphocyte levels rebound, although rarely to pre-infection levels. Specific antiviral responses involving both the humoral and cellular arms of the immune system are detected approximately 1 month after infection. With the mounting of an immune response, the infected individual generally enters an asymptomatic phase during which time few, if any, disease symptoms manifest. Recent studies indicate viral replication continues to occur at a rapid rate during clinical latency, with the production of 109-1010 virions per day and the daily turnover of up to 109 infected CD4+ T lymphocytes.30,31 Without intervention, in the case of HIV-1, the CD4+ T lymphocyte count steadily declines during the course of infection, eventually falling below 200-500/µL at which point disease symptoms become manifest and clinical AIDS ensues. In contrast, the asymptomatic phase of HIV-2 infection is substantially longer and CD4+ T lymphocyte depletion is far less pronounced. Similarly, morbidity and mortality associated

HIV-2 and SIV Vectors

87

with HIV-2 is significantly lower than that of HIV-1. In the most comprehensive prospective study yet performed to compare the rates of disease progression between HIV-1 and HIV-2 infected individuals, a 67% probability of AIDS-free survival 5 years after seroconversion was reported for the HIV-1 cohort.32 In contrast, the probability of survival over the same time period within the HIV-2 cohort was 100%. Related cross-sectional studies confirm reduced pathogenicity associated with HIV-2 as measured by a variety of indices, including CD4+ counts, CD8+ counts, and the CD4+:CD8+ ratio.33,34 Similarly, african green monkeys and sooty mangabeys fail to develop disease despite persistent, life-long infection with SIV. In addition to its reduced pathogenicity, HIV-2 is reported to be less transmissible than HIV-1. Consistent with this finding, several investigators report observing shifts in the prevalence of HIV-1 and HIV-2 among commercial sex workers. Findings in Bamako, Mali show an increase in HIV-1 prevalence (from 10 to 35.8%) and a concurrent decrease in HIV-2 prevalence (from 15 to 3.9%) between the years 1987 and 1995.16 Similar reports from countries in which HIV-2 is the predominant virus indicate the rate of HIV-1 infection to be increasing while HIV-2 prevalence declines. Viral and host factors contributing to disease progression in HIV infection have been the subject of extensive review.19 Why these same factors fail to promote disease in HIV2/SIV infection, or do so at drastically reduced rates, remains unclear at this time. Based upon these observations, however, it is possible that vectors based upon HIV-2/SIV may offer a substantial advantage over HIV-1 vectors in terms of biosafety. Efforts to address these questions will undoubtably yield insights which contribute to the development of safer lentiviral vectors.

Animal Models One of the greatest benefits to use of HIV-2/SIV vectors over other lentiviral vectors is their amenability to study in primate models for disease. Although some strains of HIV can establish persistent infection in chimpanzees, the animals fail to exhibit disease symptoms. Further, other factors prohibit the wide use of chimpanzees in animal studies. Several strains of HIV-2/SIV, however, establish persistent infection in baboons and macaques and display varying degrees of pathogenicity. Highly pathogenic strains of HIV-2/SIV produce an AIDS-like disease within macaques which culminates in death within a matter of months.35-38 Consequently, the biosafety of HIV-2/SIV vectors may be assessed within primates prior to clinical studies. Another concern with regard to lentiviral vectors is the potential for vector mobilization, a process by which transfer vector RNAs are encapsidated by replication competent lentivirus present within the same organism and subsequently spread to additional tissues. In many situations vector mobilization may prove beneficial, although this remains to be evaluated. This process, and any potential hazards it may represent, may be studied using HIV-2/SIV models in primates, providing HIV-2/SIV vectors with an advantage currently unavailable to other lentiviruses. Thus, in terms of preclinical evaluation of vector biosafety, HIV-2 and SIV offer a number of unique characteristics over other lentiviral vectors.

Genome Organization and Regulation Infectious particles of HIV-2 and SIV contain identical copies of an approximately 9 kb long single-stranded RNA genome (Fig. 1). Upon infection, a virally encoded reverse transcriptase converts this RNA genome into a linear double-stranded DNA in which identical copies of the viral long terminal repeat (LTR) flank viral genes. This DNA copy of the viral genome is subsequently integrated into the host chromosome by the virally encoded integrase. Once integrated, viral transcripts are expressed from the viral LTR using the cellular RNA Polymerase II pathway, leading to the expression of viral proteins and the accumulation of viral genomic RNAs required for viral replication. Lentiviral gene expression is biphasic, showing a temporal shift in the pattern of gene expression during early and late stages of viral replication. The early stage of viral replication is

88

Lentiviral Vector Systems for Gene Transfer

Fig. 1 Genomic organization of HIV-1 and HIV-2/SIVMAC. The relative locations of the structural, regulatory, and accessory genes are indicated, MSD, major splice donor.

characterized by the appearance of multiply spliced transcripts encoding the tat, rev and nef genes. Interaction of Rev with the RRE (see below) results in expression of the late gene products. Included among these are the structural genes gag, pol and env, and the accessory genes vif, vpr and vpu or vpx. Because many features of the viral structural, regulatory and accessory gene products are fundamentally common to all the primate lentiviruses and have been discussed elsewhere within this volume, they will only be addressed in a cursory fashion here. We will instead focus upon the relevance of these features in terms of lentiviral gene expression and/or uniqueness to HIV-2/SIV.

Structural Genes As discussed in Chapter 2, the structural genes gag, pol and env comprise the basic core of all retroviral genomes. Comparison of the HIV-1, HIV-2 and SIV structural gene sequences reveals substantial sequence variation. However, the function played by these elements remains relatively constant. One noteworthy exception involves the role of the V3 loop of gp120, also referred to as the principal neutralizing determinant. The V3 loop is among the most variable regions within HIV-1, and synthetic peptides based upon V3 elicit potent type-specific neutralizing antibodies within animals. The frequency of nonsynonomous versus synonomous mutations contributing to variability within this region in HIV-1 suggests the presence of strong selective pressures driving mutation. In contrast, the V3 region of SIV is relatively conserved and fails to elicit type-specific neutralizing antibodies, whereas the V3 region of some but not all HIV-2 strains elicit neutralizing antibodies. Greater variability is instead seen in SIV within the V1, V2 and V4 regions rather than V3. Attempts to generate a chimeric virus replacing the V3 loop of SIVMM239 with that of HIV-1MN failed to produce replication competent virus, although a comparable HIV-2KR/HIV-1 V3 chimera was capable of replication.39 These observations suggest that tolerance for variation within this region and presumably the role played by V3 in HIV-1, HIV-2 and SIV replication significantly differ. Additional aspects of the structural genes common to all lentiviruses have been reviewed.40

HIV-2 and SIV Vectors

89

Regulatory genes As discussed in Chapter 2, the primate lentiviruses encode two regulatory proteins, Tat and Rev, which play essential roles in viral replication. Tat functions as a trans-activator of transcription and strongly elevates viral RNA abundance, acting at the levels of transcription initiation and transcription elongation. Rev acts post-transcriptionally, regulating the splicing of viral transcripts and the transport of unspliced or partially spliced transcripts to the cytoplasm. Function of both the regulatory proteins requires their interaction with cis-acting RNA elements. Tat requires TAR, a RNA element which forms stem-loop structures and is found at the 5’ end of all viral transcripts. A complex RNA structure located within the env gene designated the Rev-responsive element (RRE) is required for function of Rev.

Tat Tat is an approximately 16 kD protein encoded by two exons within the lentiviral genome. Slight variation in the molecular weight of HIV-1 Tat has been described between viral isolates (86 to 101 amino acids), although truncated 58-72 amino acid forms encoded by the first exon are functional. HIV-2 Tat, in contrast, is composed of 130 amino acids and, outside of conserved cysteine-rich and arginine-rich domains, displays little homology with HIV-1 Tat.41 Also, unlike HIV-1 Tat, approximately 20% of the amino terminus and an additional 30% of the carboxy terminus of HIV-2 Tat is dispensable for function. Domains reported to be essential for the function of HIV-2 Tat include a cysteine-rich domain thought to comprise part of the protein activation domain, and an arginine-rich domain thought to mediate Tat:TAR binding. The HIV-1 TAR element is approximately 60 nucleotides in length and forms a single, stable RNA stem-loop containing a small pyrimidine-rich bulge which is essential to recognition and binding by Tat. The HIV-2/SIVMAC TAR sequence is, in contrast, approximately 120 nucleotides in length. Models of the secondary structure within this region indicate the presence of duplicate stem loops, both of which contribute to HIV-2 Tat-TAR interaction. Of note, HIV-1 and HIV-2/SIV Tat proteins display non-reciporical complementarity.42 The HIV-2 trans-activates its own LTR far more efficiently than it does the HIV-1 LTR. The HIV-1 Tat, in contrast, efficiently trans-activates either LTR with comparable efficiency. For all of the primate lentiviruses, however, the abundance of viral transcripts is elevated by several orders of magnitude in response to trans-activation by Tat.

Rev Rev is an approximately 19 kD phosphoprotein encoded by two exons within the lentiviral genome and is expressed early in the viral replication cycle. Functionally, the protein regulates a shift in the biphasic pattern in lentiviral gene expression. Early in viral replication, only multiply spliced transcripts (encoding Tat, Rev and Nef ) are detected. In the later stages of viral replication Rev binds to the cis-acting RRE found within unspliced and partially spliced transcripts (encoding viral structural proteins), promoting their transport, stability, and translation. Domains playing a role in Rev’s nuclear localization, RRE-specific RNA binding, oligomerization and post-trancriptional trans-activation of viral gene expression are highly conserved among the primate lentiviruses.43 A basic domain situated between amino acids 35-50 contributes both to Rev's nuclear localization and binding to RRE-containing transcripts. Amino acids 18-56, extending outside of the basic domain, are implicated in Rev multimerization. A conserved leucine-rich effector domain, between amino acids 75-84, plays an essential role in nuclear export, after association with cellular factors. Despite conservation of these functional domains within the primate lentiviruses, the HIV-1 and HIV-2/SIV Revs display non-reciporiced complementarity. HIV-2 Rev fails to function with the HIV-1 RRE, whereas HIV-1 Rev and HTLV I and II Rex proteins readily

90

Lentiviral Vector Systems for Gene Transfer

function with the HIV-2 RRE.44 Surprisingly, upon examination of a shorter HIV-2 RRE which more precisely defines the borders of its predicted secondary structure, the HIV-1 but not the HIV-2 Rev retained the ability to function.

The Accessory Genes The accessory genes of HIV-1 (vif, vpr, vpu and nef) and HIV-2/SIV (vif, vpr, vpx and nef) were originally identified as a series of short open reading frames which were dispensable for viral replication within established cell lines. Since their discovery, however, a growing body of evidence suggests the accessory gene products play a central role in viral replication and pathogenesis in vivo. The function of two of these genes, vif and nef, are relatively conserved among the primate lentiviruses and so will not be discussed here. Of greater interest in terms of HIV2/SIV vector design are vpr and vpx. The HIV-1 Vpr is a virion-associated 14 kD protein which interacts with p6Gag. Functionally, Vpr plays two roles: inducing cell cycle arrest at the G2/M border of the cell cycle, and facilitating nuclear transport of the PIC in non-dividing cells. In contrast, these two roles are segregated within HIV-2/SIV between Vpr and Vpx, respectively.9 The first study to show this reports that SIVSM PBj1.9 proviral clones defective in vpr efficiently infect macrophage but fail to induce cell cycle arrest within established cell lines. Comparable PBj1.9 mutants defective in vpx, conversely, stimulate cell cycle arrest at the G2/M border of the cell cycle, but fail to efficiently infect primary macrophage. Of note, PBj1.9 vpx mutants fail to efficiently replicate in primary macrophage whether or not a simultaneous mutation is introduced into the Gag matrix NLS, indicating that HIV-2/SIVSM Vpx plays a dominant role in nuclear transport of the viral PIC in monocyte-derived macrophage. This is in contrast to HIV-1, in which nuclear transport functions of the Gag matrix and Vpr overlap. With regard to vector development, these findings suggest benefits unique to HIV-2/SIV. Within HIV-2/SIV vector systems, unlike HIV-1, it is possible to eliminate the undesirable cell cycle arrest function of Vpr without sacrificing the desirable nuclear transport function of Vpx. Both features may prove to be essential in establishing optimal stable packaging lines.

LTR Expression from the lentiviral LTR is a complex process involving the interaction of cellular basal transcription factors, virally encoded trans-activators, and cis-acting viral sequences. The LTR itself is composed of three domains, U3, R and U5, which are common to all retroviral LTRs. Transcription initiates at the U3/R boundary of the 5' LTR. Basal transcription is regulated by a core promoter region found within the U3 domain of the 5' LTR. This core promoter includes a classical TATAA box and adjacent Sp-1 binding sites. The number of Sp-1 binding sites found within U3 varies among the primate lentiviruses: the HIV-2 LTR contains four, different strains of SIV contain between two and four, and the HIV-1 LTR contains three. Studies suggest the relative spacing of Sp-1 binding sites within the LTR may influence viral trans-activation mediated by Tat, and that deletion of the Sp-1 binding sites reduces enhancer activity. Also found within U3 are NF-κB enhancer elements which play a pivotal role in HIV/ SIV replication.45,46 The presence of these enhancer elements is significant, because they provide a mechanism by which other viral proteins and host mitogens and chemokines promote lentiviral gene expression. Again, the frequency of these sites varies among the primate lentiviruses. HIV-2 contains one functional NF-κB site and one non-functional site which differs slightly from the consensus GGGACTTTCC sequence. SIV contains one or two sites depending upon the viral strain examined, whereas HIV-1 has two functional NF-κB enhancer sites. The presence of a single NF-κB site, however, is sufficient to maintain wild-type enhancer activity. NF-κB-dependent trans-activation of the viral LTR is essential to viral

HIV-2 and SIV Vectors

91

transcription in both CD4+ T lymphocytes and macrophages. Cellular factors related to these processes have been the subject of several reviews.47,48 The R and U5 domains and adjacent regions play essential roles in reverse transcription (discussed in previous Chapters). The R domain participates in template switching during reverse transcription and contains TAR, the cis-acting element required for Tat-mediated transactivation. Immediately downstream of the 5’ U5 is an 18 nucleotide primer binding site which, in conjunction with tRNAlys, serves as the initiation site for minus strand synthesis during reverse transcription. Similarly, a poly-purine tract is found immediately upstream of the 3’ U3 which serves as the intiation site for second strand synthesis during reverse transcription.

The ψ Packaging Determinant/Encapsidation Signal The ψ sequence, defined by convention as the sequence between the major splice donor and the ATG initiation codon of gag, was originally described in murine oncoretroviruses as a sequence which is both necessary and sufficient for RNA encapsidation. Deletion of ψ in murine retroviruses strongly attenuates encapsidation. Conversely, attachment of ψ or ψ’ (which includes ψ and a short segment of gag) to heterologous RNAs confers nearly wild-type levels of encapsidation.49 The cis-acting sequences involved in lentiviral encapsidation, however, appear to be more complex. In contrast to murine retroviruses, determinants of HIV encapsidation have been mapped both upstream and downstream of the major splice donor.50-52 Because sequences within the leader region, upstream of the major splice donor, are found within both genomic and subgenomic RNAs, these findings strongly imply the existence of additional encapsidation signals which provide the viral packaging machinery with a mechanism for distinguishing between the two. Of greater import in terms of lentiviral vector design, the ψ (or E, for encapsulation) packaging determinant of lentiviruses extends into the gag coding region. For this reason, a short stretch of gag (250-400 nucleotides) must be incorporated in optimal lentiviral transfer vectors. This effectively makes it impossible to completely separate the cis-acting sequences required within the transfer vector from trans-acting sequences which define the packaging construct because the two overlap. Two consequences result from this. The potential for homologous recombination during vector production increases, due to the presence of identical gag sequences within both the transfer vector and packaging construct. Also, due to the presence of cis-acting repressive sequences within gag, its incorporation into the transfer vector makes necessary the inclusion of the RRE to compensate. Predicted secondary structure within the region is complex and involves multiple RNA stem loops which contribute to encapsidation, dimerization of the viral genome, and binding to the viral nucleocapsid. Recent analysis of the HIV-2 leader sequence and accompanying E region predicts between six and eight stem loop structures in contrast to the four stem loops ascribed to the HIV-1 leader.53 The functional relevance of the additional stem loops within the HIV-2 leader remains unclear. Comparison of the HIV-1 and HIV-2 leader sequences is impractical due to the lack of significant homology between the two.

Vector Systems HIV-2/SIV Production of helper-virus free vectors based upon HIV-2 or SIV requires a split-genome design. Using this approach, viral proteins required for virion assembly and morphogenesis are expressed from packaging construct(s) stripped of cis-acting sequences which participate in reverse transcription, integration, and RNA encapsidation. These same cis-acting regions are in

92

Lentiviral Vector Systems for Gene Transfer

turn introduced into a transfer vector. Under ideal circumstances co-expression of the packaging construct(s) and transfer vector results in encapsidation of only the transfer vector. Because the transfer vector lacks elements essential to viral replication, the vector particles produced are replication defective and are capable of only a single round of transduction. In practice, the extent to which vector prepared through use of a split genome design remains free of helper virus depends largely upon the precision with which cis-acting sequences required for RNA encapsidation, reverse transcription and integration are removed from the core packaging construct. We previously described development of a replication defective VSV-G pseudotyped HIV2 vector based upon HIV-2KR.54 HIV-2KR is a molecular clone that is infectious but apathogenic in pig-tailed macaques, making vectors derived from this molecular clone amenable to study within animal models. We made use of a three plasmid transient transfection system, similar to other transiently produced lentiviral vectors. The viral structural, enzymatic and regulatory functions are provided by a packaging construct. This construct contains deletions of cis-acting sequences required for encapsidation, reverse transcription and integration. Also deleted is a 776 bp fragment of env spanning the V3 loop. VSV-G is expressed from a separate construct to allow for vector pseudotyping. As previously indicated, pseudotyping of lentiviral vectors with VSV-G broadens the vector target cell range and makes it possible to concentrate retroviral vectors by ultracentrifugation. The final construct is the transfer vector which expresses a reporter gene within the context of viral cis-acting sequences required for encapsidation, reverse transcription and integration. The packaging construct we originally described expresses the viral gag, pol, tat, rev, vif, vpr, vpx and nef genes under control of the HIV-2 LTR (Fig. 2). A heterologous bovine growth hormone (BGH) polyadenylation signal replaces the 3' LTR, with the BGH p(A) being cloned precisely at the stop codon of the nef gene. A large portion of the env gene which spans the V3 loop is deleted without affecting the tat and rev coding exons or the RRE. Lastly, 61 nucleotides of the 75 nucleotide packaging signal are deleted. We found deletion of 61 of the 75 nucleotides which comprise the HIV-2KR packaging sequence sufficient to dramatically attenuate encapsidation of the genome as measured by RNase protection assay of RNA isolated from both transfected cells and from viral pellets. More importantly, no evidence of viral replication was detected over a 6 month period when proviral clones containing this deletion (with or without an additional deletion of the 3' LTR) were transiently transfected into a highly permissive T cell line. The final packaging construct, however, permitted wild-type levels of protein expression as measured by p26 ELISA. Several other recent studies point to the involvement of viral sequences outside of ψ in HIV and SIV encapsidation. For HIV-2, a 46 nucleotide deletion in ψ was reported to diminish genome encapsidation, but failed to abolish viral replication.52 A more substantial 69 nucleotide deletion of ψ analyzed within this same study had catastrophic effects on LTR driven expression, reducing p26 levels to ~10% of the wild-type control value. In a related study, deletion of the HIV-2 ψ packaging region was reported to have minimal impact on genome encapsidation. This study instead identified sequences upstream of the major splice donor as the principal packaging determinant.55 A subsequent study identified regions both upstream and downstream of the major splice donor which contribute to HIV-2 encapsidation, although no combination of deletions was found to reduce encapsidation below ~30% of the wild-type control value without drastically curtailing protein expression.53 Sequences upstream of the major splice donor have likewise been identified as essential to efficient packaging of SIVmac genomic RNAs.56 These apparent differences in the relative importance of ψ with regard to RNA encapsidation and viral infectivity may reflect variation attributable to genetic divergence of the individual HIV-2 molecular clones investigated. Because most of the desirable targets for gene therapy do not express CD4 and/or the coreceptors recognized by the native primate lentiviral envelopes, we chose to pseudotype HIV-2

HIV-2 and SIV Vectors

93

Fig. 2. A. Design of the HIV-2 packaging construct described in the text. Deletions within the packaging signal and env are indicated. The bovine growth hormone polyadenylation signal is positioned precisely at the translational termination codon of nef, replacing the 3’ LTR. B. The VSV-G expression plasmid, pCMVG. The lacZ and GFP transfer vectors described in the text. GFP is expressed in an antisense orientation relative to LTR expression, whereas lacZ expression is in a sense orientation.

with VSV-G. The receptor for VSV-G is ubiquitous. Consequently, VSV-G pseudotyping of a lentiviral vector widely broadens its tissue tropisms and target cell range. Pseudotyping with VSV-G carries with it additional advantages beyond tissue tropism. Due to the greater stability of VSV-G, pseudotyped particles may be concentrated. From the perspective of biosafety, the use of VSV-G and the absence of an intact HIV-2 env sequence in any of the three plasmids used makes it impossible to reconstitute the native virus through recombination. The transfer vector we described places a lacZ reporter gene within the context of viral cisacting sequences required for encapsidation, reverse transcription and integration. The vector contains 5' and 3' LTRs at its terminii. Viral sequences extend from the 5' LTR to include the leader sequence, ψ, and the first 373 nucleotides of gag. The short stretch of gag is followed by an approximately 1 kb env fragment which includes the RRE. Downstream of the RRE is an SV40 promoter-driven lacZ reporter placed in a sense orientation relative to the viral LTRs. A polyadenylation signal is provided for lacZ by the 3' LTR. Due to high background lacZ staining in some primary cells, we also used a CMV-IE promoter-driven GFP reporter in some instances. The GFP reporter in this vector, however, is placed in an anti-sense orientation relative to the viral LTRs and upstream of the RRE. Vector production using the three plasmid system described resulted in high titer vector (unconcentrated titer >106/mL) capable of efficiently transducing dividing and growth arrested cells, terminally differentiated neurons and primary monocyte derived macrophages. The bio-safety of any vector, however, is significantly improved by eliminating all nonessential elements from the packaging system. This is particularly true for the primate lentiviruses for which the pathogenic determinants remain incompletely defined. For this reason, we subsequently chose to delete the accessory genes within the packaging system. Deletion of vif, vpr,

94

Lentiviral Vector Systems for Gene Transfer

vpx and nef, individually or in combination, resulted in no significant change in vector titer on growth arrested cells or primary macrophages (unpublished data). A recent study of a comparable SIV vector found similarly that elimination of accessory genes within the packaging system had no effect on vector titer on dividing or growth arrested cells.57 These findings are consistent with early reports of HIV-1 vectors, although one study of HIV-1 vectors found expression of vif and vpr to be essential in order to achieve efficient transduction of hepatocytes.58 This last finding suggests that tissue specific requirements for individual accessory gene expression may prove to play an essential role in future vector design.

SIV/HIV Chimeric Vectors One novel approach to vector design involves the development of chimeric SIV/HIV vectors. A variety of such vectors have been described, involving a three plasmid design comparable to other VSV-G pseudotyped lentiviral vectors. In the chimeric vector, however, the packaging components are provided by an SIV-based packaging construct, whereas the transfer vector is based upon HIV. Cross-packaging permits production of vector at titers comparable to that achieved with more traditional lentiviral vectors.56 One such vector was recently shown to efficiently transdue a variety of cell types, including growth arrested cells, primary macrophages and primary mouse neurons.59 The principal benefit SIV/HIV chimeric vectors offer is potentially greater safety. Lack of homology between the transfer vector and the packaging construct eliminates the potential for homologous recombination during vector production. However, this possible benefit must be weighed against the possibility of generating a novel virus.

Requiem and Prospectus Although significant progress has been made in the area of lentiviral vector design, several exciting areas remain to be completely explored. Further efforts are likely to result in substantial modification of existing systems. Currently, much attention is being devoted to establishing minimal lentiviral vectors, as indicated by efforts to eliminate accessory genes from packaging systems. Now that the accessory genes have been shown to be dispensable in a number of vector systems, several investigators have turned to the regulatory genes to determine if they too might be eliminated. Particular attention has been given to the potential for replacing the RRE with the constitutive transport element (CTE) of type D retroviruses. Like RRE, CTE regulates the shuttling of viral RNAs from nucleus to cytoplasm. However, CTE makes use of an endogenous cellular pathway to perform this function, making it independent of additional viral proteins such as Rev.60 To date, a bare handful of reports have described replacing RRE with CTE in either the packaging construct or the transfer vector.61,62 Attempts to incorporate CTE within packaging constructs based upon HIV or SIV have unfortunately resulted in significant reduction of Gag-Pol expression and a 2-3 log reduction in vector titer. Attempts to incorporate CTE into a HIV-1 transfer vector have met with somewhat more success. One study reports obtaining titers comparable to more traditional Rev-dependent vectors, depending upon the placement of CTE relative to the 3’ LTR and the presence of an additional mutation of the major splice donor within the transfer vector. It is feasible to envision, then, a lentiviral transfer vector in which the LTRs and a small fragment of gag are all that remain of viral sequence. From the perspective of clinical safety, a minimal vector of this sort represents an ideal. Other recent efforts have focused on the design of self-inactivating (SIN) vectors.57,63 SIN vectors contain a deletion of U3 within the 3’ LTR of the transfer vector. Due to the template switching mechanism of reverse transcription, U3 is removed from both LTRs within the integrated vector DNA. Consequently, expression from the transfer vector within transduced cells occurs via internal promoters and not the LTR. This approach offers two distinct advantages. It prevents promoter interference within transduced cells and, it blocks expression of viral sequences, including the short stretch of gag included in all lentiviral vectors as part of the

HIV-2 and SIV Vectors

95

encapsidation signal. A number of SIN vectors based upon HIV and SIV have been described in recent years. In all cases the titers obtained are reported to be comparable to their wild-type counterparts. SIN vectors, for these reasons, are rapidly becoming a standard element within lentiviral packaging systems. Whereas the incorporation of SIN mutations within vector systems insures that viral sequences are not expressed within transduced cells, some investigators have begun to examine the use of tissue-specific promoters to insure transgene expression occurs only within the desired target cell population. Use of tissue-specific promoters to drive transgene expression provides lentiviral vectors with a greater degree of specificity and alleviates concern regarding the potential hazards associated with constitutive transgene expression within inappropriate tissues. The specificity engendered by such vector design is evident in one study by Miyoshi et al, in which the rhodopsin promoter was used to drive GFP expression.64 Use of the rhodopsin promoter resulted in photoreceptor-specific expression of GFP upon direct subretinal injection of the vector, whereas CMV-driven GFP expression was detected within a variety of cells in the subretinal compartment. Similar approaches may be envisioned for other tissues. For any vector to be clinically relevant in broad terms, however, it may be necessary to establish stable packaging lines for vector production. The transient vector preparations typically described in the literature are not readily amenable to characterization or bulk preparation, potentially limiting their applicability within a clinical setting. Stable packaging lines, conversely, may be extensively characterized and are well suited to scaling up of production. Both features may be helpful for establishing necessary quality control prior to clinical use of lentiviral vectors. The design of forthcoming vectors is likely to incorporate aspects of each of the aforementioned features. Generation of a truly minimal packaging system will improve vector bio-safety. The use of SIN vectors within this context adds an additional layer of safety by eliminating virally directed gene expression within transduced cells. Incorporation of tissue-specific promoters within transfer vectors adds an essential level of precision to vector design, targetting transgene expression to the desired population. And, lastly, incorporation of all of these elements within a stable packaging line will allow for extensive vector characterization and bulk preparation. Obviously the benefits of these approaches to improving vector design and/or bio-safety are not limited to HIV-2/SIV. They are, in fact, desirable features within any lentiviral vector system. Their value with regard to HIV-2 and SIV lies in improving a system that already offers unique benefits not found within other members of the lentiviral family. Again, these benefits include a drastically diminished pathogenicity relative to HIV-1, an amenability to study within primate models susceptible to disease, and a separation of nuclear import and cell cycle arrest functions. Further, being primate lentiviruses, HIV-2 and SIV are among the most extensively characterized viruses currently known, providing a basis by which to understand vector behavior within primates. Further modification of HIV-2/SIV vector design, in this context, will undoubtedly improve vector bio-safety and efficiency, bringing broad clinical application of gene transfer one step closer to reality.

References 1. Roe T, Reynolds TC, Yu G, Brown PO et al. Integration of murine leukemia virus DNA depends on mitosis. EMBO J 1993; 12:2099-2108. 2. Miller AD, Rosman GJ. Improved retroviral vectors for gene transfer and expression. Biotechniques 1989; 7:980-990. 3. Miller DG, Adam MA, Miller AD. Gene transfer by retrovirus vectors occurs only in cells that are actively replicating at the time of infection. Mol Cell Biol 1990; 10:4239-4242. 4. Lewis PF, Emerman M. Passage through mitosis is required for oncoretroviruses but not for the human immunodeficiency virus. J Virol 1994; 68:510-516.

96

Lentiviral Vector Systems for Gene Transfer

5. Bukrinsky MI, Haggerty S, Dempsey MP et al. A nuclear localization signal within HIV-1 matrix protein that governs infection of non-dividing cells. Nature 1993; 365:666-669. 6. Gallay P, Swingler S, Aiken C et al. HIV-1 infection of non-dividing cells: C-terminal tyrosine phosphorylation of the viral matrix protein is a key regulator. Cell 1995; 80:379-388. 7. Gallay P, Swingler S, Song J et al. HIV-1 nuclear import is governed by phospho-tyrosine mediated binding of matrix to the core domain of integrase. Cell 1995; 83:569-576 8. Heinzinger NK, Bukinsky MI, Haggerty SA et al. The vpr protein of human immunodeficiency virus type 1 influences nuclear localization of viral nucleic acids in non-dividing cells. Proc Natl Acad Sci USA 1994; 91:7311-7315. 9. Fletcher TM, Brichacek B, Sharova N et al. Nuclear import and cell cycle arrest functions of the HIV-1 vpr protein are encoded by two separate genes in HIV-2/SIV(SM). EMBO J 1996; 15:61556165. 10. Buchschacher GL, Panganiban AT. Human immunodeficiency viruses for expression of foreign genes. J Virol 1992; 66:2731-2739 11. Parolin C, Dorfman T, Palu G et al. Analysis in human immunodeficiency virus type 1 vectors of cis-acting sequences that affect gene transfer into human lymphocytes. J Virol 1994; 68:3888-3895. 12. Carroll R, Lin JT, Dacquel EJ et al. A human immunodeficiency virus type 1 (HIV-1) based retroviral vector system utilizing stable HIV-1 packaging cell lines. J Virol 1994; 68:6047-6051. 13. Naldini L, Blomer U, Gallay P et al. In vivo delivery and stable transduction of nondividing cells by a lentiviral vector. Science 1996; 272:263-267. 14. Chen Z, Telfer P, Gettie A et al. Genetic characterization of new west African simian immunodeficiency virus SIVsm: geographic clustering of household-derived SIV strains with human immunodeficiency type 2 subtypes and genetically diverse viruses from a single feral sooty mangabey troop. J Virol 1996; 70: 3617-3627. 15. Desrosiers RC, Daniel MD, Li Y. HIV related lentiviruses of nonhuman primates. AIDS Res Human Retroviruses 1989; 5: 465-473. 16. Peeters M, Koumare B, Mulanga C et al. Genetic subtypes of HIV type 1 and HIV type 2 strains in commercial sex workers from Bamako, Mali. AIDS Res Human Retroviruses 1998; 14:51-58. 17. Chen Z, Luckay A, Sodora D, Telfer P et al. HIV-2 seroprevalence and characterization of a new HIV-2 genetic subtype (F) within the natural range of SIV infected sooty mangabeys. J Virol 1996; 71:3953-3960. 18. Gardner, MB, Endres M, Barry P. Simian Retroviruses: SIV and SRV. In: Levy J, ed. The Retroviridae. New York: Plenum Press, 1994:133-276. 19. Fauci AS, Desrosiers RC. Pathogenesis of HIV and SIV. In: Coffin JM, Hughes SH, Varmus HE, eds. Retroviruses. New York: Cold Spring Harbor Laboratory Press, 1997: 587-631. 20. Gayle HD, Gnaore E, Adjorlolo G et al. HIV-1 and HIV-2 infection in children in Abidjan, Cote d’Ivoire. J Acquir Immune Defic Syndr 1992;5:513-7. 21. Dalgleish AG, Beverly PCL, Chapham PR et al. The CD4 (T4) antigen is an essential component of the receptor for the AIDS retrovirus. Nature 1984; 312:763-767. 22. Sattentau QJ, Chapham PR, Weiss RA et al. The human and simian immunodeficiency viruses HIV-1, HIV-2 and SIV interact with similar epitopes on their cellular receptor, the CD4 molecule. AIDS 1988; 2:1-9. 23. Alkhatib G, Broder CC, Combadiere C et al. CC CKR5: A RANTES, MIP-1alpha, MIP-1beta receptor as a fusion cofactor for macrophage tropic HIV-1. Science 1996; 272:1955-1958. 24. Feng Y, Broder CC, Kennedy PE et al. HIV-1 entry cofactor: Functional cDNA cloning of a seven-transmembrane, G protein-coupled receptor. Science 1996; 272:872-877. 25. Chen Z, Zhou P, Ho D et al. Genetically divergent strains of simian immunodeficiency virus use CCR5 as a coreceptor for entry. J Virol 1997; 71:2705-2715. 26. Connor RI, Sheridan KE, Ceradini D et al. Change in coreceptor use correlates with disease progression in HIV-1 infected individuals. J Exp Med 1997; 185:621-628. 27. McKnight A, Dittmar MT, Moniz-Periera J et al. A broad range of chemokine receptors are used by primary isolates of human immunodeficiency type 2 as coreceptors with CD4. J Vir 1998; 72:4065-4071.

HIV-2 and SIV Vectors

97

28. Guillon C, Van Der Ende ME, Boers PHM et al. Coreceptor usage of human immunodeficiency type 2 primary isolates and biological clones is broad and does not correlate with their syncitiuminducing capacities. J Vir 1998; 72:6260-6263. 29. Deng H, Unutmaz D, Kewelramani VN, Littman DR. Expression cloning of new receptors used by simian and human immunodeficiency viruses. Nature 1997; 388:296-300. 30. Ho DD, Neumann AU, Perelson AS,et al. Rapid turnover of plasma virions and CD4 lymphocytes in HIV-1 infection. Nature 1995; 373:123-126. 31. Wei X, Ghosh SK, Taylor ME et al. Viral dynamics in human immunodeficiency type 1 infection. Nature 1995; 373:117-122. 32. Marlink R, Kanki P, Thior I et al. Reduced rate of disease development after HIV-2 infection as compared to HIV-1. Science 1994; 265:1587-1590. 33. Kanki PJ, Travers KU, Mboup S et al. Slower heterosexual spread of HIV-2 than HIV-1. Lancet 1994; 343:943-946. 34. Pepin J, Morgan G, Dunn D et al. HIV-2-induced immunosuppression among asymptomatic West African prostitutes: Evidence that HIV-2 is pathogenic, but less so than HIV-1. AIDS 1991; 5:1165-1172. 35. Barnett SW, Murthy KK, Herndier BG et al. An AIDS-like condition induced in baboons by HIV-2. Science 1994; 266:642-646. 36. Otten RA, Brown BG, Simon M et al. Differential replication and pathogenic effects of HIV-1 and HIV-2 in Macaca nemestrina. AIDS 1994; 8:297-306. 37. Looney DJ, McClure J, Kent SJ et al. A minimally replicative HIV-2 live-virus vaccine protects M nemestrina from disease after HIV-2287 challenge. Virology 1998; 242:150-160. 38. Castro BA, Nepomuccno M, Lerche NW,et al. Persistent infection of baboons and rhesus monkeys with different strains of HIV-2. Virology 1991; 184:219-226. 39. Mamounas M, Looney DJ, Talbott R et al. An infectious chimeric human immunodeficiency virus type 2 (HIV-2) expressing the HIV-1 principal neutralizing determinant. J Virol 1995; 69:6424-6429. 40. Pavlakis GN. The molecular biology of human immunodeficiency virus type 1. In: DeVita JVT, Hellman S, Rosenberg SA, eds. AIDS: Biology, Diagnosis, Treatment and Prevention. 4th ed. Philadelphia: Lippincott-Raven, 1996. 41. Arya SK. Human immunodeficiency virus type 2 (HIV-2) trans-activator (Tat): Functional domains and the search for trans-dominant negative mutants. AIDS Research and Human Retroviruses 1993; 9:839-848. 42. Emerman M, Guyader M, Montangier L et al. The specificity of the human immunodeficiency virus type 2 trans-activator is different from that of human immunodeficiency type 1. EMBO J 1987; 6:3755-3760. 43. Luciw P. Human immunodeficiency viruses and their replication. In: Fields BN, ed. Virology. New York: Raven Press, 1995:1881-1952. 44. Lewis N, Williams J, Rekosh D et al. Identification of a cis-acting element in human immunodeficiency virus type 2 (HIV-2) that is responsive to the HIV-1 rev and human T-cell leukemia virus types 1 and 2 rex proteins. J Virol 1990; 64:1690-1697. 45. Jacque JM, Fernandez B, Arenzana-Seisdedos F, Thomas, D et al. Permanent occupancy of the human immunodeficiency virus type 1 enhancer by NF-B is needed for persistent viral replication in monocytes. J Virol 1996; 70:2930-2938. 46. Alcami J, Lain de Lera T, Folgueira L et al. Absolute dependence onB responsive elements for initiation and Tat-mediated amplification of HIV transcription in blood CD4 T lymphocytes. EMBO J 1995; 14:1552-15560. 47. Baeuerle PA, Kenkel T. Function and activation of NF-kappa B in the immune system. Ann Rev Immunol 1994; 12:141-179. 48. Thanos D, Maniatis T. NF-kappa B: A lesson in family values. Cell 1995; 80: 529-532. 49. Linial ML, Miller AD. Retroviral RNA packaging: Sequence requirements and implications. In: Swanstrom R, Vogt PK, eds. Retroviruses—-Strategies of Replication. Berlin: Springer Verlag, 1990:125-152. 50. Berkowitz RD, Hammarskjold ML, Helga-Maria C. 5’ regions of HIV-1 are not sufficient for encapsidation: implications for the HIV-1 packaging signal. Virology 1995; 212:718-723.

98

Lentiviral Vector Systems for Gene Transfer

51. McBride MS, Panganiban AT. The human immunodeficiency virus type 1 encapsidation site is a multipartite RNA element composed of functional hairpin structures. J Virol 1996; 70:2963-2973. 52. Garzino-Demo A, Gallo RC, Arya SK. Human immunodeficiency virus type 2 (HIV-2): Packaging signal and associated negative regulatory element. Hum Gene Ther 1995; 6:177-184. 53. Arya SK, Zamani M, Kundra P. Human immunodeficiency type 2 lentivirus vectors for gene transfer: expression and potential for helper virus-free packaging. Hum Gene Ther 1998; 9:1371-1380. 54. Poeschla E, Gilbert J, Li X et al. Identification of a human immunodeficiency type 2 (HIV-2) encapsidation determinant and transduction of nondividing human cells by HIV-2-based lentivirus vectors. J Virol 1998; 72:6527-6536. 55. McCann EM, Lever AM. Location of cis-acting sequences important for RNA encapsidation in the leader sequence of human immunodeficiency virus type 2. J Virol 1997; 71:4133-4137. 56. Rizvi TA, Panganiban AT. Simian immunodeficiency virus RNA is efficiently encapsidated by human immunodeficiency type 1 particles. J Virol 1993; 67:2681-2688. 57. Schnell T, Foley P, Wirth M et al. Development of a self-inactivating, minimal lentivirus vector based on simian immunodeficiency virus. Hum Gene Ther 2000; 11:439-447. 58. Kafri T, Blomer U, Peterson DA et al. Sustained expression of genes delivered directly into liver and muscle by lentiviral vectors. Nat Genet 1997; 17:314-317. 59. White SM, Renda M, Nam N et al. Lentivirus vectors using human and simian immunodeficiency virus elements. J Vir 1999; 73:2832-2840. 60. Tang H, Gaietta GM, Fischer WH et al. A cellular cofactor for the constitutive transport element of type D retrovirus. Science 1997; 276:1412-1415. 61. Mautino MR, Ramsey WJ, Reiser J et al. Modified human immunodeficiency virus-based lentiviral vectors display decreased sensitivity to trans-dominant rev. Hum Gen Ther 2000; 11:895-908. 62. Mautino M, Keiser N, Morgan RA. Improved titers of HIV-based lentiviral vectors using the SRV1 constitutive transport element. Gene Ther 2000; 7:1421-1424. 63. Miyoshi H, Blomer U, Takahashi M et al. Development of a self-inactivating lentiviral vector. J Vir 1998; 72:8150-8157. 64. Miyoshi H, Takahashi M, Gage FH et al. Stable and efficient gene transfer into the retina using and HIV-based lentiviral vector. PNAS, USA 1997; 94:10319-10323.

CHAPTER 6

FIV Vector Systems Sybille L. Sauter and Mehdi Gasmi

Abstract

W

hy is feline immunodeficiency virus (FIV) such an appealing candidate for gene therapy vector development? Phylogenetic analysis suggests FIV is only distantly related to the primate lentiviruses, and despite repeated exposure, neither seroconversion nor other detectable evidence of human infection occurs. FIV naturally infects diverse Felidae worldwide, including the domestic cat. Here, the disease progression parallels the immunodeficiency caused by HIV, and for that reason, FIV and the cat provide an excellent model for anti-virals and AIDS vaccine research. Simple genome organization also facilitates vector development and analysis: FIV has only three accessory/regulatory proteins. To overcome FIV’s cat-specific tropism, feline vectors are equipped with hybrid LTRs, since the FIV LTR shows low activity in human cells. Recombinant FIV vectors generate titers comparable to other lentiviral systems, are capable of incorporating heterologous envelopes and efficiently transduce dividing and nondividing cells in the presence and absence of the accessory proteins in vitro. Compared to HIV vectors, FIV vector development is still in its infancy, but initial in vivo data in various species and tissues indicate long-term gene expression at therapeutic levels, and thus FIV vectors hold great promise. Future efficacy studies in animal models and primates will determine the FIV vectors’ suitability for gene therapy. The design of recombinant FIV vectors incorporates safety features described for primate lentiviral vectors with the benefit that biosafety testing of FIV vectors can occur in the natural host. Currently, FIV vectors are generated in a transient fashion, but the availability of a stable producer system amenable to better characterization and scale-up will considerably increase the potential for use of FIV vectors in the clinic.

Epidemiology and Pathogenesis of FIV Infection FIV was initially isolated in Davis, California from the peripheral blood lymphocytes (PBLs) of a domestic cat (Felis catus) presenting a syndrome of immunodeficiency transmissible to specific pathogen free (SPF) cats.1 Structural and biochemical virology studies revealed FIV to be a retrovirus with particle morphology, Mg2+-dependent reverse transcriptase and genome organization characteristic of a lentivirus. FIV infection has been observed worldwide with peak incidences in Australia (21%)2, Japan (30%)3 and England (47%).4 Prevalence of infection is highest in populations of free-roaming cats and among cats with signs of deteriorated health. While natural transmission mainly occurs horizontally through biting and scratching, sexual and vertical transmissions have also been reported in experimental settings.5,6 FIV strains are classified into five subtypes, A through E, according to envelope protein sequence Lentiviral Vector Systems for Gene Transfer, edited by Gary L. Buchschacher, Jr. ©2003 Eurekah.com.

100

Lentiviral Vector Systems for Gene Transfer

variability.7-9 Subtypes appear to segregate geographically; however, the notion of different pathogenic potential between subtypes is unclear.10,11 Evidence of infection with variant strains of FIV has been observed in free ranging or captive non-domestic felids,12,13 but has not been associated with pathology14 except in the case of the Pallas cat.15

Clinical Manifestations of FIV Infection The clinical course of FIV infection in cats is divided into four stages that coincide almost exactly to those defined for HIV infection in humans (for reviews see 16,17). Primary infection (stage 1) may be associated with an acute viral infection syndrome, including generalized lymphadenopathy, fever and neutropenia. This acute infection phase is in most cases followed by a several-year-long clinically silent period referred to as the asymptomatic phase (stage 2), during which the presence of FIV can be detected in peripheral blood mononuclear cells (PBMCs), plasma and saliva of infected animals.18 The asymptomatic phase is followed by a third period (stage 3) characterized by various nonspecific clinical signs such as persistent pyrexia, generalized lymphadenopathy and secondary chronic oral infections. At this stage, hematology shows a decrease in absolute white cell counts, and an inversion of the CD4/CD8 ratio due to a decline in the CD4 positive lymphocyte subset. Stage 4 corresponds to the full-blown immunodeficiency, brought about by the rapid drop in CD4 positive cells. This period develops in several months and results in the emergence of opportunistic infections, neurological disorders and tumors of various etiologies, which cause the rapid demise of the animal. Throughout the course of the disease, the increase in plasma FIV RNA copy number correlates with the decline in CD4 cells and serves as an indicator of disease progression.19

Cellular Tropism and Virus Receptor FIV possesses a broad tropism for PBMCs and cells derived from the monocyte/macrophage lineage in the CNS. In vivo, FIV is found in CD4 positive T lymphocytes, but also in CD8 positive T lymphocytes, B lymphocytes and circulating monocytes.1,20-22 Macrophages, astrocytes and microglia may also be sites of FIV replication.23,24 Certain molecular clones of FIV can productively infect the feline fibroblast cell line CrFK in vitro.25 FIV tropism determinants are found in the viral envelope protein26-29 but also in the accessory proteins Vif and Orf-2, which play an important role in primary target cell infectivity.30,31 While the CD4 positive lymphocyte subset represents a major target of FIV infection in vivo, the feline CD4 molecule on the surface of target cells need not be present for productive infection.32 Interestingly, both primary isolates and CrFK adapted FIV strains can interact specifically with the feline or human chemokine receptor CXCR4 to enter target cells.33-36 When engineered to express human CXCR4 molecules, cells of various species can become permissive for FIV penetration.33,37,38 That CXCR4 can mediate both human and feline immunodeficiency virus cell entry suggests this interaction could be involved in the pathogenicity of these viruses in their respective species. Yet, unlike HIV, CXCR4 utilization by FIV does not seem to correlate with the animals’ clinical status.35,39 To date, only one study suggests the possibility of a CXCR4-independent mechanism for FIV cell entry.34 While further studies are required to confirm this hypothesis, in light of the most recent data, CXCR4 may constitute the major partner of the viral envelope for target cell recognition, with possible intervention of CC type chemokine receptor(s) as recently suggested by Lerner et al.29

Pathogenesis In domestic cats, FIV pathogenesis is primarily characterized by a progressive incapacitation of the immune system. Functional immunodeficiency begins early in the acute phase of infection, as evidenced by a defect in T lymphocyte helper activity prior to any quantitative defect in the CD4 positive T cell population.40 Unlike HIV-1, FIV does not exclusively infect

FIV Vectors

101

CD4 positive cells; therefore, the direct cytotoxic effect of FIV infection observed in vitro41 cannot in itself account for the rapid decline in CD4 cell counts in vivo. Similar to HIV, mechanisms of CD4 cell depletion such as indirect triggering of apoptosis in uninfected cells42,43 and impaired T-cell regeneration due to premature involution of the thymus in juvenile cats44 are induced by FIV infection. In addition, the altered cytokine network induced by FIV infection can also account for immune dysfunction, notably increased expression of IFN-γ, TNF-α and IL-10 mRNAs and decreased expression of type 1 cytokines IL-2 and IL-12 mRNAs45,46 Therefore. similar to HIV-1 infection,47 the apparent decline in Th1 immunity combined with an increase in IL-10 expression seems to constitute a key factor in FIV-induced immunodeficiency. Another clinical hallmark of FIV pathogenesis is the development of neurological abnormalities in some infected animals,1 similar to what is seen in HIV-1 infection.48 The first direct evidence of FIV neurotropism was reported by Dow et al (1990) who were able to isolate FIV from cultures of neural cells from different areas of infected animals’ central nervous system (CNS).24 Anatomically, FIV-infected animals show signs of neural tissue injury with loss of neurons in the frontal cortex49,50 and gliosis of gray and white matter.49,51 Neurological symptoms include altered behavior and are correlated to signs of dysfunctional physiology of the central and peripheral nervous systems.52-57 Like HIV-1, FIV is believed to penetrate the CNS via infected cells capable of crossing the blood-brain barrier (BBB), although the disruption of the BBB observed in some animals during acute infection may also allow penetration of cellfree FIV particles in the CNS.55 To date there is no evidence of direct neuronal infection by FIV, although, astrocytes and microglia have been found to be infected in vivo, however at low levels.51,58 FIV’s neurotoxicity mechanisms are poorly understood, however FIV infection inhibits astrocytes’ uptake of glutamate, whose increased concentration in the microenvironment is responsible for neuron damage.56,59,60 FIV infection also induces abnormal microglial expression of cytokines, notably TNF-α, which is believed to play an important role in FIV neuropathogenesis, as it is the case for HIV-1 and SIV infections.61,62

Immune Response to FIV Infection Most studies focusing on the development of an immune response to FIV infection derive from experimental infection of SPF cats. A robust humoral response to various FIV antigens appears as early as two weeks post-infection and persists throughout the clinical stages of the disease.63 Antibodies capable of neutralizing FIV replication and syncytia formation in vitro are detectable in sera of symptomatic as well as asymptomatic animals;18,64-68 however, their role in the control of FIV infection and pathogenesis is unclear. High-titer virus neutralizing antibody (VNA)-containing sera capable of inhibiting FIV infection of CrFK cells did not inhibit infection of lymphoid cells in vitro. Furthermore, FIV particles preincubated in VNAcontaining-sera remained infectious in vivo.69 On the other hand, animals could be protected from infection after passive immunization with sera of infected cats or of cats immunized against FIV epitopes of the same virus isolate.70 Regarding the cellular immune response to FIV infection, MHC class I-restricted CD8 positive cytotoxic T lymphocyte (CTL) precursors to FIV antigens are detected in stimulated circulating lymphocytes early after infection, but only in lymph nodes during the later course of infection.71 A non-cytolytic CD8 mediated antiviral activity is observed in the peripheral blood lymphocytes of infected cats.72-74 This anti-FIV cellular response is not restricted by MHC and appears to be mediated by a soluble factor(s) secreted by CD8 positive cells.75 This factor exerts its antiviral activity at the level of viral expression.73 Such an activity was also demonstrated for HIV infection76 and is believed to constitute a mechanism whereby viral load is reduced without eliminating infected cells, which avoids further depletion of CD4 positive lymphocytes.77

102

Lentiviral Vector Systems for Gene Transfer

FIV Vaccines In the past decade, many types of vaccine preparations consisting of either inactivated virus, inactivated FIV infected cells, subunit recombinant proteins, peptides, plasmid DNA or viral vectors encoding FIV epitopes have been tested in cats for their potential protective effect against FIV infection (for review see ref. 78).79-81 Inactivated whole virus preparations and infected cells have been shown to successfully protect animals from infection with homologous or closely related isolates.82-84 Interestingly, subunit vaccines administered either as recombinant proteins alone or via immuno-stimulating complexes have shown little or no protective activity, and in some cases were associated with increased susceptibility to subsequent challenge infection.85-87 More recently, DNA immunization with a live attenuated FIV showed protection from infection with wild-type homologous virus.88 When achieved, long-term protection appears to be mediated by CTLs,89,90 which can be transferred to naïve animals and confer protection to homologous challenge in a MHC restricted manner.91 While high levels of virus-neutralizing antibodies in vaccinated animals do not correlate with enhanced protection from challenge by homologous strains, they do seem to be important for relative protection against infection by heterologous strains.92 Overall, FIV vaccination studies parallel those derived from nonhuman primate lentivirus studies in that broadly cross-neutralizing intersubtype protection constitutes a challenging goal. Still, recent encouraging results suggest, that vaccination trials in the FIV/cat system could provide important information useful for the design of a candidate vaccine for HIV infection.92,93

Antiviral Drugs Given the common structural and biochemical properties between HIV and FIV enzymatic proteins, drugs that inhibit HIV replication, such as reverse transcriptase (RT) and certain protease inhibitors, also inhibit FIV enzymes activity in vitro.94-98 In vivo, RT inhibitors improve clinical manifestations of FIV infection, notably in young cats.99,100 Unfortunately, the similarities between HIV and FIV extend in their capacity to develop drug resistance by evolving escape mutants harboring amino acid substitution in the protein sequences of RT.101,102

FIV Genome Organization and Pattern of Expression

FIV molecular clones were sequenced soon after the first description of the virus.103,104 Phylogenetic analyses have revealed that FIV is more closely related to non-primate lentiviruses than to the HIV/SIV subgroup,103 although as described above, pathologies associated with FIV infection are more reminiscent of those of HIV and SIV than of those of the equine infectious anemia virus (EIAV), visna-maedi virus (VMV) and caprine arthritis-encephalitis virus (CAEV). The 9.2 kilobase genome of FIV harbors typical retroviral/lentiviral structures (Fig. 1). As for all retroviruses, the FIV proviral genome is flanked by two long terminal repeats (LTRs) generated by the duplication of the genomic terminal sequences during reverse transcription. The coding sequences include the gag, pol and env retroviral genes along with various other open reading frames (ORFs), whose expression relies on alternative splicing mechanism (Fig. 2). Most of these ORFs encode accessory/regulatory proteins that have been characterized and have revealed striking structural and/or functional similarities with proteins encoded by other lentiviruses more or less distant in evolution.

Long Terminal Repeats (LTR) The 5’ LTR U3 region of the FIV proviral genome carries the viral enhancer/promoter that drives transcription of FIV RNA in infected cells. Transcription regulatory elements consist of putative binding sequences for the transcription factors AP-1, AP-4, ATF, C/EBP NF1 and NFkB.105 In addition, two imperfect direct repeats, a CCAAT box and a TATA box have

FIV Vectors

103

Fig. 1. FIV proviral genome organization relative to other lentiviruses. The relative positions of characterized open reading frames and Rev responsive elements within each genome are shown.

also been identified in the promoter region of U3. DNaseI footprinting experiments have revealed the AP-1, AP-4, ATF and C/EBP elements’ capacity to interact with cellular proteins, although their binding properties appear to vary in different cell types.106-108 AP-1, ATF and C/EBP sites are important for basal promoter activity in various in vitro systems.105,106,108,109 The ATF binding site is responsible for FIV LTR increased activity following stimulation of the protein kinase A signal transduction pathway and the AP-1 site confers responsiveness of the viral promoter to phorbol-esters and to lymphoid cell activation signals.105,110 Deleting the AP-1 site does not impair FIV replication in feline PBMCs or macrophages,110,111 however virus replication and pathogenesis are affected in vivo.112 On the other hand, deleting the ATF1 binding site alone or in combination with an AP-1 deletion is detrimental to virus replication in PBMCs and macrophages111 consistent with the transactivation studies by transfection.106 Furthermore, deleting the entire enhancer region virtually abolishes virus replication in otherwise permissive cells. Various studies conducted to determine whether the FIV genome encoded its own transactivator had led to conflicting results about the orf2 gene product’s ability to activate FIV LTR driven expression.30,105,108 However, de Parseval et al recently have brought strong evidence that Orf-2 was capable of transactivating the basal activity of FIV enhancer/promoter by up to 20-fold depending on the cell type.113 The mechanism of this transactivation observed in feline as well as in human fibroblasts has yet to be defined. While no direct binding of Orf2 protein to the FIV LTR has been observed, deletions of the AP-1, C/EBP tandem or ATF sites alone or the deletion of the region encompassing AP-1 through ATF decreased or abolished the transactivating effect of Orf2 respectively, suggesting the involvement of these regions in Orf2mediated LTR transactivation.

104

Lentiviral Vector Systems for Gene Transfer

Fig. 2. FIV genome expression pattern. Functional splice sites generating alternately spliced transcripts are shown. The size of transcripts and the nature of encoded genes are indicated. p15rev was recently identified by de Pareseval et al and was shown to exert partial Rev activity in feline cells.113

Structural Proteins (gag Gene Products) The FIV Gag polyprotein (50kDa) is issued from the translation of the unspliced genomic RNA (Fig. 2). The viral protease processes it into three proteins—matrix (MA), capsid (CA) and nucleocapsid (NC)—that form the mature FIV particle. Mass spectrometry analyses allowed the determination of exact proteolytic sites within the Gag precursor.114 MA (14 kDa) is issued from the N-terminal portion of the polyprotein by cleavage between Tyr 135 and Pro 136. The protein is myristoylated at the Gly 2 residue, which promotes its interaction with the inner layer of the cell membrane, similar to some other retroviruses.114,115 As observed for HIV-1, a double cleavage between CA (24kDa) and NC (7 kDa) releases both proteins as well as a 2 kDa peptide whose function is unknown.116 While FIV lacks the p6 peptide generated by cleavage of the NC C-terminus in HIV, Elder et al have found that the mature form of FIV NC is 2 kDa lighter than the protein deduced from the gag coding sequence,114 presumably due to the proteolytic cleavage of the C-terminal region of NC. In the case of HIV-1, a similar cleavage occurs post-infection and is involved in efficient proviral synthesis. 117 Whether this phenomenon applies to FIV remains to be determined. While the FIV gag gene product shares common features with the primate lentiviruses (pattern of cleavage, myristoylation of MA), antigenic cross reactivity has only been demonstrated between CA of FIV and CA of non-primate lentiviruses,118 which reinforces the discrepancy between FIV evolution links and induced pathogenesis.

Polymerase (pol Gene Products) The pol ORF encodes the viral enzymes protease (PR), reverse transcriptase (RT), deoxyuridine triphosphatase (DU) and integrase (IN). pol translation is achieved by ribosomes that undergo a frameshift induced by a typical RNA tertiary pseudoknot structure in the 3’ end region of NC.119 The resulting molecule is a Gag-Pol multipolyprotein of approximately 158 kDa, matured by the viral protease. As in other retroviruses the synthesis of Gag to Pol products ratio is 20 to 1.114

FIV Vectors

105

Protease (PR) The 13 kDa FIV PR is responsible for processing the Gag and Gag-Pol polyproteins into their mature products. Although FIV and HIV protease share many structural features (homodimeric nature and similar quaternary structure120), they demonstrate distinct specificities for substrates and inhibitors.98,121 Comparative studies between primate lentivirus proteases and the FIV protease, combined with mutagenesis analyses, have led to the identification of common amino acid residues involved in the recognition of protease ligands—important for the development of protease inhibitors that are effective against different virus species and against resistant mutants.97,122,123

Reverse Transcriptase (RT) FIV RT is a heterodimeric enzyme composed of two polypeptides of 66 kDa and 51 kDa, both generated by cleaving the Gag-Pol polyprotein by PR.124 FIV RT structure and proteolytic cleavage sites were predicted by sequence homology with HIV RT and were validated using synthetic substrates containing the putative cleavage sites.114 These studies confirmed that in the C-terminal portion the 66 kDa protein contains the fragment predicted to encode RNase H activity, absent in the 51 kDa subunit. Separate expression of both p51 and p66 in E. coli allowed precise biochemical characterization of subunit assembly and enzymatic activity. Notably, the capacity of FIV RT to perform DNA positive strand displacement, is similar to HIV RT.125 FIV RT is sensitive to nucleoside analog inhibitors, and specific resistance mutations have been identified within the protein sequence.101,102

Deoxyuridine Triphosphatase (DU) The presence of DU within the FIV genome constitutes additional evidence of FIV’s relation to nonprimate lentiviruses, which also encode DU unlike HIV or SIV. The recent determination of the FIV DU crystal structure has revealed that the protein is a trimer of 14.3 kDa subunits.126 DU is found in prokaryotic and eukaryotic cells and promotes the hydrolysis of dUTP to dUMP in order to prevent misincorporation of deoxyuridine during DNA synthesis. FIV DU-defective [DU(-)] mutants can still replicate in CrFK cells but show markedly reduced growth levels in primary macrophages127 as do EIAV DU mutants.128 In vivo, cats infected with DU(-) FIV show lower viral loads, notably in the spleen, and mutation frequency increases in proviral sequences found in lymphocytes, consistent with increased uracil incorporation into viral DNA. In addition, cats infected with a DU(-) mutant developed less pronounced neurological symptoms than cats infected with wild-type FIV.54 Therefore low levels of endogenous DU that could compromise FIV replication in nondividing cells are supplemented by viral DU provided by the virus itself.129

Integrase (IN) The 32 kDa FIV IN is located at the C-terminus of the Gag-Pol polyprotein. FIV IN shows catalytic activities typical of integrase enzymes in vitro, i.e. site-specific cleavage of FIV viral DNA ends, DNA strand-transfer and disintegration.130 Functional studies suggest that FIV IN is a multimer, like other retroviral integrases.131 The FIV IN functional domains include the N-terminal domain, which contains a His-Cys zinc-finger motif involved in DNA interaction, and the central domain that contains the sequence D-X39-58-D-X35-E specific of integrase catalytic sites (for review see ref. 132). Additionally, the FIV IN C-terminus region also plays an important part in the enzyme’s interaction with its DNA substrate.131 An FIV molecular clone harboring a 121 amino-acid deletion in the C-terminus of the protein transfected into CrFK cells produced immature noninfectious particles that contained high levels of unprocessed Gag precursor and no detectable RT activity, indicating a role for FIV integrase protein in the late stages of FIV replication133 as suggested for HIV-1 integrase.134

106

Lentiviral Vector Systems for Gene Transfer

FIV Accessory Protein Vif The vif gene has been found in all lentiviral genomes except for that of EIAV. Although sequence analyses show little homology between vif genes of different lentiviruses, their size and location within viral genomes is well conserved. Vif is expressed from a singly spliced 5.2 kb mRNA (Fig. 2), and is involved in cell-free virus infectivity, as evidenced by the incapacity of Vif deleted mutants [Vif(-)] to replicate in certain cell lines notably CrFK (fibroblasts), G355-5 (embryonic brain cells),135,136 feline PBMCs and primary macrophages.31 However, Vif(-) FIV could be amplified in the lymphocytic Mya-1 cells grown in the presence of CrFK cells transfected with a Vif(-) FIV molecular clone.135 This phenomenon is reminiscent of what is seen in the case of HIV, in that target cells are either permissive or non-permissive for Vif deficient virus growth. In vivo, Vif(-) FIV failed to replicate in PBMCs, although replication was observed in lymph nodes, probably due to cell-to-cell virus transmission enhanced by close cellular contact in the tissue.112 During the course of the study (16 weeks), no immunological or histological change was observed in cats infected with Vif(-) FIV compared to wild-type FIV-infected animals, suggesting the role of Vif in the pathogenesis of FIV infection. Furthermore, cats injected with a Vif(-) FIV molecular clone were protected from challenge infection with a wild-type homologous strain.88 The reasons for the attenuated phenotype of Vif(-) FIV are unknown. Experiments using sensitive detection techniques failed to localize Vif in the virions, but identified the protein in the nucleus of productively infected cells.137 This suggests that the effect of Vif on virus infectivity could occur at the nuclear level prior to virion release from the producer cells.

FIV Accessory Protein Orf2 The 9.3 kDa Orf2 protein is expressed from multiply spliced mRNAs (Fig. 2). Its gene location is similar to that of the first exon of the CAEV and VMV tat gene; consequently it was postulated that the Orf2 protein could constitute a Tat-like factor that stimulates FIV enhancer/promoter basal transcription activity. After several controversial reports, a study reported by de Parseval et al demonstrated the transactivating effect of Orf2 on FIV LTR-driven transcription and confirmed that, like other lentiviruses, FIV encodes its own transactivating factor. Similar to the VMV Tat protein, orf2 encodes a cysteine-rich region in its N-terminus but lacks both the core and basic domains essential for HIV Tat protein transactivation activity.138 Although a stem-loop structure analogous to the Tat responsive element (TAR) of primate lentiviruses has been described in the FIV LTR,139 its involvement in Orf2-mediated transactivation has not yet been determined. Rather, Orf2 transactivation appears to implicate transcription factor binding sites present in the FIV promoter/enhancer,113 similar to VMV.140 Efficient FIV replication in feline T cell lines and in PBMCs141 requires Orf2. In addition, the replication of the molecular clone p34TF10 in T cells is restored after substitution of the orf2 premature stop codon with a tryptophane codon.30 In vivo, cats infected with an Orf2 defective virus exhibit delayed antibody response to FIV and a lower proviral load in various tissues, in comparison to cats infected with wild-type virus.112 The exact mechanism by which Orf2 enhances FIV pathogenesis in infected cats is at present unknown.

Envelope FIV Env is encoded in a singly spliced mRNA of 4.4 kb (Fig. 2). FIV Env precursor processing into a glycoprotein arranged in a non-covalent association between SU (gp120) and TM (gp40) subunits results from post-translational modifications common to all retroviral envelopes. However, like CAEV Env, FIV env undergoes an additional proteolytic processing that is responsible for the cleavage of a 20 kDa polypeptide present at the N-terminus of the SU domain142 and which includes the 80 N-terminal residues of the rev first exon. Little is known about this 20 kDa fragment structure or about its subcellular localization. However,

FIV Vectors

107

Vahlenkamp et al recently reported that a FIV mutant partially deleted in the 20 kDa polypeptide region exhibited dramatically decreased replication levels in feline primary astrocytes and in feline astrocyte-derived cell lines, while replication of the same mutant was unaffected in other cell types including PBMCs and CrFK cells.143 Whether the deletion impaired the viral envelope function in astrocytes at an early or late stage of viral replication remains to be determined. Env is the most divergent structural protein of FIV.144 Similar to HIV, FIV Env is structured in conserved and variable regions.145,146 Nine variable regions are scattered throughout the FIV Env molecule: two of them (V1, V2) are located in the 20 kDa domain, four (V3-V6) in the SU domain and three (V7-V9) in the TM coding region. Domains V1 through V5 are classified as hypervariable;146 V3 and V4 domains as well as the ectodomain of TM of FIV Env are implicated in FIV tropism.26-28 As with HIV, a principal neutralization domain and an immunodominant domain have been identified in FIV V3 and in the ectodomain of TM respectively.147 FIV Env also holds virus tropism determinants within its sequence26-29 and although HIV and FIV interact with CXCR4 to enter target cells, they share little homology in their envelope protein sequences. Interestingly, Serres148 recently reported structural and physical analogies between the trimeric models of TM ectodomains of HIV, SIV and FIV and the IL-2 molecule of their respective species. These findings suggest that three moreorless distantly related immunodeficiency viruses have evolved a common structural feature to specifically interact with their host target cells. Further studies are needed to determine to what extent this contributes to FIV, HIV and SIV pathogenesis.

Rev and the Rev Responsive Element The rev gene is found in all lentiviral genomes, and its product—the Rev protein—is a key factor in lentivirus replication. The 29 kDa Rev protein is encoded in two exons present on multiply spliced mRNAs (Fig. 2). The first exon of rev overlaps and shares the same coding frame as the 5’ region of the env gene. The second rev exon is located downstream of the env coding region and extends into the U3 region of the 3’LTR. Much of what is known about HIV Rev function is applicable to the Rev proteins of all lentiviruses (for review see Chapter 2 and refs. 149,150). Rev constitutes the switch that triggers late gene expression in the virus life cycle when present at a sufficient concentration within the cell nucleus. Rev promotes the nuclear export of unspliced and singly spliced mRNAs, i.e., the genomic RNA which also encodes Gag-Pol and the Vif and Env mRNAs, enabling their translation. Rev function is mediated by its interaction with a highly structured sequence, the Rev responsive element (RRE) present on target mRNAs. The RRE typically located rather internally within the envcoding region of the different lentiviruses was mapped in the 3’ end of the FIV env gene.151 Various HIV-1 Rev functions have been attributed to different domains of the protein. A highly basic amino acid domain responsible for RRE-binding and nucleolar localization and a nuclear export signal (NES) rich in leucine residues involved in Rev nuclear export allow Rev shuttling in and out of the nucleus; a third domain is involved in HIV Rev multimerization. FIV Rev has been shown to contain a basic domain151 and an atypical NES that can substitute for HIV NES.152 If it exists, the FIV Rev multimerization domain remains to be identified. Contrary to early reports, the FIV Rev/RRE system is functional in human cells, provided sufficient expression levels are achieved.37

Development of FIV-Based Vectors Why develop non-primate FIV vectors? To provide an alternative to the primate lentiviral vectors, and offer possible solutions to some issues raised with HIV-based vectors. FIV is nonpathogenic in humans; despite exposure to FIV, seroconversion in human populations has not occurred,153-156 and phylogenetic analysis suggests FIV is only distantly related to primate lentiviruses.10,103,104 The generation and in vitro evaluation of FIV-based vectors, first described

108

Lentiviral Vector Systems for Gene Transfer

in 1998,157 was followed by the optimization of FIV vector technology including the first in vivo studies in 1999.158 Curran et al describe a potential benefit of FIV vectors, suggesting that FIV does not efficiently cross-package HIV genomic RNA. If confirmed, this trait may be desirable for the gene therapy of HIV-infected subjects.159 Further in vivo studies established FIV vectors as an efficient gene therapy vector. Transduction of the primate and murine central nervous system160-162 and rabbit airway epithelium have resulted in longterm expression at therapeutic levels. Although these initial studies were carried out with VSV-G pseudotyped FIV vectors, FIV vectors have also been successfully pseudotyped with the amphotropic envelope. In fact, MLV, HIV-1 and FIV-based vectors carrying the amphotropic envelope have a prolonged half-life in human serum compared to these vectors pseudotyped with VSV-G,163 a particularly important finding for those applications requiring direct injection into blood vessels. When FIV vectors were developed, relatively little was known about the sequence requirements for efficient packaging of the FIV RNA genome or the activity of certain cis elements in human cells. By analogy with other lentiviruses, the location of the putative packaging signal for FIV was suspected downstream of the 5’LTR, possibly extending into gag coding sequences. More was known about the activity of the FIV promoter in human cells: reports described that feline-specific tropism of FIV is governed by low transcriptional activity of the FIV LTR in human cells as well as its envelope.105,141,164-167 More recent data indicate that wild-type FIV can penetrate into certain human cells via the interaction of its envelope with human CXCR4. While there are reports of FIV infection of human cells without spreading of the virus in culture,164,167,168 low levels of infection of human primary PBMCs are observed albeit without evidence for FIV integration into the genome. Currently, we still have little understanding as to what receptor(s) other than CXCR4 may be involved in FIV wild-type infection, and because of that we suggest the state-of-the-art safety measures developed for other lentiviral vectors will also be incorporated into FIV vectors despite any evidence for disease in humans caused by exposure to FIV. Low FIV LTR activity in human cells along with other potential roadblocks caused researchers to focus on designing an FIV vector system to overcome possible restrictions inherent to the FIV virus. The first steps in FIV vector development addressed basic questions such as i) the ability to accommodate heterologous human-tropic envelopes; ii) FIV LTR activity in human cells; iii) the packaging signal’s size and location; iv) the need for RNA export elements in the transfer vector and packaging constructs; and v) the influence of accessory proteins on titer and transduction efficiency.157-159 All presently described FIV vector systems have been derived from FIV-34TF10, a variant of the Petaluma strain,12,103,104 which productively infects feline kidney and neuronal cells but not PBLs or macrophages. Also, the FIV-34TF10 molecular clone is readily available from the NIH AIDS Research and Reference Reagent Program (Cat# 1236). To generate FIV vectors, the plasmid components of the recombinant vector system were transfected into the human 293T cell line.169 This same highly transfectable producer line has been used for the transient production of MLV and lentiviral vectors and allows rapid, high-titer viral particle production. Using human cell lines for FIV vector production reduces possible problems from recombination of FIV vector components with endogenous feline retroviral sequences or potential infectious agents associated with less well-characterized feline producer lines. In addition, production in a human line avoids destruction of viral particles by human serum complement as documented for MLV vectors.170-173 Since clinical trials in 1990,174 experience gathered with MLV-based retroviral vectors suggests that, in addition to overall quality and titer, there is another critical issue in vector production: safety regarding the generation of replication competent virus. The challenge is generating FIV constructs with minimal sequence overlap in an effort to reduce the potential for homologous recombination while maintaining the high-titer capacity. To address these safety issues, FIV vector development followed the general design of state-of-the-art lentiviral

FIV Vectors

109

vector technology, i.e., separating the cis-acting sequences involved in the transfer of the viral genome to target cells from the sequences encoding the trans-acting viral proteins. This “splitgenome” strategy results in separate expression cassettes for the FIV transfer vector, the packaging plasmid and the envelope plasmid. Because of this, currently described FIV vector systems consist of three or four components depending on whether FIV or HIV Rev protein is expressed from its own plasmid (Fig. 3). The resulting recombinant FIV vectors are replication-incompetent and limited to a single round of the infection process in the target cell without spreading.

FIV Transfer Vector For FIV, the total capacity to accommodate transgenes is approximately 8 kb when the sequences coding for all structural and enzymatic proteins are removed. To overcome the FIV LTR’s low transcriptional activity in human cells, a strong ubiquitous promoter such as CMV replaces the U3 region of the 5’ LTR. This hybrid LTR promoter strategy, previously used for MLV and HIV-based vectors,175-178 allows FIV vector production in human cells.157-159 Downstream of the 5’ hybrid promoter, the putative FIV packaging signal is anticipated, yet interestingly no information on its size and exact location was available until the question was raised in the context of FIV vector design. Currently, all described FIV vectors include the 270 nucleotides between the 5’LTR and the gag start codon plus various lengths of gag coding sequence ranging from 250 bp159 to 1,250 bp.157 All FIV vectors with any of these putative packaging signal sequences achieve comparable titers. Another cis-element of the FIV transfer vector is the RNA export system necessary to transport unspliced or singly spliced RNA messages from the nucleus to the cytoplasm. Firstgeneration FIV vectors used the FIV Rev/RRE export system with the RRE located either upstream of the internal expression cassette or downstream at its natural location just before the 5’ terminal PPT (polypurine tract). Heterologous export elements such as the cytoplasmic transport element (CTE) from Mason Pfizer monkey virus, which facilitates the transport of intron-containing mRNAs by a Rev-independent mechanism179,180 or the HIV Rev/RRE export system, were explored in the next generation FIV vectors.158,159 A prototype of the FIV transfer vectors is shown in Figure 3A. Future efforts to improve FIV transfer-vector safety may include deleting even more sequences non-essential to vector production or function and the generation of self-inactivating SIN vectors previously described for MLV and HIV vectors.181-186 FIV SIN vectors would eliminate any remaining transcriptional activity of the wildtype FIV LTR in human target cells, abolishing production of genomic FIV RNA in the target cell. This in turn will decrease possible recombination on the RNA level and reduce the chance of insertional activation of genomic DNA transcription. An added benefit may be increased expression of the transgene since possible interference between the LTR with internal promoters is eliminated. Improving the potency and boost transgene expression of FIV-mediated gene transfer may come from incorporating such additional cis-acting elements as the posttranscriptional regulator from the woodchuck hepatitis virus (WPRE;187,188) and the central DNA flap, which mediates nuclear import of the pre-integration complex. Other retro- and lentiviral systems have previously shown the positive effects of these elements.189-192 Although the FIV genome does not feature a central polypurine tract or termination sequence characteristic for the central DNA flap, several AG-rich regions are present in the FIV pol region, and that region corresponds to the location of the central DNA flap in HIV. Whether FIV has a cis element functionally similar to the central DNA flap in HIV vectors awaits further analysis.

FIV Packaging Plasmid The packaging plasmid supplies all necessary enzymatic and structural core proteins required for the FIV vector particle in trans. A heterologous promoter and poly A site at the 5’

110

Lentiviral Vector Systems for Gene Transfer

Fig. 3. Schematic presentation of the plasmid components for the FIV-based gene delivery system. Represented are the FIV transfer vector (A), the FIV packaging plasmid (B), and a plasmid coding for a heterologous viral envelope (C). The FIV packaging plasmid can be Rev/RRE independent using heterologous export elements such as the CTE, or the packaging functions use an Rev/RRE export element with the Rev protein being driven by a separate plasmid (B). Prom.: promoter.

and 3’ end, respectively, replaces both FIV LTRs and the putative packaging signal. In addition, the plasmid also provides an RNA export element in the form of the original FIV Rev/RRE or the heterologous CTE or HIV Rev/RRE sequences. First-generation FIV vectors retained the FIV-34TF10-derived accessory proteins Vif (functional) and Orf2 with the premature stop codon (nonfunctional), while later generations of FIV vectors158,159describe packaging plasmids without either of the accessory proteins. A prototype of the FIV packaging plasmid with a Rev/RRE export element included in the same packaging plasmid (three plasmid system) or provided on a separate plasmid (four plasmid system) is shown in Figure 3B. Future efforts to improve the safety and expression level of the FIV core proteins may include the use of degenerate code previously described for retroviral vectors. Degenerate codons will not only eliminate sequence homology in the 5’ area of the packaging plasmid with the packaging signal of the FIV vector but may also increase expression levels of the Gag-Pol proteins and generate an mRNA independent of any export element as described for the expression of HIV core proteins.177,193-196

Envelope Plasmid The flexibility to incorporate non-FIV envelope glycoprotein into the FIV viral particle provides an exciting opportunity for cell and tissue targeting using the natural or modified tropism of the heterologous viral envelope. So far, the pantropic VSV-G197 and the 4070 Aderived amphotropic envelopes198,199 have efficiently pseudotyped FIV vectors.157-159,163 A heterologous promoter and poly A signal regulate the expression of the envelope of choice. A prototype of the envelope plasmid is shown in Figure 3C, and it is not surprising that it does not differ in general design from envelope cassettes described for other lentiviral vectors. Compared to VSV-G pseudotyped FIV vectors, pseudotyping with the amphotropic envelope results in an approximately five- to tenfold lower titer in unconcentrated supernatant. This is also true for MLV- or HIV-based vectors (unpublished data). The higher titers observed for VSV-G pseudotyped vectors may be a result of increased transduction efficiencies rather

FIV Vectors

111

Fig. 4. In vivo transduction by a VSV-G pseudotyped FIV vector of (A) hamster muscle;158 (B) Purkinje cells in the murine cerebellar lobule (2x105 cfu, day 21), courtesy of Dr. Beverly Davidson; and (C) rabbit airway epithelium transduced from the apical side.214 Various cell types of the airway epithelium including ciliated and basal cells are transduced (indicated by arrows).

than higher recombinant viral particle output. Possible explanations for increased transduction include the higher prevalence of VSV-G receptors compared to amphotropic receptors on target cells, different mechanisms for viral particle uptake and/or increased half-life of VSV-G pseudotyped particles in vitro. Incorporating VSV-G into the recombinant viral particle results in increased resistance to mechanical stresses—ultracentrifugation, for example.200 This property is not observed in amphotropic vectors and may contribute to a reduced half-life in culture media and increased losses of infectivity during vector concentration. Like many animal viruses that bind to specific receptors on the plasma membrane of cells, the amphotropic envelope receptor Pit-2 is a phosphate transporter that interacts with the amphotropic envelope in a very specific manner.198,199,201,202 In contrast, the receptor for VSV-G is a rather nonspecific, ubiquitous phospholipid molecule203-205 present in all cell membranes, which results in such a broad host range that it efficiently transduces almost all species of vertebrate and insect cells.200,206,207 Furthermore, viruses with VSV-G envelopes enter the target cells through general endocytosis via clathrin-coated pits and pH-induced endosomal fusion208 while viruses with amphotropic envelopes fuse directly at the cell membrane after binding to the receptor. The exact mechanism(s) of reduced titer of amphotropic compared to VSV-G tropic vectors remains to be elucidated. The general pseudotyping capability of the FIV particle is not yet explored in great detail, but the high titer VSV-G and amphotropic FIV vectors hold promise for other envelopes. Interestingly, the Gibbon Ape Leukemia Virus (GALV) envelope efficiently pseudotypes MLV but not HIV vectors, and our observations show that the GALV envelope does not pseudotype FIV vectors either (unpublished data). However, a recent report describes the successful pseudotyping of HIV vectors with the GALV envelope after exchange of the cytoplasmic tail of GALV with that of the MLV envelope in an effort to mimic the amphotropic envelope.209 Although HIV vector titers with the GALV hybrid envelope are low, and the most commonly used envelope for lentiviral vectors is VSV-G, alternative envelopes and envelope hybrids widen the opportunity to transduce specific cell types and enhance the FIV vector’s utility for a range of applications where more restricted tropism is beneficial. Despite the higher titer of VSV-G pseudotyped FIV vectors, we propose that the amphotropic envelope may have advantages over VSV-G, in particular for developing clinicalgrade FIV vectors. In addition to the excellent safety profile of the amphotropic envelope

112

Lentiviral Vector Systems for Gene Transfer

demonstrated by the safe administration of amphotropic MLV vectors to over 1,000 patients210 further benefits are i) lack of toxicity caused by the fusogenic property of VSV-G; ii) resistance to human serum inactivation; iii) the possibility of generating stable packaging and producer lines without the need for regulated vector production and; iv) reduced pseudotransduction.211,212

Analysis of FIV Vectors In Vitro VSV-G pseudotyped vectors have been useful, though. First proof-of-concept studies, as well as further optimization of an FIV-based gene delivery system, used vectors pseudotyped with VSV-G. This allows efficient transduction of a wide range of target cells and concentration via simple centrifugation.200 Recombinant VSV-G pseudotyped FIV vectors coding for marker genes (β-gal, GFP) were used to analyze the requirements for various cis elements and accessory proteins in the FIV vector and packaging plasmids. The following paragraphs review the various aspects of FIV vector technology investigated and published so far in more detail.

FIV LTR Activity in Human Cells Analyzing FIV LTR activity in human and feline cells established that the sole restriction to FIV replication in human cells is the U3 region of the FIV LTR.157 The 50- to 200-fold drop in titer observed from recombinant FIV vectors driven by the 5’FIV LTR compared to a hybrid LTR with the CMV promoter/enhancer158,159 supports this data. The U3 element of the 5’ FIV LTR was replaced by fusing the heterologous CMV promoter/enhancer region to the R repeat in such a way that the natural spacing between the CMV-derived TATA box and the FIV mRNA cap site is preserved.157-159 Replacing the U3 region with the CMV enhancer only gave intermediate titers, indicating that the CMV promoter provides additional cis elements for increased transcription in human cells.159 Such hybrid LTR promoters have been described before for other retroviral vectors175-177,213 and have made it possible to generate high titer FIV vectors in human cells.

FIV Vector Titer Recombinant FIV vectors are generated via transient transfection in the highly transfectable human epithelial kidney cell line 293T169 and titers of approximately 1x106 cfu/ml in unconcentrated supernatants were achieved.158,159 Major factors influencing the titer of FIV vectors include the choice of the production cell line, the design and expression level of the structural and enzymatic helper genes and the levels of vector RNA. Another critical factor is the molar ratio of the vector, packaging and envelope components since this ratio strongly influences the percentage of infectious versus noninfectious viral particles. Relative transduction efficiencies of FIV vectors using a panel of human, feline and rodent target cells varied up to two logs.157,159 suggesting that the choice of cell line used for titration (and titering method) are additional factors for consideration. We show a comparison of commonly used titering cell lines and clearly demonstrate the influence of the target cell on FIV vector titer (Table 1). Although the feline CrFK cells consistently gave the highest titer results, we generally use a human titering line since a human target cell seems most relevant for gene therapy applications. High titer FIV vector production is very robust with a variety of transgenes including those encoding β-galactosidase, GFP, β-glucuronidase, CFTR, hGH and erythropoietin achieving titers of about 1x106 cfu/ml unconcentrated supernatant (unpublished data). To facilitate the FIV titer assay for those transgenes without convenient read-out, a real-time PCR titering method was developed.214 This PCR-based FIV titer protocol can be applied to any given FIV vector preparation and to determine the transducing units per ml.

FIV Vectors

113

Transduction of Nonproliferating Cells FIV vectors transduce proliferating and nonproliferating cells arrested in the G0/G1, G1/ S or G2/M phase at similar efficiencies: a variety of established cell lines as well as primary human cells have been tested.157-159,214 The test panel of established lines includes cells of human origin such as fibrosarcoma (HT-1080), rhabdomyosarcoma (RD), carcinoma (HeLa), osteosarcoma (HOS), kidney (293), lung fibroblast (WI-38); cells of feline origin such as tongue epithelium (Fc3Tg) or kidney epithelium (CrFK); and rodent cells such as rat and murine fibroblasts. Human primary cells transduced with FIV vectors include monocyte-derived macrophages, postmitotic neurons, skin fibroblasts, aortic smooth muscle cells, hepatocytes, airway epithelia and dendritic cells.157-159,214 The dividing established cell lines were growth arrested either through γ-irradiation (G2/M), aphidicolin (G1/S) or taxol (G2/M). The dividing primary skin fibroblasts were arrested in the G0/G1 phase by contact inhibition, and the other primary cells divide very slowly or not at all in culture. So far, there is no report of a cell type refractory to FIV vector transduction or a cell type that displays a large discrepancy in transduction efficiency of nonproliferating versus proliferating cells of 5-fold or higher. To monitor the FIV vectors’ overall transducing capability relative to other well-characterized retroviral vectors, MLV-based vectors at the same titers were included in some experiments. Compared to MLV vector controls no difference in transduction efficiency of proliferating target cells was observed, indicating that FIV vectors are equally able to confer transgene expression to proliferating cells. In addition, MLV vectors served as a control for growth arrest due to their virtually abolished capability to transduce nonproliferating cells.

Accessory Proteins The simple FIV genome, with only two accessory proteins (Vif, Orf2) and one regulatory (Rev) protein, facilitates analyzing their importance for the FIV vector system. To study possible effects of vif and orf2 coding regions and/or their proteins, packaging plasmids were generated containing either vif or orf2 alone, both genes or none. Since FIV-34TF10 is naturally dysfunctional in Orf2 expression,104 its orf2 coding region was substituted with that of the fully functional orf2 of the molecular FIV strain FIV14.158 Resulting FIV vectors were used to analyze the titer and transduction efficiency of various proliferating and nonproliferating cells. Transduction efficiencies of vectors prepared without the FIV vif and orf2 genes did not differ substantially from those of vectors prepared with accessory gene expression in either dividing or nondividing cells (ref. 158 and Table 2). This finding is comparable to the situation in HIV vectors where efficient transduction does not require any of the accessory proteins except for one reported case of diminished transduction of liver after in vivo injection with recombinant HIV-1 vectors lacking the accessory Vpr and Vpu proteins.215 Interestingly, studies also report efficient transduction of primary human hepatocytes by FIV vectors irrespective of the presence or absence of sequences coding for FIV accessory proteins.159 Based on the function of FIV Vif and the fact that HIV Vif(-) vectors do not show diminished titer production or transduction efficiency in vivo,216 we anticipated that there may not be a need for FIV accessory proteins for efficient FIV vector performance either. The transactivating activity of Orf2 becomes superfluous in the context of FIV vectors with a hybrid LTR, suggesting that retaining Orf2 in FIV vectors may not have an advantage over FIV Orf2 vectors. Altogether it may not be surprising that FIV vectors deleted in accessory proteins show undiminished titer and transduction efficiency although it cannot be ruled out that accessory proteins or their sequences may play a role in the transducibility of certain species or tissues that have not yet been examined.

114

Lentiviral Vector Systems for Gene Transfer

Table 1. Effect of titering cell line on FIV vector titer Titering Cell Line

FIV/βgal, VSV-G FIV/βgal, Ampho Mean Titera (CFU/ml of virus stock) ± SD

HeLa

1.9 (± 0.20) x 107

9.8 (± 1.44) x 105

HT-1080

1.6 (± 0.68) x 108

1.6 (± 0.30) x 107

293T

2.3 (± 0.66) x 108

2.2 (± 0.08) x 107

CrFK

1.3 (± 0.20) x 109

1.3 (± 0.23) x 108

3T3

2.6 (± 0.27) x 108

5.8 (± 1.05) x 107

FIV vectors coding for β-galactosidase were pseudotyped with the envelope from Vesicular Stomatitis Virus (VSV-G) or the amphotropic envelope from 4070A (Ampho). Vectors were concentrated and virus stocks titered on a variety of cell lines. a Each titer value represents the average of duplicate titers determined from 2-8 titer values scored at different dilutions each. Titer is expressed as LacZ colonyforming units per milliliter of virus stock.

Export Elements To understand the requirement for the FIV vector system regarding RNA export elements, FIV’s natural RNA export system (FIV Rev/RRE), as well as substitutions with heterologous export elements were analyzed in the context of the FIV vector and packaging plasmid.158,159 The natural location of the FIV RRE at the 3’ end of the genome facilitated the design of FIV vectors and packaging plasmids since the RRE site could be retained in its original position as shown in early generation vectors.157-159 An alternative location for the FIV RRE in the vector construct upstream of the internal expression cassette of the transgene resulted in slightly increased titers.158 Heterologous export systems such as the Rev/RRE system from HIV-1 achieved 1.5- to 5-fold increased titers,158,159 whereas the CTE export element gave titers comparable to the FIV Rev/RRE system as long as the CTE was located less than 250 bp upstream of the 3’FIV LTR.159 In the absence of any export system in the FIV vector and packaging construct, the titer dropped by as much as five logs. Lack of the FIV RRE in the FIV vector alone while maintaining the FIV Rev/RRE system in the packaging construct reduced the titer by approximately 1-1.5 logs.158 In general, both the FIV vector and packaging plasmid require an export element for high titer vector production, although the nature and location of the export element is flexible and certain heterologous export systems may help to increase titers relative to FIV’s own export system. Future modifications of the packaging plasmid may include the use of degenerate codons to increase expression levels and gain independence from any export element as described for HIV Gag-Pol.194

Safety Modifications The major safety concern for lentiviral vectors is the emergence of replication-competent virus, which usually arises from homologous and to a lesser extent nonhomologous recombination.217,218 In an effort to reduce the substrate for homologous recombination, the sequence homology between FIV vector components is minimized. Corresponding safety modifications of the FIV vector system include: i) replacement of the FIV LTRs with heterologous promoters

FIV Vectors

115

Table 2. Effect of FIV accessory gene expression on transduction of human primary skin fibroblasts Mean Transduction Efficiencya (CFU/ml of virus stock) ± SD Dividing Nondividing

Packaging Construct

Vector Construct

pCFIV

pVETLCβ

3.2 (± 0.36) x 104 [100%]

1.6 (± 0.27) x 104 [100%]c

pCFIV∆orf2

pVETLCβ

3.3 (± 0.47) x 104 [104%]

1.7 (± 0.36) x 104 [106%]

pCFIV∆vif

pVETLCβ

3.3 (± 0.11) x 104 [104%]

1.7 (± 0.37) x 104 [106%]

pCFIV∆orf2∆vif

pVETLCβ

2.9 (± 0.30) x 104 [ 93%]

1.7 (± 0.16) x 104 [108%]

pMLVgagpol

pMLVCMVβ

3.8 (± 0.26) x 104 [121%]

7.3 (± 0.90) x 101 [0.5%]

a Results are from a representative experiment, with each value representing the average transduction

efficiency of three replicate vector preparations done in triplicate. Transduction efficiency is expressed as LacZ-forming units per milliliter of virus stock. Virus stock was used to transduce dividing and nondividing human skin fibroblasts at an MOI of 2.0. The transduction efficiency of vector stocks prepared from pCFIV and pVETLCβ was arbitrarily given a value of 100%, to which the transduction efficiencies of the other vector stocks were compared. FIV packaging constructs code either for Vif and Orf2 (pCFIV), Vif only (pCFIV∆orf2), Orf2 only (pCFIV∆vif) or neither accessory protein (pCFIV∆orf2∆vif). The FIV and MLV vectors pVETLCβ and pMLVCMVβ, respectively, code for β-galactosidase.

and poly A signal in the packaging plasmid; ii) separation of viral structural and enzymatic genes into at least two expression cassettes, gag/pol and a heterologous env; iii) reduction of sequence homology between the individual retroviral components; and iv) the use of human producer cells instead of feline cells. All presently described FIV vectors were generated in human cells and follow the design of the split-genome approach with minimal sequence overlap.157-159 To further separate the FIV components, expression of the FIV Rev protein was driven by its own expression plasmid, resulting in a four-plasmid transfection system for FIV vector production (Fig. 3).158,159 Alternatively, heterologous export elements such as the CTE eliminated the need for FIV Rev and RRE sequences altogether;159 however, one area of sequence overlap currently remains between the 3’ area of the putative packaging signal in the FIV vector and the 5’ part of the packaging plasmid. Following these general guidelines, minimal FIV packaging constructs were designed and the production of accessory proteins dispensable for vector production and infectivity was eliminated. A minimal packaging construct with only 6 bp of noncoding sequence upstream of the major splice donor (MSD) site while deleting the 17 bp normally located between the MSD and the gag start codon is as functional as a packaging plasmid with additional 100 bp upstream of the MSD and the 17 original bp between MSD and gag start remaining. Downstream of the FIV gag/pol coding region, deletions in the envelope and accessory protein sequence of various sizes were introduced to disable the generation of the Env and/or accessory

116

Lentiviral Vector Systems for Gene Transfer

proteins. The most advanced packaging constructs contain heterologous HIV Rev/RRE or CTE export elements.158,159 FIV vector constructs are generally deleted in all FIV structural and enzymatic genes. However, due to its dual function as a part of the putative packaging signal as well as the 5’ coding part of gag, partial sequence of the gag-coding region is retained. The shortest packaging signal reported in an FIV vector without titer loss includes the 270 bp between the 5’ LTR and the gag codon plus the first 350 bp of the gag coding region.158 These first definitions of the putative packaging signal await further, more detailed analysis of the minimal requirements for maximal packaging. Additional safety features include the introduction of a stop codon around 300 bp downstream of gag start to avoid production of a Gag-Pol multiprotein in case recombination between the FIV vector and packaging plasmid occurs.157 The next generation vectors eliminated most of the remaining envelope and FIV Rev/RRE sequences present in the first generation FIV vectors.157-159 Table 3 summarizes in more detail the molecular structure of the currently published FIV vector systems, each describing the most advanced versions of the FIV gene delivery system. Extensive testing for replication-competent virus generated during FIV vector production has not yet been described, but our efforts to monitor FIV p27 (capsid) expression levels in human and feline target cells exposed to high titer FIV vectors and passaged for a period of six weeks did not result in any p27 expression levels above background (unpublished data). Analogous to the RCR testing of primate lentiviral vectors, additional RCR assays for FIV vectors including quantitative PCR need to be developed to assure this nonprimate lentiviral gene delivery system’s safety.

FIV Vector Mobilization A concern when using HIV-based vectors in human subjects has been the potential for HIV vector mobilization by wild-type HIV. To address the question of whether HIV can mobilize FIV vectors, assays to test cross-packaging using FIV-, MLV- and HIV-based vector components were carried out: the conclusion was that HIV cannot efficiently mobilize FIV vectors.159 If these preliminary studies can be confirmed, FIV vectors have a potential benefit in particular for those patients with established HIV infections.

Envelope-Specific Serum Inactivation MLV-based vectors produced in certain human cells are resistant to inactivation by human complement,170,171,173,219,220 whereas vectors produced in a nonhuman (canine) cell were rapidly susceptible to inactivation both in vitro and in vivo.171 Besides the producer cell, factors such as the envelope contribute in determining complement sensitivity also.170,221 With that in mind, the relative complement sensitivity of MLV, HIV and FIV vectors to human serum was tested for vectors produced in human cells and pseudotyped with either VSV-G or amphotropic envelopes. In contrast to their VSV-G counterparts, all three amphotropic pseudotyped vectors survived when incubated with human serum (survival of all three VSV-G pseudotyped vectors was less than 3%).163 Sensitivity of these vectors to human serum seems to result from an assault by VSV-G specific antibodies directly or via complement activation. These data suggest that FIV vectors with an amphotropic envelope may have substantial advantages compared to VSV-G pseudotyped vectors for certain human gene transfer applications, particularly those requiring intravenous administration.

In Vivo Gene Delivery by FIV Vectors The promise of FIV vectors for direct gene delivery was demonstrated by efficient transduction of hamster muscle,158 mouse and primate brain160-162 and rabbit airway epithelium.214 Extensive transgene expression and duration in a variety of tissues and species is seen in the

FIV Vectors

117

Table 3. Comparison of molecular components of published FIV vector systemsa FIV vector components

FIV transfer vector Remaining gag-coding sequence in packaging signal (ψ)

gag-coding sequences of ψ interrupted by stop codon Heterologous export element FIV packaging plasmid Remaining FIV sequence upstream of gag/pol start (bp) Sequence between MSD and start of gag eliminated Heterologous export elements Rev coded on a separate expression plasmid Sequence coding for accessory proteins deleted

Poeschla et al, 1998157

Johnston et al, 1999 158

Curran et al, 2000159

1,250 bp

350 bp

830b bp

+

-

-

-

+

+

120 bp -

10 bp +

120 bp -

-

+ +

+ +

-

+

+

a Compared were the most advanced vector components in the respective publications. b Reported length of the 3rd generation transfer vectors. However, earlier generation transfer vectors

with 250 bp of gag-coding sequences are mentioned. MSD = major splice donor

absence of inflammatory responses.214 Particularly promising for a gene therapy approach to cystic fibrosis is the stable expression of a transgene in 5-10% of airway epithelium transduced from the apical side,214 since that level of transduction is in a range considered to be therapeutic.222 Detailed analysis of the cell types transduced by FIV vectors following injection into mouse cerebrum revealed that Purkinje cells among others were successfully transduced.160 This suggests that FIV vectors may be applied for diseases affecting the cerebellum such as spinocerebellar ataxias (SCA). Stable expression of β-glucuronidase in a β-gluc-deficient mouse model after FIV/β-gluc vector treatment resolved characteristic disease symptoms of mucopolysaccharidosis VII.161 These data further validate FIV vectors for gene therapy applications of the mammalian central nervous system. Efficient gene transfer to the retina and liver was observed as well (personal communication, Drs. Beverly Davidson and Paul McCray). Figure 4 shows some examples of transduced organs and cells after direct injection of FIV vectors. For in vivo studies, transiently generated FIV vectors were concentrated to achieve titers ranging from 1x107-1x109 cfu/ml using a variety of methods including ultrafiltration, centrifugation, ion exchange chromatography and PEG-precipitation. Table 4 shows an example for FIV vector concentration using PEG precipitation followed by centrifugation. Final titer and general quality of the preparation is an important factor ultimately influencing the in vivo efficacy of retroviral/lentiviral vectors. Vector preparations can vary greatly in their quality, especially concerning the ratio of infectious versus noninfectious vector as well as protein and other contaminants, for example FBS. A detailed description on the production of high quality MLV-based vector is noted elsewhere,223 however, these observations most likely

Lentiviral Vector Systems for Gene Transfer

118

Table 4. Yields and losses during concentration of FIV vectors FIV/epo vector material

Amount (ml)

Titer (TU/ml)

Concentration (x-fold) recovery

% step recovery

Crude supernatant

5,000

3.1 x 105

Clarified

5,000

6.8 x 105

1

221

PEG concentrate

200

1.1 x 107

25

62.9

Centrifuged, final

3

2.0 x 108

1667

27.9

% overall

38.8

Representative concentration procedure for FIV vectors using PEG precipitation followed by pelleting. Unconcentrated viral supernatant is clarified using a filtration step (0.45 um) followed by polyethylene glycol precipitation (10% PEG final) for at least 4 hours at 4oC followed by pelleting and resuspension in a smaller volume. FIV vectors were then further concentrated to a smaller volume by a final centrifugation step. FIV vectors were pseudotyped with VSV-G, coded for erythropoietin and titers were determined by quantitative PCR.

apply to FIV vectors as well. The development of stable FIV packaging and producer cell lines is expected to improve the quality and reproducibility of FIV vector preparations. Extensive screening of producer lines for high titer and general scalable growth characteristics usually identifies a candidate with a high ratio of infectious to noninfectious particles. High-titer producer cell lines will facilitate a commercially feasible manufacturing process that will in turn deliver good quality vector at the clinically relevant titers required for therapeutic responses. Will lentiviral vector systems derived from various species perform in a similar fashion in primates? Some lentiviruses naturally favor certain tissues and cell types, a feature often defined by the envelope component. However, as in the case of FIV-34TF10 where deletion of functional orf2 results in modified cell tropism, it is conceivable that differences in the activity of ciselements and non-envelope proteins such as the integrase or reverse transcriptase modulate the overall efficiency of a given lentiviral vector in primates. Comparing different lentiviral systems is difficult since inequalities in vector material flaws most comparisons. To assess relative efficiencies of lentiviral vectors from different species, all components going into the final vector preparation must be generated and treated in the same manner, starting with the quality of the plasmids, the nature of the vector components (i.e. export elements and promoters), vector production, purification and titration. Apart from using FIV vectors for human gene therapy, the technology is beneficial for anyone who needs to efficiently introduce transgenes into any type target cell for further study, or to generate stable cell lines. FIV vector production is both simple and fast, and researchers may use FIV vectors to study basic FIV virology. In summary, primary results from in vitro and in vivo studies of FIV vectors indicate that this nonprimate gene delivery system is highly efficient and holds promise for many human gene therapy applications.

Acknowledgements This work was supported in part by the National Institute of Allergy and Infectious Diseases grant AI45992.

FIV Vectors

119

References 1. Pedersen NC, Ho EW, Brown ML, Yamamoto JK. Isolation of a T-lymphotropic virus from domestic cats with an immunodeficiency-like syndrome. Science 1987; 235:790-793. 2. Malik R, Kendall K, Cridland J, Coulston S, Stuart AJ, Snow D et al. Prevalences of feline leukaemia virus and feline immunodeficiency virus infections in cats in Sydney. Aust Vet J 1997; 75:323-327. 3. Ishida T, Washizu T, Toriyabe K, Motoyoshi S, Tomoda I, Pedersen NC. Feline immunodeficiency virus infection in cats of Japan. J Am Vet Med Assoc 1989; 194:221-225. 4. Carpenter MA, Brown EW, MacDonald DW, O’Brien SJ. Phylogeographic patterns of feline immunodeficiency virus genetic diversity in the domestic cat. Virology 1998; 251:234-243. 5. Jordan HL, Howard J, Sellon RK, Wildt DE, Tompkins WA, Kennedy-Stoskopf S. Transmission of feline immunodeficiency virus in domestic cats via artificial insemination. J Virol 1996; 70:8224-8228. 6. O’Neil LL, Burkhard MJ, Hoover EA. Frequent perinatal transmission of feline immunodeficiency virus by chronically infected cats. J Virol 1996; 70:2894-2901. 7. Sodora DL, Shpaer EG, Kitchell BE, Dow SW, Hoover EA, Mullins JI. Identification of three feline immunodeficiency virus (FIV) env gene subtypes and comparison of the FIV and human immunodeficiency virus type 1 evolutionary patterns. J Virol 1994; 68:2230-2238. 8. Kakinuma S, Motokawa K, Hohdatsu T, Yamamoto JK, Koyama H, Hashimoto H. Nucleotide sequence of feline immunodeficiency virus: classification of Japanese isolates into two subtypes which are distinct from non-Japanese subtypes. J Virol 1995; 69:3639-3646. 9. Pecoraro MR, Tomonaga K, Miyazawa T, Kawaguchi Y, Sugita S, Tohya Y et al. Genetic diversity of Argentine isolates of feline immunodeficiency virus. J Gen Virol 1996; 77:2031-2035. 10. Bachmann MH, Mathiason-Dubard C, Learn GH, Rodrigo AG, Sodora DL, Mazzetti P et al. Genetic diversity of feline immunodeficiency virus: Dual infection, recombination, and distinct evolutionary rates among envelope sequence clades. J Virol 1997; 71:4241-4253. 11. Nishimura Y, Goto Y, Pang H, Endo Y, Mizuno T, Momoi Y et al. Genetic heterogeneity of env gene of feline immunodeficiency virus obtained from multiple districts in Japan. Virus Res 1998; 57:101-112. 12. Olmsted RA, Langley R, Roelke ME, Goeken RM, Adger-Johnson D, Goff JP et al. Worldwide prevalence of lentivirus infection in wild feline species: Epidemiologic and phylogenetic aspects. J Virol 1992; 66:6008-6018. 13. Barr MC, Calle PP, Roelke ME, Scott FW. Feline immunodeficiency virus infection in nondomestic felids. J Zoo Wildl Med 1989; 20:265-272. 14. VandeWoude S, O’Brien SJ, Hoover EA. Infectivity of lion and puma lentiviruses for domestic cats. J Gen Virol 1997; 78:795-800. 15. Zou L, Barr MC, Hoose WA, Avery RJ. Characterization of the transcription map and Rev activity of a highly cytopathic feline immunodeficiency virus. Virology. 1997; 236:266-278. 16. Hartmann K. Feline immunodeficiency virus infection: an overview. Vet J 1998; 155:123-137. 17. Bendinelli M, Pistello M, Lombardi S, Poli A, Garzelli C, Matteucci D et al. Feline immunodeficiency virus: An interesting model for AIDS studies and an important cat pathogen. Clin Microbiol Rev 1995; 8:87-112. 18. Matteucci D, Baldinotti F, Mazzetti P, Pistello M, Bandecchi P, Ghilarducci R et al. Detection of feline immunodeficiency virus in saliva and plasma by cultivation and polymerase chain reaction. J Clin Microbiol 1993; 31:494-501. 19. Goto Y, Nishimura Y, Mizuno T, Endo Y, Baba K, Momoi Y et al. Quantification of viral ribonucleic acid in plasma of cats naturally infected with feline immunodeficiency virus.[In Process Citation]. Am J Vet Res 2000; 61:1609-1613. 20. Brown WC, Bissey L, Logan KS, Pedersen NC, Elder JH, Collisson EW. Feline immunodeficiency virus infects both CD4+ and CD8+ T lymphocytes. J Virol 1991; 65:3359-3364. 21. English RV, Johnson CM, Gebhard DH, Tompkins MB. In vivo lymphocyte tropism of feline immunodeficiency virus. J Virol. 1993; 67:5175-5186. 22. Dow SW, Mathiason CK, Hoover EA. In vivo monocyte tropism of pathogenic feline immunodeficiency viruses. J Virol 1999; 73:6852-6861.

120

Lentiviral Vector Systems for Gene Transfer

23. Brunner D, Pedersen NC. Infection of peritoneal macrophages in vitro and in vivo with feline immunodeficiency virus. J Virol 1989; 63:5483-5488. 24. Dow SW, Poss ML, Hoover EA. Feline immunodeficiency virus: a neurotropic lentivirus. J Acquir Immune Defic Syndr 1990; 3:658-668. 25. Phillips TR, Talbott RL, Lamont C, Muir S, Lovelace K, Elder JH. Comparison of two host cell range variants of feline immunodeficiency virus. J Virol 1990; 64:4605-4613. 26. Vahlenkamp TW, Verschoor EJ, Schuurman NN, van Vliet AL, Horzinek MC, Egberink HF et al. A single amino acid substitution in the transmembrane envelope glycoprotein of feline immunodeficiency virus alters cellular tropism. J Virol 1997; 71:7132-7135. 27. Verschoor EJ, Boven LA, Blaak H, van Vliet AL, Horzinek MC, de Ronde A. A single mutation within the V3 envelope neutralization domain of feline immunodeficiency virus determines its tropism for CRFK cells. J Virol 1995; 69:4752-4757. 28. Vahlenkamp TW, de Ronde A, Schuurman NN, van Vliet AL, van Drunen J, Horzinek MC et al. Envelope gene sequences encoding variable regions 3 and 4 are involved in macrophage tropism of feline immunodeficiency virus. J Gen Virol. 1999; 80:2639-2646. 29. Lerner DL, Elder JH. Expanded host cell tropism and cytopathic properties of feline immunodeficiency virus strain PPR subsequent to passage through interleukin-2-independent T cells. J Virol 2000; 74:1854-1863. 30. Waters AK, De Parseval AP, Lerner DL, Neil JC, Thompson FJ, Elder JH. Influence of ORF2 on host cell tropism of feline immunodeficiency virus. Virology 1996; 215:10-16. 31. Lockridge KM, Himathongkham S, Sawai ET, Chienand M, Sparger EE. The feline immunodeficiency virus vif gene is required for productive infection of feline peripheral blood mononuclear cells and monocyte-derived macrophages. Virology 1999; 261:25-30. 32. Willett BJ, Hosie MJ, Dunsford TH, Neil JC, Jarrett O. Productive infection of T-helper lymphocytes with feline immunodeficiency virus is accompanied by reduced expression of CD4. Aids 1991; 5:1469-1475. 33. Willett BJ, Picard L, Hosie MJ, Turner JD, Adema K, Clapham PR. Shared usage of the chemokine receptor CXCR4 by the feline and human immunodeficiency viruses. J Virol 1997; 71:6407-6415. 34. Hosie MJ, Broere N, Hesselgesser J, Turner JD, Hoxie JA, Neil JC et al. Modulation of feline immunodeficiency virus infection by stromal cell-derived factor. J Virol 1998; 72:2097-2104. 35. Richardson J, Pancino G, Merat R, Leste-Lasserre T, Moraillon A, Schneider-Mergener J et al. Shared usage of the chemokine receptor CXCR4 by primary and laboratory-adapted strains of feline immunodeficiency virus. J Virol 1999; 73:3661-3671. 36. Egberink HF, De Clercq E, Van Vliet AL, Balzarini J, Bridger GJ, Henson G et al. Bicyclams, selective antagonists of the human chemokine receptor CXCR4, potently inhibit feline immunodeficiency virus replication. J Virol 1999; 73:6346-6352. 37. Poeschla EM, Looney DJ. CXCR4 is required by a nonprimate lentivirus: Heterologous expression of feline immunodeficiency virus in human, rodent, and feline cells. J Virol 1998; 72:6858-6866. 38. Willett BJ, Adema K, Heveker N, Brelot A, Picard L, Alizon M et al. The second extracellular loop of CXCR4 determines its function as a receptor for feline immunodeficiency virus [published erratum appears in J Virol 1998 Oct; 72(10):8460]. J Virol 1998;72:6475-6481. 39. Endo Y, Goto Y, Nishimura Y, Mizuno T, Watari T, Hasegawa A et al. Inhibitory effect of stromal cell derived factor-1 on the replication of divergent strains of feline immunodeficiency virus in a feline T-lymphoid cell line. Vet Immunol Immunopathol 2000; 74:303-314. 40. Bishop SA, Williams NA, Gruffydd-Jones TJ, Harbour DA, Stokes CR. Impaired T-cell priming and proliferation in cats infected with feline immunodeficiency virus. Aids 1992; 6:287-293. 41. Tochikura TS, Hayes KA, Cheney CM, Tanabe-Tochikura A, Rojko JL, Mathes LE et al. In vitro replication and cytopathogenicity of the feline immunodeficiency virus for feline T4 thymic lymphoma 3201 cells. Virology 1990; 179:492-497. 42. Bishop SA, Gruffydd-Jones TJ, Harbour DA, Stokes CR. Programmed cell death (apoptosis) as a mechanism of cell death in peripheral blood mononuclear cells from cats infected with feline immunodeficiency virus (FIV). Clin Exp Immunol 1993; 93:65-71. 43. Johnson CM, Benson NA, Papadi GP. Apoptosis and CD4+ lymphocyte depletion following feline immunodeficiency virus infection of a T-lymphocyte cell line. Vet Pathol 1996; 33:195-203.

FIV Vectors

121

44. Woo JC, Dean GA, Pedersen NC, Moore PF. Immunopathologic changes in the thymus during the acute stage of experimentally induced feline immunodeficiency virus infection in juvenile cats. J Virol 1997; 71:8632-8641. 45. Dean GA, Pedersen NC. Cytokine response in multiple lymphoid tissues during the primary phase of feline immunodeficiency virus infection. J Virol 1998; 72:9436-9440 46. Levy JK, Ritchey JW, Rottman JB, Davidson MG, Liang YH, Jordan HL et al. Elevated interleukin10-to-interleukin-12 ratio in feline immunodeficiency virus-infected cats predicts loss of type 1 immunity to Toxoplasma gondii. J Infect Dis 1998; 178:503-511. 47. Clerici M, Shearer GM. A TH1—>TH2 switch is a critical step in the etiology of HIV infection. Immunol Today 1993; 14:107-111. 48. Krebs FC, Ross H, McAllister J, Wigdahl B. HIV-1-associated central nervous system dysfunction. Adv Pharmacol 2000; 49:315-385. 49. Poli A, Abramo F, Di Iorio C, Cantile C, Carli MA, Pollera C et al. Neuropathology in cats experimentally infected with feline immunodeficiency virus: A morphological, immunocytochemical and morphometric study. J Neurovirol 1997; 3:361-368. 50. Podell M, Maruyama K, Smith M, Hayes KA, Buck WR, Ruehlmann DS et al. Frontal lobe neuronal injury correlates to altered function in FIV-infected cats. J Acquir Immune Defic Syndr 1999; 22:10-18. 51. Boche D, Hurtrel M, Gray F, Claessens-Maire MA, Ganiere JP, Montagnier L et al. Virus load and neuropathology in the FIV model. J Neurovirol 1996; 2:377-387. 52. Podell M, Oglesbee M, Mathes L, Krakowka S, Olmstead R, Lafrado L. AIDS-associated encephalopathy with experimental feline immunodeficiency virus infection. J Acquir Immune Defic Syndr 1993; 6:758-771. 53. Prospero-Garcia O, Herold N, Waters AK, Phillips TR, Elder JH, Henriksen SJ. Intraventricular administration of a FIV-envelope protein induces sleep architecture changes in rats. Brain Res 1994; 659:254-258. 54. Phillips TR, Prospero-Garcia O, Wheeler DW, Wagaman PC, Lerner DL, Fox HS et al. Neurologic dysfunctions caused by a molecular clone of feline immunodeficiency virus, FIV-PPR. J Neurovirol 1996; 2:388-396. 55. Podell M, Hayes K, Oglesbee M, Mathes L. Progressive encephalopathy associated with CD4/CD8 inversion in adult FIV-infected cats. J Acquir Immune Defic Syndr Hum Retrovirol 1997; 15:332-340. 56. Power C, Moench T, Peeling J, Kong PA, Langelier T. Feline immunodeficiency virus causes increased glutamate levels and neuronal loss in brain. Neuroscience 1997; 77:1175-1185. 57. Podell M, Chen E, Shelton GD. Feline immunodeficiency virus associated myopathy in the adult cat. Muscle Nerve 1998; 21:1680-1685. 58. Hein A, Martin JP, Koehren F, Bingen A, Dorries R. In vivo infection of ramified microglia from adult cat central nervous system by feline immunodeficiency virus. Virology. 2000; 268:420-429. 59. Yu N, Billaud JN, Phillips TR. Effects of feline immunodeficiency virus on astrocyte glutamate uptake: Implications for lentivirus-induced central nervous system diseases. Proc Natl Acad Sci USA 1998; 95:2624-2629. 60. Billaud JN, Selway D, Yu N, Phillips TR. Replication rate of feline immunodeficiency virus in astrocytes is envelope dependent: Implications for glutamate uptake. Virology 2000;266:180-188. 61. Boche D, Gray F, Khatissian E, Hurtrel M, Montagnier L, Hurtrel B. Contribution of animal models in the understanding of AIDS encephalopathy. Arch Anat Cytol Pathol 1997; 45:75-85. 62. Poli A, Pistello M, Carli MA, Abramo F, Mancuso G, Nicoletti E et al. Tumor necrosis factoralpha and virus expression in the central nervous system of cats infected with feline immunodeficiency virus. J Neurovirol 1999; 5:465-473. 63. Kohmoto M, Uetsuka K, Ikeda Y, Inoshima Y, Shimojima M, Sato E et al. Eight-year observation and comparative study of specific pathogen-free cats experimentally infected with feline immunodeficiency virus (FIV) subtypes A and B: Terminal acquired immunodeficiency syndrome in a cat infected with FIV petaluma strain. J Vet Med Sci 1998; 60:315-321. 64. Fevereiro M, Roneker C, Laufs A, Tavares L, de Noronha F. Characterization of two monoclonal antibodies against feline immunodeficiency virus gag gene products and their application in an assay to evaluate neutralizing antibody activity. J Gen Virol 1991; 72:617-622.

122

Lentiviral Vector Systems for Gene Transfer

65. Tozzini F, Matteucci D, Bandecchi P, Baldinotti F, Poli A, Pistello M et al. Simple in vitro methods for titrating feline immunodeficiency virus (FIV) and FIV neutralizing antibodies. J Virol Methods 1992; 37:241-252. 66. Tozzini F, Matteucci D, Bandecchi P, Baldinotti F, Siebelink K, Osterhaus A et al. Neutralizing antibodies in cats infected with feline immunodeficiency virus. J Clin Microbiol 1993; 31:1626-1629. 67. Fevereiro M, Roneker C, de Noronha F. Enhanced neutralization of feline immunodeficiency virus by complement viral lysis. Vet Immunol Immunopathol 1993; 36:191-206. 68. Inoshima Y, Ikeda Y, Kohmoto M, Pecoraro MR, Shimojima M, Shimojima Y et al. Persistence of high virus neutralizing antibody titers in cats experimentally infected with feline immunodeficiency virus. J Vet Med Sci 1996; 58:925-927. 69. Baldinotti F, Matteucci D, Mazzetti P, Giannelli C, Bandecchi P, Tozzini F et al. Serum neutralization of feline immunodeficiency virus is markedly dependent on passage history of the virus and host system. J Virol 1994; 68:4572-4579. 70. Hohdatsu T, Pu R, Torres BA, Trujillo S, Gardner MB, Yamamoto JK. Passive antibody protection of cats against feline immunodeficiency virus infection. J Virol 1993; 67:2344-2348. 71. Beatty JA, Willett BJ, Gault EA, Jarrett O. A longitudinal study of feline immunodeficiency virusspecific cytotoxic T lymphocytes in experimentally infected cats, using antigen-specific induction. J Virol 1996; 70:6199-6206. 72. Jeng CR, English RV, Childers T, Tompkins MB, Tompkins WA. Evidence for CD8+ antiviral activity in cats infected with feline immunodeficiency virus. J Virol 1996; 70:2474-2480. 73. Hohdatsu T, Okubo M, Koyama H. Feline CD8+ T cell non-cytolytic anti-feline immunodeficiency virus activity mediated by a soluble factor(s). J Gen Virol 1998; 79:2729-2735. 74. Hokanson RM, TerWee J, Choi I, Coates J, Dean H, Reddy DN et al. Dose response studies of acute feline immunodeficiency virus PPR strain infection in cats [In Process Citation]. Vet Microbiol 2000; 76:311-327. 75. Flynn JN, Cannon CA, Sloan D, Neil JC, Jarrett O. Suppression of feline immunodeficiency virus replication in vitro by a soluble factor secreted by CD8+ T lymphocytes. Immunology 1999; 96:220-229. 76. Mackewicz CE, Blackbourn DJ, Levy JA. CD8+ T cells suppress human immunodeficiency virus replication by inhibiting viral transcription. Proc Natl Acad Sci USA 1995; 92:2308-2312. 77. Levy JA, Mackewicz CE, Barker E. Controlling HIV pathogenesis: The role of the noncytotoxic anti-HIV response of CD8+ T cells. Immunol Today 1996; 17:217-224. 78. Elyar JS, Tellier MC, Soos JM, Yamamoto JK. Perspectives on FIV vaccine development. Vaccine 1997; 15:1437-1444. 79. Gonin P, Fournier A, Oualikene W, Moraillon A, Eloit M. Immunization trial of cats with a replication-defective adenovirus type 5 expressing the ENV gene of feline immunodeficiency virus. Vet Microbiol 1995; 45:393-401. 80. von Schwedler U, Townsend K, Chada S, Jolly DJ, Elder J, Chang SM et al. Retroviral-mediated expression of FIV envelope/Rev induces CD8 + CTL responses in mice. Intervirology 1997; 40:271-276. 81. Flynn JN, Hosie MJ, Rigby MA, Mackay N, Cannon CA, Dunsford T et al. Factors influencing cellular immune responses to feline immunodeficiency virus induced by DNA vaccination. Vaccine 2000; 18:1118-1132. 82. Yamamoto JK, Okuda T, Ackley CD, Louie H, Pembroke E, Zochlinski H et al. Experimental vaccine protection against feline immunodeficiency virus. AIDS Res Hum Retroviruses 1991; 7:911-922. 83. Matteucci D, Pistello M, Mazzetti P, Giannecchini S, Del Mauro D, Zaccaro L et al. Vaccination protects against in vivo-grown feline immunodeficiency virus even in the absence of detectable neutralizing antibodies. J Virol 1996; 70:617-622. 84. Matteucci D, Pistello M, Mazzetti P, Giannecchini S, Del Mauro D, Lonetti I et al. Studies of AIDS vaccination using an ex vivo feline immunodeficiency virus model: Protection conferred by a fixed-cell vaccine against cell-free and cell-associated challenge differs in duration and is not easily boosted. J Virol 1997; 71:8368-8376. 85. Hosie MJ, Osborne R, Reid G, Neil JC, Jarrett O. Enhancement after feline immunodeficiency virus vaccination. Vet Immunol Immunopathol 1992; 35:191-197.

FIV Vectors

123

86. Siebelink KH, Tijhaar E, Huisman RC, Huisman W, de Ronde A, Darby IH et al. Enhancement of feline immunodeficiency virus infection after immunization with envelope glycoprotein subunit vaccines. J Virol 1995; 69:3704-3711. 87. Osterhaus AD, Tijhaar E, Huisman RC, Huisman W, Darby IH, Francis MJ et al. Accelerated viremia in cats vaccinated with recombinant vaccinia virus expressing envelope glycoprotein of feline immunodeficiency virus. AIDS Res Hum Retroviruses 1996; 12:437-441. 88. Lockridge KM, Chien M, Dean GA, Stefano Cole K, Montelaro RC, Luciw PA et al. Protective immunity against feline immunodeficiency virus induced by inoculation with vif-deleted proviral DNA. Virology 2000; 273:67-79. 89. Hosie MJ, Flynn JN. Feline immunodeficiency virus vaccination: characterization of the immune correlates of protection. J Virol 1996; 70:7561-7568. 90. Flynn JN, Keating P, Hosie MJ, Mackett M, Stephens EB, Beatty JA et al. Env-specific CTL predominate in cats protected from feline immunodeficiency virus infection by vaccination. J Immunol 1996; 157:3658-3665. 91. Pu R, Omori M, Okada S, Rine SL, Lewis BA, Lipton E, Yamamoto JK. MHC-restricted protection of cats against FIV infection by adoptive transfer of immune cells from FIV-vaccinated donors. Cell Immunol 1999; 198:30-43. 92. Hosie MJ, Dunsford T, Klein D, Willett BJ, Cannon C, Osborne R et al. Vaccination with inactivated virus but not viral DNA reduces virus load following challenge with a heterologous and virulent isolate of feline immunodeficiency virus. J Virol 2000; 74:9403-9411. 93. Pistello M, Matteucci D, Cammarota G, Mazzetti P, Giannecchini S, Del Mauro D et al. Kinetics of replication of a partially attenuated virus and of the challenge virus during a three-year intersubtype feline immunodeficiency virus superinfection experiment in cats. J Virol 1999; 73:1518-1527. 94. Nrth TW, Cronn RC, Remington KM, Tandberg RT. Direct comparisons of inhibitor sensitivities of reverse transcriptases from feline and human immunodeficiency viruses. Antimicrob Agents Chemother 1990; 34:1505-1507. 95. Cronn RC, Remington KM, Preston BD, North TW. Inhibition of reverse transcriptase from feline immunodeficiency virus by analogs of 2'-deoxyadenosine-5'-triphosphate. Biochem Pharmacol 1992; 44:1375-1381. 96. North TW, LaCasse RA. Testing anti-HIV drugs in the FIV model. Nat Med 1995; 1:410-411. 97. Lee T, Laco GS, Torbett BE, Fox HS, Lerner DL, Elder JH et al. Analysis of the S3 and S3' subsite specificities of feline immunodeficiency virus (FIV) protease: Development of a broad-based protease inhibitor efficacious against FIV, SIV, and HIV in vitro and ex vivo. Proc Natl Acad Sci USA 1998; 95:939-944. 98. Dunn BM, Pennington MW, Frase DC, Nash K. Comparison of inhibitor binding to feline and human immunodeficiency virus proteases: Structure-based drug design and the resistance problem. Biopolymers 1999; 51:69-77. 99. Hartmann K, Donath A, Beer B, Egberink HF, Horzinek MC, Lutz H et al. Use of two virustatica (AZT, PMEA) in the treatment of FIV and of FeLV seropositive cats with clinical symptoms. Vet Immunol Immunopathol 1992; 35:167-175. 100. Hayes KA, Lafrado LJ, Erickson JG, Marr JM, Mathes LE. Prophylactic ZDV therapy prevents early viremia and lymphocyte decline but not primary infection in feline immunodeficiency virusinoculated cats. J Acquir Immune Defic Syndr 1993; 6:127-134. 101. Smith RA, Remington KM, Lloyd RM, Jr., Schinazi RF, North TW. A novel Met-to-Thr mutation in the YMDD motif of reverse transcriptase from feline immunodeficiency virus confers resistance to oxathiolane nucleosides. J Virol 1997; 71:2357-2362. 102. Smith RA, Remington KM, Preston BD, Schinazi RF, North TW. A novel point mutation at position 156 of reverse transcriptase from feline immunodeficiency virus confers resistance to the combination of (-)-beta-2',3'-dideoxy-3'-thiacytidine and 3'-azido-3'-deoxythymidine. J Virol 1998; 72:2335-2340. 103. Olmsted RA, Hirsch VM, Purcell RH, Johnson PR. Nucleotide sequence analysis of feline immunodeficiency virus: genome organization and relationship to other lentiviruses. Proc Natl Acad Sci USA 1989; 86:8088-8092.

124

Lentiviral Vector Systems for Gene Transfer

104. Talbott RL, Sparger EE, Lovelace KM, Fitch WM, Pedersen NC, Luciw PA et al. Nucleotide sequence and genomic organization of feline immunodeficiency virus. Proc Natl Acad Sci USA 1989; 86:5743-5747. 105. Sparger EE, Shacklett BL, Renshaw-Gegg L, Barry PA, Pedersen NC, Elder JH et al. Regulation of gene expression directed by the long terminal repeat of the feline immunodeficiency virus. Virology 1992; 187:165-177. 106. Ikeda Y, Inoshima Y, Kawaguchi Y, Maeda K, Kohmoto M, Kai C et al. Protein-binding properties of the putative AP-1 and ATF sequences in the feline immunodeficiency virus long terminal repeat. J Gen Virol 1998; 79:95-99. 107. Kawaguchi Y, Tomonaga K, Maeda K, Ono M, Miyazawa T, Kohmoto M et al. The C/EBP site in the feline immunodeficiency virus (FIV) long terminal repeat (LTR) is necessary for its efficient replication and is also involved in the inhibition of FIV LTR-directed gene expression by pseudorabies virus ICP4. Virology 1995; 208:492-499. 108. Thompson FJ, Elder J, Neil JC. Cis- and trans-regulation of feline immunodeficiency virus: Identification of functional binding sites in the long terminal repeat. J Gen Virol 1994; 75:545-554. 109. Kawaguchi Y, Norimine J, Miyazawa T, Kai C, Mikami T. Sequences within the feline immunodeficiency virus long terminal repeat that regulate gene expression and respond to activation by feline herpesvirus type 1. Virology 1992; 190:465-468. 110. Miyazawa T, Kohmoto M, Kawaguchi Y, Tomonaga K, Toyosaki T, Ikuta K et al. The AP-1 binding site in the feline immunodeficiency virus long terminal repeat is not required for virus replication in feline T lymphocytes. J Gen Virol 1993; 74:1573-1580. 111. Bigornia L, Lockridge KM, Sparger EE. Construction and in vitro characterization of attenuated feline immunodeficiency virus long terminal repeat mutant viruses [In Process Citation]. J Virol 2001; 75:1054-1060. 112. Inoshima Y, Kohmoto M, Ikeda Y, Yamada H, Kawaguchi Y, Tomonaga K et al. Roles of the auxiliary genes and AP-1 binding site in the long terminal repeat of feline immunodeficiency virus in the early stage of infection in cats. J Virol 1996; 70:8518-8526. 113. de Parseval A, Elder JH. Demonstration that orf2 encodes the feline immunodeficiency virus transactivating (Tat) protein and characterization of a unique gene product with partial rev activity. J Virol 1999; 73:608-617. 114. Elder JH, Schnolzer M, Hasselkus-Light CS, Henson M, Lerner DA, Phillips TR et al. Identification of proteolytic processing sites within the Gag and Pol polyproteins of feline immunodeficiency virus. J Virol 1993; 67:1869-1876. 115. Morikawa S, Booth TF, Bishop DH. Analyses of the requirements for the synthesis of virus-like particles by feline immunodeficiency virus gag using baculovirus vectors. Virology 1991; 183:288-297. 116. Henderson LE, Bowers MA, Sowder RC, II, Serabyn SA, Johnson DG, Bess JW et al. Gag proteins of the highly replicative MN strain of human immunodeficiency virus type 1: Post-translational modifications, proteolytic processings, and complete amino acid sequences. J Virol 1992; 66:1856-1865. 117. Baboonian C, Dalgleish A, Bountiff L, Gross J, Oroszlan S, Rickett G et al. HIV-1 proteinase is required for synthesis of pro-viral DNA. Biochem Biophys Res Commun 1991; 179:17-24. 118. Egberink HF, Ederveen J, Montelaro RC, Pedersen NC, Horzinek MC, Koolen MJ. Intracellular proteins of feline immunodeficiency virus and their antigenic relationship with equine infectious anaemia virus proteins. J Gen Virol 1990; 71:739-743. 119. Morikawa S, Bishop DH. Identification and analysis of the gag-pol ribosomal frameshift site of feline immunodeficiency virs. Virology 1992; 186:389-397. 120. Wlodawer A, Guschina A, Reshetnikova L, Lubkowski J, Zdanov A, Hui KY et al. Structure of an inhibitor complex of the proteinase from feline immunodeficiency virus. Nat Struct Biol 1995; 2:480-488. 121. Schnolzer M, Rackwitz HR, Gustchina A, Laco GS, Wlodawer A, Elder JH, Kent SB. Comparative properties of feline immunodeficiency virus (FIV) and human immunodeficiency virus type 1 (HIV-1) proteinases prepared by total chemical synthesis. Virology 1996; 224:268-275. 122. Li M, Morris GM, Lee T, Laco GS, Wong CH, Olson AJ et al. Structural studies of FIV and HIV-1 proteases complexed with an efficient inhibitor of FIV protease. Proteins 2000; 38:29-40.

FIV Vectors

125

123. Lin YC, Beck Z, Lee T, Le VD, Morris GM, Olson AJ et al. Alteration of substrate and inhibitor specificity of feline immunodeficiency virus protease. J Virol 2000; 74:4710-4720. 124. North TW, Cronn RC, Remington KM, Tandberg RT, Judd RC. Characterization of reverse transcriptase from feline immunodeficiency virus. J Biol Chem 1990; 265:5121-5128. 125. Amacker M, Hottiger M, Hubscher U. Feline immunodeficiency virus reverse transcriptase: Expression, functional characterization, and reconstitution of the 66- and 51-kilodalton subunits. J Virol 1995; 69:6273-6279. 126. Prasad GS, Stura EA, McRee DE, Laco GS, Hasselkus-Light C, Elder JH et al. Crystal structure of dUTP pyrophosphatase from feline immunodeficiency virus. Protein Sci 1996; 5:2429-2437. 127. Wagaman PC, Hasselkus-Light CS, Henson M, Lerner DL, Phillips TR, Elder JH. Molecular cloning and characterization of deoxyuridine triphosphatase from feline immunodeficiency virus (FIV). Virology 1993; 196:451-457. 128. Steagall WK, Robek MD, Perry ST, Fuller FJ, Payne SL. Incorporation of uracil into viral DNA correlates with reduced replication of EIAV in macrophages. Virology 1995; 210:302-313. 129. Lerner DL, Wagaman PC, Phillips TR, Prospero-Garcia O, Henriksen SJ, Fox HS et al. Increased mutation frequency of feline immunodeficiency virus lacking functional deoxyuridine-triphosphatase. Proc Natl Acad Sci USA 1995; 92:7480-7484. 130. Vink C, van der Linden KH, Plasterk RH. Activities of the feline immunodeficiency virus integrase protein produced in Escherichia coli. J Virol 1994; 68:1468-1474. 131. Shibagaki Y, Holmes ML, Appa RS, Chow SA. Characterization of feline immunodeficiency virus integrase and analysis of functional domains. Virology 1997; 230:1-10. 132. Polard P, Chandler M. Bacterial transposases and retroviral integrases. Mol Microbiol 1995; 15:13-23. 133. Tomonaga K, Itagaki SI, Kashiwase H, Kawaguchi Y, Inoshima Y, Ikeda Y et al. Characterization of an integrase mutant of feline immunodeficiency virus. Arch Virol 1998; 143:1-14. 134. Engelman A, Englund G, Orenstein JM, Martin MA, Craigie R. Multiple effects of mutations in human immunodeficiency virus type 1 integrase on viral replication. J Virol 1995; 69:2729-2736. 135. Tomonaga K, Norimine J, Shin YS, Fukasawa M, Miyazawa T, Adachi A et al. Identification of a feline immunodeficiency virus gene which is essential for cell-free virus infectivity. J Virol 1992; 66:6181-6185. 136. Shacklett BL, Luciw PA. Analysis of the vif gene of feline immunodeficiency virus. Virology 1994; 204:860-867. 137. Chatterji U, Grant CK, Elder JH. Feline immunodeficiency virus Vif localizes to the nucleus. J Virol 2000; 74:2533-2540. 138. Bayer P, Kraft M, Ejchart A, Westendorp M, Frank R, Rosch P. Structural studies of HIV-1 Tat protein. J Mol Biol 1995; 247:529-535. 139. Morikawa S, Lutz H, Aubert A, Bishop DH. Identification of conserved and variable regions in the envelope glycoprotein sequences of two feline immunodeficiency viruses isolated in Zurich, Switzerland. Virus Res 1991; 21:53-63. 140. Hess JL, Small JA, Clements JE. Sequences in the visna virus long terminal repeat that control transcriptional activity and respond to viral trans-activation: involvement of AP-1 sites in basal activity and trans-activation. J Virol 1989; 63:3001-3015. 141. Tomonaga K, Miyazawa T, Sakuragi J, Mori T, Adachi A, Mikami T. The feline immunodeficiency vrus ORF-A gene facilitates efficient viral replication in established T-cell lines and peripheral blood lymphocytes. J Virol 1993; 67:5889-5895. 142. Verschoor EJ, Hulskotte EG, Ederveen J, Koolen MJ, Horzinek MC, Rottier PJ. Post-translational processing of the feline immunodeficiency virus enveope precursor protein. Virology 1993; 193:433-438. 143. Vahlenkamp TW, de Ronde A, Rottier PJ, Horzinek MC, Egberink HF, Verschoor EJ. The feline immunodeficiency virus envelope protein precursor: Functional analysis of a leader deletion mutant. Vet Microbiol 1999; 69:115-116. 144. Maki N, Miyazawa T, Fukasawa M, Hasegawa A, Hayami M, Miki K et al. Molecular characterization and heterogeneity of feline immunodeficiency virus isolates. Arch Virol 1992; 123:29-45.

126

Lentiviral Vector Systems for Gene Transfer

145. Rigby MA, Holmes EC, Pistello M, Mackay A, Brown AJ, Neil JC. Evolution of structural proteins of feline immunodeficiency virus: Molecular epidemiology and evidence of selection for change. J Gen Virol 1993; 74:425-436. 146. Pancino G, Fossati I, Chappey C, Castelot S, Hurtrel B, Moraillon A et al. Structure and variations of feline immunodeficiency virus envelope glycoproteins. Virology 1993; 192:659-662. 147. de Ronde A, Stam JG, Boers P, Langedijk H, Meloen R, Hesselink W et al. Antibody response in cats to the envelope proteins of feline immunodeficiency virus: Identification of an immunodominant neutralization domain. Virology 1994; 198:257-264. 148. Serres PF. Molecular mimicry between the trimeric ectodomain of the transmembrane protein of immunosuppressive lentiviruses (HIV-SIV-FIV) and interleukin 2. C R Acad Sci III 2000; 323:1019-1029. 149. Pollard VW, Malim MH. The HIV-1 Rev protein. Annu Rev Microbiol 1998; 52:491-532. 150. Hope TJ. The ins and outs of HIV Rev. Arch Biochem Biophys 1999; 365:186-191. 151. Phillips TR, Lamont C, Konings DA, Shacklett BL, Hamson CA, Luciw PA et al. Identification of the Rev transactivation and Rev-responsive elements of feline immunodeficiency virus. J Virol 1992; 66:5464-5471. 152. Mancuso VA, Hope TJ, Zhu L, Derse D, Phillips T, Parslow TG. Posttranscriptional effector domains in the Rev proteins of feline immunodeficiency virus and equine infectious anemia virus. J Virol 1994; 68:1998-2001. 153. Donaldson, LJ, Rankin J, Proctor S. Is it possible to catch leukaemia from a cat? Lancet 1994; 344:971-972. 154. Lutz, H. Feline retroviruses: A brief review. Vet Microbiol 1990; 23:131-46. 155. Nowotny, N, Uthman A, Haas OA et al. Is it possible to catch leukemia from a cat? Lancet 1995; 346:252-253. 156. Yamamoto, JK, Hansen H, Ho EW et al. Epidemiologic and clinical aspects of feline immunodeficiency virus infection in cats from the continental United States and Canada and possible mode of transmission. J Am Vet Med Assoc 1989; 194:213-220. 157. Poeschla, EM, Wong-Staal F, Looney DJ. Efficient transduction of nondividing human cells by feline immunodeficiency virus lentiviral vectors. Nat Med 1998; 4:354-357. 158. Johnston, JC, Gasmi M, Lim LE et al. Minimum requirements for efficient transduction of dividing and nondividing cells by feline immunodeficiency virus vectors. J Virol 1999; 73:4991-5000. 159. Curran, MA, Kaiser SM, Achacoso PL et al. Efficient transduction of nondividing cells by optimized feline immunodeficiency virus vectors. Mol Ther 2000; 1:31-38. 160. Alisky, JM, Hughes SM, Sauter SL et al. Transduction of murine cerebellar neurons with recombinant FIV and AAV5 vectors. Neuroreport 2000; 11:2669-2673. 161. Davidson BL , Brooks AI, Stein CS et al. Correction of cellular pathology and behavioral deficits in adult b-glucuronidase-deficient mice after FIV vector-mediated gene transfer to brain. 1. 2000:S250-S251. 162. Derksen TA, Ghodsi A, Redmond EE et al. Gene delivery to primate brain using recombinant FIV-based vectors. 1. 2000:S107. 163. DePolo, NJ, Reed JD, Sheridan PL et al. VSV-G pseudotyped lentiviral vector particles produced in human cells are inactivated by human serum. Mol Ther 2000; 2:218-222. 164. Ikeda, Y, Tomonaga K, Kawaguchi Y et al. Feline immunodeficiency virus can infect a human cell line (MOLT-4) but establishes a state of latency in the cells. J Gen Virol 1996; 77(Pt 8):1623-1630. 165. Miyazawa, T, Kawaguchi Y, Kohmoto M et al. Comparative functional analysis of the various lentivirus long terminal repeats in human colon carcinoma cell line (SW480 cells) and feline renal cell line (CRFK cells). J Vet Med Sci 1994; 56:895-899. 166. Miyazawa, T, Kawaguchi Y, Kohmoto M et al. Production of feline immunodeficiency virus in feline and non-feline non-lymphoid cell lines by transfection of an infectious molecular clone. J Gen Virol 1992; 73(Pt 6):1543-1546. 167. Tochikura, TS, Tanabe-Tochikura A, Hayes KA et al. Fusion activity dissociated from replication ability in feline immunodeficiency virus (FIV) in human cells. J Acquir Immune Defic Syndr 1993; 6:1301-1310. 168. Johnston, J, Power C. Productive infection of human peripheral blood mononuclear cells by feline immunodeficiency virus: implications for vector devlopment. J Virol 1999; 73:2491-2498.

FIV Vectors

127

169. DuBridge, RB, Tang P, Hsia HC et al. Analysis of mutation in human cells by using an EpsteinBarr virus shuttle system. Mol Cell Biol 1987; 7:379-387. 170. Takeuchi, Y, Cosset FL, Lachmann PJ et al. Type C retrovirus inactivation by human complement is determined by both the viral genome and the producer cell. J Virol 1994; 68:8001-8007. 171. DePolo, NJ, Harkleroad CE, Bodner M et al. The resistance of retroviral vectors produced from human cells to serum inactivation in vivo and in vitro is primate species dependent. J Virol 1999; 73:6708-6714. 172. Takeuchi, Y, Liong SH, Bieniasz PD et al. Sensitization of rhabdo-, lenti-, and spumaviruses to human serum by galactosyl(alpha1-3)galactosylation. J Virol 1997; 71:6174-6178. 173. Cosset, FL, Takeuchi Y, Battini JL et al. High-titer packaging cells producing recombinant retroviruses resistant to humanserum. J Virol 1995; 69:7430-7436. 174. Blaese, RM, Culver KW, Miller AD et al. T lymphocyte-directed gene therapy for ADA- SCID: Initial trial results after 4 years. Science 195; 270:475-480. 175. Soneoka, Y, Cannon PM, Ramsdale EE et al. A transient three-plasmid expression system for the production of high titer retroviral vectors. Nucleic Acids Res 1995; 23:628-633. 176. Chang, LJ, McNulty E, Martin M. Human immunodeficiency viruses containing heterologous enhancer/promoters are replication competent and exhibit different lymphocyte tropisms. J Virol 1993; 67:743-752. 177. Kim, VN, Mitrophanous K, Kingsman SM et al. Minimal requirement for a lentivirus vector based on human immunodeficiency virus type 1. J Virol 1998; 72:811-816. 178. Bukrinsky, MI, Haggerty S, Dempsey MP et al. A nuclear localization signal within HIV-1 matrix protein that governs infection of non-dividing cells. Nature 1993; 365:666-669. 179. Bray, M, Prasad S, Dubay JW et al. A small element from the Mason-Pfizer monkey virus genome makes human immunodeficiency virus type 1 expression and replication Rev-independent. Proc Natl Acad Sci USA 1994; 91:1256-1260. 180. Ernst, RK, Bray M, Rekosh D et al. A structured retroviral RNA element that mediates nucleocytoplasmic export of intron-containing RNA. Mol Cell Biol 1997; 17:135-144. 181. Yu, SF, von Ruden T, Kantoff PW et al. Self-inactivating retroviral vectors designed for transfer of whole genes into mammalian cells. Proc Natl Acad Sci USA 1986; 83:3194-3198. 182. Yee, JK, Moores JC, Jolly DJ et al. Gene expression from transcriptionally disabled retroviral vectors. Proc Natl Acad Sci USA 1987; 84:5197-5201. 183. Julias, JG, Hash D, Pathak VK . E- vectors: Development of novel self-inactivating and self-activating retroviral vectors for safer gene therapy. J Virol 1995; 69:6839-6846. 184. Naviaux, RK, Costanzi E, Haas M et al. The pCL vector system: rapid production of helper-free, high-titer, recombinant retroviruses. J Virol 1996; 70:5701-5705. 185. Miyoshi, H, Blomer U, Takahashi M et al. Development of a self-inactivating lentivirus vector. J Virol 1998; 72:8150-8157. 186. Deglon, N, Tseng JL, Bensadoun JC et al. Self-inactivating lentiviral vectors with enhanced transgene expression as potential gene transfer system in Parkinson’s disease. Hum Gene Ther 2000; 11:179-190. 187. Zufferey, R, Donello JE, Trono D et al. Woodchuck hepatitis virus posttranscriptional regulatory element enhances expression of transgenes delivered by retroviral vectors. J Virol 1999; 73:2886-2892. 188. Donello, JE, Loeb JE, Hope TJ . Woodchuck hepatitis virus contains a tripartite post-transcriptional regulatory element. J Virol 1998; 72:5085-5092. 189. Schambach, A, Wodrich H, Hildinger M et al. Context dependence of different modules for posttranscriptional enhancement of gene expression from retroviral vectors. Mol Ther 2000; 2:435-2445. 190. Ramezani, A, Hawley TS, Hawley RG. Lentiviral vectors for enhanced gene expression in human hematopoietic cells. Mol Ther 2000; 2:458-49. 191. Zennou, V, Petit C, Guetard D et al. HIV-1 genome nuclear import is mediated by a central DNA flap. Cell 2000; 101:173-85. 192. Follenzi, A, Ailles LE, Bakovic S et al. Gene transfer by lentiviral vectors is limited by nuclear translocation and rescued by HIV-1 pol sequences. Nat Genet 2000; 25:217-222. 193. Schneider, R, Campbell M, Nasioulas G et al. Inactivation of the human immunodeficiency virus type 1 inhibitory elements allows Rev-independent expression of Gag and Gag/protease and particle formation. J Virol 1997; 71:4892-4903.

128

Lentiviral Vector Systems for Gene Transfer

194. Schwartz, S, Campbell M, Nasioulas G et al. Mutational inactivation of an inhibitory sequence in human immunodeficiency virus type 1 results in Rev-independent gag expression. J Virol 1992; 66:7176-7182. 195. Haas, J, Park EC, Seed B. Codon usage limitation in the expression of HIV-1 envelope glycoprotein. Curr Biol 1996; 6:315-324. 196. zur Megede, J, Chen MC, Doe B et al. Increased expression and immunogenicity of sequencemodified human immunodeficiency virus type 1 gag gene. J Virol 2000; 74:2628-2635. 197. Yee, JK, Miyanohara A, LaPorte P et al. A general method for the generation of high-titer, pantropic retroviral vectors: Highly efficient infection of primary hepatocytes. Proc Natl Acad Sci USA 1994; 91:9564-9568. 198. Miller, DG, Edwards RH, Miller AD. Cloning of the cellular receptor for amphotropic murine retroviruses reveals homology to that for gibbon ape leukemia virus. Proc Natl Acad Sci USA 1994; 91:78-82. 199. van Zeijl, M, Johann SV, Closs E et al. A human amphotropic retrovirus receptor is a second member of the gibbon ape leukemia virus receptor family. Proc Natl Acad Sci USA 1994; 91:1168-1172. 200. Burns, JC, Friedmann T, Driever W et al. Vesicular stomatitis virus G glycoprotein pseudotyped retroviral vectors: concentration to very high titer and efficient gene transfer into mammalian and nonmammalian cells. Proc Natl Acad Sci USA 1993; 90:8033-8037. 201. Kavanaugh, MP, Miller DG, Zhang W et al. Cell-surface receptors for gibbon ape leukemia virus and amphotropic murine retrovirus are inducible sodium-dependent phosphate symporters. Proc Natl Acad Sci USA 1994; 91:7071-7075. 202. Kozak, SL, Sess DC, Kavanaugh MP et al. The envelope glycoprotein of an amphotropic murine retrovirus binds specifcally to the cellular receptor/phosphate transporter of susceptible species. J Virol 1995;69:3433-3440. 203. Schlegel, R, Tralka TS, Willingham MC et al. Inhibition of VSV binding and infectivity by phosphatidylserine: is phosphatidylserine a VSV-binding site? Cell 1983; 32:639-646. 204. Mastromarino, P, Conti C, Goldoni P et al. Characterization of membrane components of the erythrocyte involved in vesicular stomatitis virus attachment and fusion at acidic pH. J Gen Virol 1987; 68(Pt 9):2359-2369. 205. Schlegel, R, Willingham MC, Pastan IH. Saturable binding sites for vesicular stomatitis virus on the surface of Vero cells. J Virol 1982; 43:871-875. 206. Lin, S, Gaiano N, Culp P et al. Integration and germ-line transmission of a pseudotyped retroviral vector in zebrafish. Science 1994; 265:666-669. 207. Matsubara, T, Beeman RW, Shike H et al. Pantropic retroviral vectors integrate and express in cells of the malaria mosquito, Anopheles gambiae. Proc Natl Acad Sci USA 1996; 93:6181-6185. 208. Matlin, KS, Reggio H, Helenius A et al. Pathway of vesicular stomatitis virus entry leading to infection. J Mol Biol 1982; 156:609-631. 209. Stitz, J, Buchholz CJ, Engelstadter M et al. Lentiviral vectors pseudotyped with envelope glycoproteins derived from gibbon ape leukemia virus and murine leukemia virus 10A1. Virology 2000; 273:16-20. 210. Human gene marker/therapy clinical protocols (complete updated listings). Hum Gene Ther 2000; 11:2543-2617. 211. Liu, ML, Winther BL, Kay MA. Pseudotransduction of hepatocytes by using concentrated pseudotyped vesicular stomatitis virus G glycoprotein (VSV-G)-Moloney murine leukemia virusderived retrovirus vectors: Comparison of VSV-G and amphotropic vectors for hepatic gene transfer. J Virol 1996; 70:2497-2502. 212. Haas, DL, Case SS, Crooks GM et al. Critical factors influencing stable transduction of human CD34(+) cells with HIV-1-derived lentiviral vectors. Mol Ther 2000; 2:71-80. 213. Robinson, D, Elliott JF, Chang LJ. Retroviral vector with a CMV-IE/HIV-TAR hybrid LTR gives high basal expression levels and is up-regulated by HIV-1 Tat. Gene Ther 1995; 2:269-278. 214. Wang, G, Slepushkin V, Zabner J et al. Feline immunodeficiency virus vectors persistently transduce nondividing airway epithelia and correct the cystic fibrosis defect. J Clin Invest 1999; 104:R55-R62.

FIV Vectors

129

215. Kafri, T, Blomer U, Peterson DA et al. Sustained expression of genes delivered directly into liver and muscle by lentiviral vectors. Nat Genet 1997; 17:314-317. 216. Zufferey, R, Nagy D, Mandel RJ et al. Multiply attenuated lentiviral vector achieves efficient gene delivery in vivo. Nat Biotechnol 1997; 15:871-875. 217. Zhang, J, Temin HM. Rate and mechanism of nonhomologous recombination during a single cycle of retroviral replication. Science 1993; 259:234-238. 218. Hu, WS, Temin HM. Genetic consequences of packaging two RNA genomes in one retroviral particle: Pseudodiploidy and high rate of genetic recombination. Proc Natl Acad Sci USA 1990; 87:1556-1560. 219. Pensiero, MN, Wysocki CA, Nader K et al. Development of amphotropic murine retrovirus vectors resistant to inactivation by human serum. Hum Gene Ther 1996; 7:1095-1101. 220. Rigg, RJ, Chen J, Dando JS et al. A novel human amphotropic packaging cell line: high titer, complement resistance, and improved safety. Virology 1996; 218:290-295. 221. Mason, JM, Guzowski DE, Goodwin LO et al. Human serum-resistant retroviral vector particles from galactosyl (alpha1-3) galactosyl containing nonprimate cell lines. Gene Ther 1999; 6:1397-1405. 222. Johnson, LG, Olsen JC, Sarkadi B et al. Efficiency of gene transfer for restoration of normal airway epithelial function in cystic fibrosis. Nat Genet 1992; 2:21-25. 223. Fong, TC, Sauter SL, Ibanez CE et al. The use and development of retroviral vectors to deliver cytokine genes for cancer therapy. Crit Rev Ther Drug Carrier Syst 2000; 1

CHAPTER 7

EIAV, CAEV and Other Lentivirus Vector Systems John C. Olsen

Abstract

L

entiviruses that infect non-primates make up a diverse collection of viruses. Although these viruses have some features in common with HIV and other primate viruses, differences in genome organization and viral gene function have made the successful derivation of vectors from non-primate lentiviruses unpredictable. This Chapter discusses the construction and application of gene transfer systems derived from four non-primate lentiviruses including equine infectious anemia virus (EIAV), caprine arthritis encephalitis virus (CAEV), visna virus, and Jembrana disease virus (JDV).

Introduction A general property of lentiviruses is the ability to integrate their genomes into the chromosomal DNA of non-dividing infected cells. This property, which is not shared by simple retroviruses such as murine leukemia virus or avian leukosis/sarcoma virus, is the main rationale for developing lentiviruses for gene transfer applications.1 At present vector systems have been derived from eight different lentiviruses including HIV-1, HIV-2, SIV, FIV, EIAV, CAEV, visna virus, and JDV. Earlier Chapters in this volume describe the development and properties of the first four of these vector systems, while this Chapter describes the construction and properties of the latter four gene transfer systems.

Biological Similarities and Differences of Lentiviruses An important question is why so many different lentiviruses are being explored for application as gene transfer systems. As a group lentiviruses have a number of properties in common; however, there are enough differences between individual lentiviruses that make it worthwhile to explore whether these differences translate into useful gene transfer systems. All lentiviruses share the property of replicating in non-dividing terminally differentiated cells. Lentiviruses are highly species-specific and primary virus isolates replicate efficiently only in cells isolated from their species. Lentiviruses share the property of infecting cells of the monocyte/macrophage lineage although a subset of lentiviruses causing immunodeficiencies also target CD4 lymphocytes. The infection of cells involved in the immune system accounts for dissemination of virus throughout the body and the multi-organ involvement typically seen in lentivirus infections. The progression to disease following natural infection can be rapid, on the order of weeks as with EIAV or Jembrana disease virus, or can take months to years to develop, as with visna virus, CAEV or HIV. Lentiviral Vector Systems for Gene Transfer, edited by Gary L. Buchschacher, Jr. ©2003 Eurekah.com.

132

Lentiviral Vector Systems for Gene Transfer

Fig. 1. Genomic organization of selected lentiviruses. Shaded boxes indicate gag, pol and env genes. Solid boxes indicate accessory genes.

EIAV, CAEV and Other Lentiviral Vectors

133

The genetic organization for these lentiviruses is shown in Figure 1. The lentiviral genome contains gag, pol, and env genes that encode the structural and enzymatic proteins that make up infectious viral particles. These viruses also contain tat and rev accessory genes that encode proteins that regulate gene expression. While these viruses typically have multiple exons encoding tat and rev some differences can be noted for the placement of the coding exons for these genes. For example, the first exon of EIAV tat is located 5' to the gag gene before the major splice donor site. An unusual CUG codon is used to initiate EIAV tat translation.2 An noteworthy feature in the organization of JDV is that both env and the second exon encoding rev extend into the 3' LTR sequence,3 similar to the nef gene of HIV-1 and other primate lentiviruses. EIAV, CAEV, visna, and JDV all contain a third accessory gene. For CAEV, visna, and JDV this gene appears to be analogous to HIV vif. The placement of vif for each is similar and overlaps the 3' end of pol. Based on comparisons of the predicted protein, it is thought that the non-primate Vif proteins are functionally analogous to HIV Vif.

EIAV S2 Gene EIAV differs from the other lentiviruses in lacking a gene encoding a recognizable Vif protein. Instead EIAV contains a gene called S2, which is located in the pol-env intergenic region. The S2 protein is expressed from a singly spliced tricistronic mRNA, also potentially coding for Tat and Env.2,4 Translation of this mRNA to make S2 uses an AUG codon 10 nt downstream of the open reading frame for Tat and 26 nt upstream of the env AUG in an alternate reading frame. S2 encodes a 65 amino acid protein and has several potential functional motifs that are conserved among various isolates and remain conserved during virus genome evolution within an infected animal.5 These sequences conform to nucleoporin (GLFP), SH3 domain binding (PXXP) and nuclear localization (RRKQETKK) motifs. In vitro synthesized S2 has been shown to react with anti-serum from EIAV infected horses,2 so S2 must be expressed in vivo. The role of S2 in virus replication remains unclear. Mutagenesis of S2 in an infectious, pathogenic molecular clone of EIAV had no affect on viral replication kinetics in equine monocyte/macrophage cultures.6 However, mutation of S2 severely affected viral replication and virulence of EIAV in experimentally inoculated ponies.5

dUTPase A common feature of non-primate lentiviruses is the presence of sequences encoding a dUTPase protein. For FIV, EIAV, CAEV and visna virus the dUTPase is encoded by sequences within the pol gene between sequences encoding RT and IN.7 As with RT and IN, dUTPase is incorporated into viral particles. The bovine lentiviruses, BIV and JDV, lack dUTPase.3 Primate lentiviruses including HIV and SIV strains also lack a functional dUTPase, although sequences related to dUTPase genes can be found within the env gene of HIV-1.8 Type B and D retroviruses also contain sequences encoding dUTPase but these sequences are located in the pro reading frame.9,10 Other viruses including herpesviruses11 and poxviruses12 have dUTPase activity as do most, if not all, prokaryotic and eukaryotic organisms, including humans.13 Cellular dUTPase is part of an important biosynthetic pathway that catalyzes dUTP hydrolysis to dUMP and PPi with dUMP being converted to TTP by thymidylate synthase. The dUTPase helps to maintain a low dUTP/TTP ratio (typically ~10-5) which is thought to be important for minimizing the incorporation of uracil into DNA.14,15 Cellular dUTPase levels appear to be regulated during the cell cycle and as a function of differentiation; as a norm quiescent cells and terminally differentiated cells have lower levels of dUTPase than dividing cells and non-differentiated cells.16-21 This provides a rationale for the presence of dUTPase in lentiviruses whose target cells may be non-dividing. The role of viral dUTPase may be analogous to cellular dUTPase to prevent incorporation of uracil into reverse transcribed DNA.

134

Lentiviral Vector Systems for Gene Transfer

Since uracil can base pair with guanosine this could lead to high G-to-A mutations if dU were incorporated during first strand synthesis. A number of studies support a role for dUTPase during lentiviral replication. Mutation of dUTPase in FIV,22 EIAV,23,24 CAEV,25 and visna virus25 leads to delayed replication in nondividing primary macrophage cultures, but normal kinetics of replication in dividing primary monocytes or permissive cell lines. In animals, dUTPase mutants show reduced viral loads in EIAV infected ponies,24 visna virus infected sheep,26 attenuated severity of arthritic lesions in CAEV infected goats,27 and decreased viral burden in some tissues of FIV infected cats.28 However, the neuropathogenic effect of dUTPase mutants was not different than wild-type virus in FIV or visna virus infected animals.26,29 Mechanistic studies support the role of non-primate lentiviral dUTPase in preventing the incorporation of dUTP into DNA during reverse transcription. G-to-A transition rates were significantly elevated with dUTPase mutants compared to wild type virus during virus evolution in CAEV infected goats27 or FIV infected cats.28 Steagall et al investigated the molecular defect causing the delay of replication in EIAV infected macrophage cultures where dUTPase mutants replicate to 1-2% of wild type levels by 7 days post-infection.30 It was found that reverse transcription was normal with full-length linear DNA evident 24 hr after infection. Integration of dUTPase mutant DNA was slightly depressed (~2-fold) by 72 hr post-infection. Transcription levels, however, were significantly depressed with the dUTPase mutant. By 72 hr post-infection analysis, of RNA by Northern blotting showed about a 25-fold decrease in multiply spliced messages and an 85-fold to 300-fold decrease in full-length RNA in cells infected with the dUTPase mutant compared to wild type virus. The decrease in RNA synthesis could account for the decrease in virus production in these cultures. At present it is not clear how RNA synthesis is depressed with the dUTPase mutant. Possibly, significant incorporation of dU into the enhancer/promoter region could affect the interaction of transcription factors with their binding elements. Using a PCR assay of uracil-DNA glycosylase treated DNA, Steagall et al provided evidence that dU was incorporated into proviral DNA in macrophage cultures infected with the EIAV dUTPase mutant.30

Recent Vector Developments Equine Lentiviruses Equine infectious anemia virus (EIAV) causes a persistent infection and a chronic, usually non-fatal disease in horses and closely related equines (donkeys and mules) worldwide.31 The clinical course of the disease can be somewhat variable, depending upon the dose and virulence of the particular strain and the susceptibility of the horse. Since insects commonly transmit the virus, the disease is most prevalent in warm climates. The virus does not replicate in insects; instead, blood is carried from animal-to-animal on the mouthparts of biting insects.32 During the acute phase of EIAV infection, fever and viremia result due to infection of peripheral and tissue macrophages. Acute viremia generally occurs within the first 1-4 weeks post-infection. The chronic phase of the disease follows and is characterized by recurring cycles of viremia and associated clinical symptoms including fever, anorexia, edema, lethargy, diarrhea, anemia, leukopenia, glomerulonephritis, thrombocytopenia, and hemorrhaging due to severe depletion of platelets.31 Neurologic symptoms can also occur although are relatively infrequent. Each clinical episode lasts from 3-6 days. The period between each episode is irregular, lasting from weeks to months. The frequency of episodes generally declines with time. After 6-8 episodes over the course of about a year the chronic stage of the diseases ends and >90% of infected animals become clinically quiescent and remain life-long inapparent carriers. During the course of infection within an animal, evolution of EIAV genomic quasispecies occurs in concert with a maturation of the host immune system.33 Dynamic changes in both

EIAV, CAEV and Other Lentiviral Vectors

135

host and virus contribute to the eventual regulation of virus replication in inapparent carriers. Some recent work has focussed on the viral determinants that influence the virulence. This work has implicated the viral LTR, env, S2, and rev as having important roles in pathogenesis.5,34-36 The roles that these determinants have in disease progression are as yet incompletely elucidated. The recent construction of pathogenic molecular clones of EIAV will be important tools for dissecting viral elements that contribute to the disease process.34,35

EIAV Vector Production

Two groups have described lentiviral vector systems derived from EIAV.37,38 The design of these systems is based on separation of cis-acting elements in the EIAV genome from sequences encoding trans-acting proteins (Fig. 2). The viral cis-acting elements for replication and integration are contained in the gene transfer vector plasmid. Two other plasmids, the viral protein expression cassette and an envelope expression plasmid, provide the genes encoding the viral proteins necessary for assembling the RNA form of the gene transfer vector into infectious viral particles.

EIAV Gene Transfer Vectors The initial EIAV gene transfer vectors were designed to be produced by transient threeplasmid transfection of human 293 cells.37 Since the EIAV promoter elements function poorly in these cells, the immediate early CMV promoter/enhancer region was substituted for the U3 region of the 5' LTR. In these CMV-R-U5 constructs the transcription start site was placed to be identical to the EIAV transcription start site at the boundary of the R region (Fig. 2). An internal promoter (CMV or murine leukemia virus) was used for expression of a lacZ reporter gene or the puromycin resistance gene in target cells. All of the EIAV U3 sequences in the 5' LTR were deleted. A complete EIAV 3' LTR was included downstream of the transgene. The titers of these initial constructs were in the range of 6x104 and 2x105 transducing units/ml and vectors were shown to efficiently transduce aphidicolin-arrested human cells.37 In an effort to improve titers, sequences including the EIAV RRE region were placed in the downstream position between the transgene and the 3' LTR (Fig. 2). In general, RRE containing constructs have a 5-20 fold improvement in vector titers (M Patel and JC Olsen, unpublished data).38 To minimize expression of gag-pol sequences present in the vector other, modifications have been made. First, the AUG start codon of the gag open reading frame was changed to a UAG stop codon without adversely affecting vector titers (M Patel and JC Olsen, unpublished data). Second, deletion analysis of the gag sequence has shown that for optimal titers about 373 nucleotides of gag sequence are required.38 The requirement for gag sequences is not yet completely understood, but it is likely that that sequences in this region contribute to RNA encapsidation and/or RNA stability.

EIAV Viral Protein Expression Cassettes The first EIAV viral protein expression cassettes contained intact open reading frames for Gag-Pol and the accessory genes tat, rev, and S2.37,38 In the pEV53 construct, env was disrupted by a 736 nt Hind III internal deletion of the SU coding region and a 168 nt deletion at the 3' end of the gene effectively truncating the TM coding region (Fig. 2).37 Ideally, the protein expression cassette used for expression of viral proteins should be devoid of all viral cisacting elements including the RNA encapsidation signal and cis-acting sequences involved in DNA synthesis and integration. Deletion of viral cis-acting elements in the protein expression cassette is important in that it minimizes the chances for generation of replication competent viruses by recombination events. The initial protein expression cassettes used for vector production lacked many cis-acting sequences rendering the viral protein cassette defective for replication. However, not all cis-acting sequences were removed because of uncertainties regarding

136

Lentiviral Vector Systems for Gene Transfer

Fig. 2. Schematic of EIAV vector system. SD, major splice donor site. pA, polyadenylation signal. R-U5, sequence domains from EIAV LTR.

possible effects on viral protein expression. Thus the pEV53 viral expression vector contained an essentially intact 5' untranslated region including a portion of the viral LTR, the tRNA primer binding site for first strand DNA synthesis and sequences in the 5' leader sequence predicted for RNA encapsidation.37 Certain cis-acting sequences required for replication were deleted, however, rendering pEV53 incapable of replication. These include cis-acting sequences from the 3' end of the genome including the polypurine tract adjacent to the 3’LTR used for initiation of second strand DNA synthesis, and the entire 3' LTR sequence, including cisacting sequences required for integration. More recent modifications have been made to further disable the replication potential of the protein expression cassette. This work has also addressed the role of accessory protein expression for EIAV vector production. Deletion of the 5' leader region, including all LTR sequences and the tRNA primer binding site up to the major splice donor sequence can be accomplished without adversely affecting expression of gag-pol in human 293 cells (M Patel and JC Olsen, unpublished data). This deletion also spans the first exon of tat. Since Tat transactivation of the viral LTR has been shown to be non-functional in human cells, this finding is not surprising. However, exclusion of tat from the vector system is desirable since this eliminates any possible unknown adverse affects that might be caused by delivery of Tat to host cells. Mitrophanous and colleagues have shown that S2 gene can be eliminated from the vector system.38 Deletion of S2 sequences from both the protein expression cassette and from the RRE region of the gene transfer vector had no effect on vector titers or the ability to transfer genes to growth arrested cells or to cultured rat neurons. This result is consistent with the results of others who have shown that mutations in S2 have no affect on the ability of the wildtype virus to infect equine cells in culture, including monocyte-derived macrophages.6 Since EIAV and other non-primate lentiviruses encode sequences for a dUTPase activity not contained in HIV-1, it is of interest to determine the role of this gene in vector production. As previous studies with the wild-type virus had indicated that the dUTPase activity was important for infection of equine macrophages but not dividing monocytes,23,24,30 it was thought that the dUTPase activity might be generally important for transduction of non-dividing cells. To test this, the active site for dUTPase activity was mutated in the pONY3.1 protein expression cassette using an identical mutation that had been shown to adversely affect virus

EIAV, CAEV and Other Lentiviral Vectors

137

replication in quiescent equine macrophage cultures.30 Interestingly, it was found that the EIAV dUTPase was not required for gene transfer to aphidicolin arrested cells or to rat neurons in culture.38 It is unknown, however, if the dUTPase activity will be dispensable for all gene transfer targets or if it will be important for specific applications.

Pseudotyping EIAV Vectors The natural host range of EIAV is restricted to specific cell types and tissues in infected equines. Infected cells in the host include monocytes and macrophages, which are also efficiently infected in culture. Expression of EIAV in monocytes is very low, and increases greatly when cells differentiate to macrophages.39,40 Other cells including endothelial cells have also been shown to be infected in vivo.41 Presumably, the viral tropism reflects, in part, the distribution of receptors recognized by EIAV Env. At present the natural receptors for EIAV have not been identified. Also, it is not known if vectors containing the EIAV Env proteins can infect human cells. Primary isolates of virus from infected equines can be adapted to replicate on equine fibroblasts and canine cell lines so it is possible that the natural receptor is present on these cells or that the adaptive process results in an expanded host range. To extend the host range to human cells, EIAV vectors were pseudotyped with the envelope glycoprotein (G) from vesicular stomatitis virus (VSV), a rhabdovirus.37,38 VSV-G pseudotyped EIAV particles appear to be stable and can be concentrated by high-speed centrifugation42 or by ultracentrifugation.38,43 EIAV vectors have also been successfully pseudotyped with the G protein from a second rhabdovirus, rabies virus, and with the ecotropic envelope and amphotropic envelope (4070A) from murine leukemia virus, although some variability in efficiency was noted.38 Since current efforts aimed at achieving cell-type specific targeting are being done with the MLV envelopes, it seems feasible that this developing technology could be applied for targeted gene transfer by EIAV vectors as well.

Gene Transfer Applications for EIAV Vectors VSV-G pseudotyped vectors have been shown to mediate transfer of GFP or lacZ reporter genes to cultured rat neurons or to adult rat brain with transduction efficiencies similar to HIV-derived vectors.38 Transduction of cultured differentiated hippocampal neurons was shown to be extensive even at a moderate (5-10) multiplicity of infection. Co-localization of GFP with the Neu-N neuronal specific marker was used to demonstrate that post-mitotic neurons could be transduced. Injection of concentrated EIAV vectors (0.6-15 x 104 infectious units) into the caudate putamen of rat brain striatum resulted in an average of 104 transduced cells spanning the striatum. Cells of both glial and neuronal appearance were transduced. Transduction of neuronal nuclei of the neighboring lateral globus pallidus were also observed. No pathological change in the injected brain area was noted 2, 7 or 28 days following gene transfer. Gene expression persisted for 28 days, the longest time point examined. These results are significant in that they demonstrate that EIAV vectors can achieve efficient direct in vivo gene transfer and expression in post-mitotic cells. For development of lentiviral vectors for gene therapy it will be important to evaluate lentiviral vectors from non-primate and primate viruses to determine if there is a scientific basis to choose one vector system over another. These studies may also be useful to improve individual vector systems. In a study by Yamada and colleagues, complementation of the defect in cells from Fanconi anemia (FA) patients was investigated using vectors derived from EIAV, HIV, and MLV.44 Although differences in the vector backbone, including different promoters, precluded a true head-to-head comparison, some interesting observations were made. FA is a disease characterized by bone marrow failure secondary to hematopoietic stem cell dysfunction. Previous studies have shown that MLV-retroviruses could be used in an ex vivo approach

138

Lentiviral Vector Systems for Gene Transfer

using transduction of bone marrow cells for successful treatment of a FA mouse model.45 Clinical trials, however, using MLV vectors have not yet demonstrated lasting improvement.46 Thus the Yamada study represents an initial effort to determine the utility of using lentiviral vectors for treatment of this disease. EBV-transformed lymphoblasts derived from FA group C patients were transduced with vectors containing the normal FA-C cDNA. Phenotypic correction of FA cells, as measured by drug resistance to mitomycin C, was demonstrated for EIAV, HIV and MLV-derived vectors. Cells transduced by all three vectors showed normalized cell cycle kinetics and significantly less chromosomal damage after exposure to mitomycin C.44 Differences were noted, however, in gene expression by the various vectors. EIAV vectors expressed about one-half to one-third of the steady state levels of RNA in transduced lymphoblasts, compared to HIV or MLV vectors. Also, the EIAV vector directed the synthesis of an apparently full length RNA at levels about equivalent to RNA directed from an internal CMV promoter. This result suggests that the EIAV LTR, even in the absence of EIAV Tat, can actively direct transcription in human cells. Thus in order to prevent transcription from the 5' LTR in transduced cells, which may interfere with transcription from internal promoters, removal of transcription elements from the U3 region to generate self-inactivating (SIN) vectors seems advisable.

Lentiviruses of Sheep and Goats The ovine-caprine lentiviruses are separable into two distinct groups. These include caprine arthritis-encephalitis virus (CAEV) and maedi/visna virus. These viruses cause chronic diseases often with long incubation periods.31,47 Caprine arthritis-encephalitis virus (CAEV) causes subacute progressive encephalomyelitis in young goats and chronic arthritis in adult animals. CAEV is widespread in goat populations and can spread rapidly in industrialized countries where overcrowding of animals and the pooling of milk to feed young animals enhances virus spread. The virus infects monocytes/macrophages and transmission occurs when young goats ingest macrophages in milk. Clinical signs of lameness or ataxia can be observed as early as one month in animals infected at birth. The adult disease starts out as synovitis which progresses to crippling arthritis over time. A progressive neurological disease has also been documented.48 Visna virus causes chronic pneumonitis (called maedi) and a progressive demyelinating disease in sheep that leads to wasting and paralysis (referred to as visna).47,49 These diseases develop over the course of months to years after the initial infection. Visna virus is transmitted in body fluids, primarily in aerosols from respiratory exudates and in milk. The major cell targets in vivo for CAEV and visna are monocytes and tissue macrophages. As with EIAV, expression of CAEV or visna is very low in monocytes, but increases greatly when the cells differentiate into macrophages.50-53

CAEV-Based Vectors An initial attempt to generate vectors from caprine arthritis encephalitis virus (CAEV) resulted in vectors that transferred genes with low transduction efficiencies.54 In this study, two gene transfer vectors were constructed in which reporter genes were used to replace all of the viral genes from the start of gag to all but about 180 nucleotides of the env gene (Fig. 3). One gene transfer vector (pBNL2) contained a neo gene expressed from full length RNA transcripts and a lacZ gene expressed by splicing using the major CAEV splice donor site in the 5' leader region and a heterologous splice donor site between neo and lacZ. A second vector (pCSHL) contained the major splice donor, no splice acceptor and a phleomycin-lacZ fusion gene (Fig. 3). For virus production, the gene transfer vectors were first transfected into goat embryo fibroblasts and cell lines generated using the selectable markers within the vectors. To rescue vector sequences into virus, the permissive cell lines were infected with a replication competent

EIAV, CAEV and Other Lentiviral Vectors

139

Fig. 3. Schematic of CAEV gene transfer vectors. SD, major splice donor site. SA, splice acceptor. SH, phleomycin selection marker.

strain of CAEV and virus stocks were harvested four days later. Rescue of the gene transfer vectors was found to be very inefficient, resulting in lacZ titers of 10-187 TU/ml for the pBNL2 vector, and 17-40 TU/ml for the pCSHL vector, whereas the helper virus was produced with titers ranging from 105 to 7x105 TCID50/ml.54 A detailed analysis of vector RNA in CAEV-infected producer cells showed that for the pBNL2 vector, both vector-length and spliced RNA could be detected by Northern blot analysis of total cellular RNA; however, only spliced sub-genomic RNA was detected in the cytoplasm.54 Subsequent analysis of virion RNA or RNA in viral vector transduced cells confirmed that the spliced RNA form was packaged, albeit at low efficiency. Vector-length pCSHL RNA accumulated in the cytoplasm at higher concentrations than vector length pBNL2 RNA, suggesting that CAEV RNA containing a splice donor but no splice acceptor can be exported from the nucleus. The greater apparent RNA encapsidation efficiency of vector-length pCSHL relative to sub-genomic spliced pBNL2 RNA might suggest that CAEV sequences downstream of the splice donor contribute to RNA encapsidation.54 The overall poor vector production by the CAEV vector system might in part reflect the absence of an RRE within the vector, resulting in poor accumulation of vector-length CAEV RNA in the cytoplasm.

Visna Virus-Based Gene Transfer Vectors In a detailed study Berkowitz and colleagues described efforts to derive vectors from visna virus.55 Although a considerable effort was put forth into generating vectors, difficulties were encountered in generating a vector system that could transduce cells efficiently. In an initial approach, a two-plasmid system was used to evaluate modifications aimed at improving expression of visna virus protein expression cassettes in 293-based producer cells.55 A series of visna virus genomes were constructed capable of expressing visna viral proteins and a GFP reporter. A representative vector is shown in Figure 4. The panel of vectors was modified in several ways. First, the CMV promoter was fused to the R-U5 sequence of the 5' visna LTR, either joined at the predicted transcription start sites or joined at the TATA box. Second, about 800 bp were deleted from the middle of env, leaving the rev coding region intact. Finally a GFP reporter gene with an internal promoter was placed at the site of the env deletion (Fig. 4). Constructs were evaluated in a transient 293-cell expression system for cytoplasmic Gag mRNA,

140

Lentiviral Vector Systems for Gene Transfer

Fig. 4. Schematic of visna virus vector system. Top, an example of a vector used in two-plasmid vector system. R-U5, sequence domains from visna virus LTR. SD, major virus splice donor site. PGK pro, the murine phosphoglycerate kinase promoter. Middle and bottom, examples of visna virus protein expression cassette and gene transfer vectors used in three-plasmid vector system. MND pro, the myeloid proliferative sarcoma virus promoter.

cell-associated Gag protein levels, and virion-associated Gag protein and virion reverse transcriptase activity. It was found that replacing the visna virus LTR with the immediate early CMV promoter led to a 4-fold increase in cell-associated Gag expression. Virion production from 293 cells was also found to be very good as assessed by virion-associated Gag protein, vector length RNA and reverse transcriptase activity. In fact, reverse transcriptase activity was as good or exceeded reverse transcriptase activity from parallel cultures transfected with a CMVdriven HIV protein expression cassette vector. To assess gene transfer ability, human 293T cells, human lymphoid CEM cells or sheep choroid plexus cells (permissive for visna virus replication) were used as transduction targets for GFP-containing vectors. It was found that while HIV vectors transduced all three cell targets readily, very little transduction was observed by the visna vectors.55 Further visna virus constructs were made to use in a three-plasmid vector system.55 These constructs, similar to the design of other lentiviral three-plasmid systems, separated the viral genes and the gene tranfer vector onto two plasmids (Fig. 4, middle and bottom). The viral gene expression constructs (e.g., VH2) were deleted of cis-acting sequences including LTR sequences and the 3' PPT tract required for replication. Gene transfer vectors contained the CMV promoter fused to the visna TATA box, various amounts of sequences from gag to provide an RNA encapsidation signal, and various amounts of sequences from env to serve as a

EIAV, CAEV and Other Lentiviral Vectors

141

source of the RRE. Constructs were identified that yielded good levels of vector-length cytoplasmic RNA and, by comparison to HIV derived vector controls, good encapsidation of vector length RNA into virions (e.g., vector VV2-PG, Fig. 4). However, transduction efficiency was 100-300 fold less than expected compared to HIV vector production carried out in parallel. The mechanism for the low transduction was investigated.55 Improvements in relative infectivity were not seen when visna vectors were produced in cells permissive for visna infection, suggesting that producing these vectors in human cells was not necessarily detrimental. Similar levels of VSV-G were found in visna and HIV virion preparations, as assessed by Western blot analysis, so inefficient transduction was not due to lack of G envelope incorporation. Using a PCR strategy to detect nearly complete DNA products in transduced cells, it was determined that visna vector transduced cells contained about 30-fold less viral DNA than predicted. Furthermore the visna vector DNA was 10-fold less efficiently integrated than HIV vector DNA. Thus the low transduction efficiency appears to be due to problems occurring at early steps in the transduction process. Once visna vector DNA has integrated, however, expression levels remained stable for at least several weeks.

Bovine Lentiviruses The bovine lentiviruses contain two members: bovine immunodeficiency virus (BIV) and Jembrana disease virus (JDV). Despite it’s name, BIV has been associated with only a mild clinical syndrome in infected taurine cattle (Bos taurus).56 In contrast, JDV causes a severe acute febrile illness in infected Bali cattle (Bos javanicus).57 These cattle are raised mainly in the Jembrana district on the island of Bali in Indonesia.58 The acute disease associated with JDV infection has a short incubation time (5-12 days) and a duration of about 7 days. Infected animals show signs of fever, lymphadenopathy and lymphopenia. The mortality rate of infected animals is significant, about 17%.59 The disease is characterized pathologically as an intense lymphoproliferative disorder in which proliferating infected lymphoblastoid cells are found in lymphoid tissues of most organs, particularly in enlarged lymph nodes and spleen.59,60 Cellular infiltrates are also found in the parenchyma of other organs including the lungs, liver, and kidneys.59 In fatal infections death is attributed to multi-organ failure. In non-fatal infections animals recover about 4-5 weeks post-infection and exhibit no more clinical signs. Recovered animals are resistant to further infection.61

Bovine Lentiviral Vectors While gene transfer vectors have yet to be described for BIV, vectors have been generated from molecular clones of Jembrana disease virus (JDV).62 The strategy was to construct a three-plasmid vector system, similar to approaches described above. The packaging construct, pC4.gpe (Fig. 5), contains all of the viral genes. Several modifications were made to disable the construct for replication. The env start codon was mutated to a stop codon. To disable encapsidation of the packaging construct RNA into virions, a 20-nucleotide sequence was deleted from the 5' leader region between the major splice donor and the start of gag. However, it was not determined to what extent this mutation inhibited encapsidation. To prevent the packaging construct from replicating, the 5' LTR sequences and adjacent tRNA primer binding site were replaced with a CMV promoter. Indeed PCR analysis using primers to tat failed to detect packaging construct DNA in cells transduced with JDV vectors.62 Gene transfer vectors contained an internal CMV promoter driving a GFP-IRES-neo cassette (e.g., pJLCGIN, Fig. 5). The JDV LTR is apparently quite robust in human 293 cells and was used in producer cells to drive synthesis of vector length RNA. A portion of gag was retained in the gene transfer vectors to ensure efficient encapsidation. The sequence at the gag start codon was mutated by causing a frameshift at that site.

142

Lentiviral Vector Systems for Gene Transfer

Fig. 5. Schematic of JDV vector system. SD, major splice donor site. RRE, presumed rev responsive element from the 3' region of JDV env.

VSV-G pseudotyped JDV vectors were produced having titers in the range of 0.4-1.2 x 106 G418 colony forming units/ml using 293 cells to titer virus.62 Transduction of GFP was demonstrated for several cell lines including human 293 and HeLa cells, monkey COS7 cells, murine B16 cells and primary fetal bovine lung cells. These cell types were transduced with efficiencies ranging from 28-78%, based upon GFP fluorescence, at a relatively low MOI of 5. The vectors showed a slight preference for fetal bovine lung cells among the cell types tested. Transduction of aphidicolin treated HeLa cells was as efficient as untreated cells indicating that JDV vectors can transduce non-dividing human cells.

Future Prospects In summary, significant progress has been made in deriving gene transfer vector systems from EIAV, CAEV, visna and JDV. Due to the unpredictability of deriving simple vector systems from complicated viruses, variability in the success of constructing efficient vector systems has been observed. Relatively efficient vector systems have been made from EIAV and JDV.37,38,62 It is likely that future work will focus on improving the efficiency and safety of these systems. In contrast, gene transfer titers of vectors derived from CAEV and visna virus have been disappointing.54,55 It is likely that future work will address potential reasons for this in more detail. A recent study has shown that VSV-G pseudotyping of otherwise wild-type CAEV allowed an efficient single round of infection of human cells.63 It is notable that in these studies human 293 cells were used for the efficient production of VSV-G pseudotyped virus. Thus, there does not appear to be any obvious obstacle in developing efficient vector systems from CAEV. Preliminary reports have appeared describing VSV-G packaging cell lines capable of stable production of EIAV vectors.64,65 Stable vector producing cell lines will be important for quality control analyses of cells and vector preparations and for scaling up vector production for human clinical trials. The eventual use of these vectors in clinical trials will depend upon steady improvement in gene transfer efficiency and the safety validation of these novel vector systems. Note added in proof: A recent study has described the construction of gene transfer vectors based on bovine immunodeficiency virus.66

EIAV, CAEV and Other Lentiviral Vectors

143

References 1. Naldini L, Verma IM. Lentiviral vectors. Adv Virus Res 2000; 55:599-609. 2. Schiltz RL, Shih DS, Rasty S et al. Equine infectious anemia virus gene expression: characterization of the RNA splicing pattern and the protein products encoded by open reading frames S1 and S2. J Virol 1992; 66(6):3455-3465. 3. Chadwick BJ, Coelen RJ, Wilcox GE et al. Nucleotide sequence analysis of Jembrana disease virus: A bovine lentivirus associated with an acute disease syndrome. J Gen Virol 1995; 76(Pt 7):1637-1650. 4. Maury W. Regulation of equine infectious anemia virus expression. J Biomed Sci 1998; 5(1):11-23. 5. Li F, Leroux C, Craigo JK et al. The S2 gene of equine infectious anemia virus is a highly conserved determinant of viral replication and virulence properties in experimentally infected ponies. J Virol 2000; 74(1):573-579. 6. Li F, Puffer BA, Montelaro RC. The S2 gene of equine infectious anemia virus is dispensable for viral replication in vitro. J Virol 1998; 72(10):8344-8348. 7. Elder JH, Lerner DL, Hasselkus-Light CS et al. Distinct subsets of retroviruses encode dUTPase. J Virol 1992; 66(3):1791-1794. 8. Abergel C, Robertson DL, Claverie JM. “Hidden” dUTPase sequence in human immunodeficiency virus type 1 gp120. J Virol 1999; 73(1):751-753. 9. Bergman AC, Bjornberg O, Nord J et al. The protein p30, encoded at the gag-pro junction of mouse mammary tumor virus, is a dUTPase fused with a nucleocapsid protein. Virology 1994; 204(1):420-424. 10. Koppe B, Menendez-Arias L, Oroszlan S. Expression and purification of the mouse mammary tumor virus gag-pro transframe protein p30 and characterization of its dUTPase activity. J Virol 1994; 68(4):2313-2319. 11. Caradonna SJ, Cheng YC. Induction of uracil-DNA glycosylase and dUTP nucleotidohydrolase activity in herpes simplex virus-infected human cells. J Biol Chem 1981; 256(19):9834-9837. 12. Broyles SS. Vaccinia virus encodes a functional dUTPase. Virology 1993; 195(2):863-865. 13. Baldo AM, McClure MA. Evolution and horizontal transfer of dUTPase-encoding genes in viruses and their hosts. J Virol 1999; 73(9):7710-7721. 14. Vassylyev DG, Morikawa K. Precluding uracil from DNA. Structure 1996; 4(12):1381-1385. 15. Richards RG, Sowers LC, Laszlo J et al. The occurrence and consequences of deoxyuridine in DNA. Adv Enzyme Regul 1984; 22:157-185. 16. Hokari S, Horikawa S, Tsukada K et al. Expression of deoxyuridine triphosphatase during liver regeneration in rat. Biochem Mol Biol Int 1995; 37(3):583-590. 17. Pang JH, Chen KY. Global change of gene expression at late G1/S boundary may occur in human IMR-90 diploid fibroblasts during senescence. J Cell Physiol 1994; 160(3):531-538. 18. Spector R, Boose B. Development and regional distribution of deoxyuridine 5'-triphosphatase in rabbit brain. J Neurochem 1983; 41(4):1192-1195. 19. Strahler JR, Zhu XX, Hora N et al. Maturation stage and proliferation-dependent expression of dUTPase in human T cells. Proc Natl Acad Sci USA 1993; 90(11):4991-4995. 20. Sugita Y, Yamauchi H, Omine M et al. Elevated deoxyuridine triphosphate nucleotidohydrolase (dUTPase) activity in the cobalamin-deficient megaloblastic bone marrow cells. Int J Hematol 1996; 64(3-4):203-212. 21. Vilpo JA. Mitogen induction of deoxyuridine triphosphatase activity in human T and B lymphocytes. Med Biol 1983; 61(1):54-58. 22. Wagaman PC, Hasselkus-Light CS, Henson M et al. Molecular cloning and characterization of deoxyuridine triphosphatase from feline immunodeficiency virus (FIV). Virology 1993; 196(2):451-457. 23. Threadgill DS, Steagall WK, Flaherty MT et al. Characterization of equine infectious anemia virus dUTPase: Growth properties of a dUTPase-deficient mutant. J Virol 1993; 67(5):2592-2600. 24. Lichtenstein DL, Rushlow KE, Cook RF et al. Replication in vitro and in vivo of an equine infectious anemia virus mutant deficient in dUTPase activity. J Virol 1995; 69(5):2881-2888. 25. Turelli P, Petursson G, Guiguen F et al. Replication properties of dUTPase-deficient mutants of caprine and ovine lentiviruses. J Virol 1996; 70(2):1213-1217.

144

Lentiviral Vector Systems for Gene Transfer

26. Petursson G, Turelli P, Matthiasdottir S et al. Visna virus dUTPase is dispensable for neuropathogenicity. J Virol 1998; 72(2):1657-1661. 27. Turelli P, Guiguen F, Mornex JF et al. dUTPase-minus caprine arthritis-encephalitis virus is attenuated for pathogenesis and accumulates G-to-A substitutions. J Virol 1997; 71(6):4522-4530. 28. Lerner DL, Wagaman PC, Phillips TR et al. Increased mutation frequency of feline immunodeficiency virus lacking functional deoxyuridine-triphosphatase. Proc Natl Acad Sci USA 1995; 92(16):7480-7484. 29. Phillips TR, Prospero-Garcia O, Wheeler DW et al. Neurologic dysfunctions caused by a molecular clone of feline immunodeficiency virus, FIV-PPR. J Neurovirol 1996; 2(6):388-396. 30. Steagall WK, Robek MD, Perry ST et al. Incorporation of uracil into viral DNA correlates with reduced replication of EIAV in macrophages. Virology 1995; 210(2):302-313. 31. Campbell RS, Robinson WF. The comparative pathology of the lentiviruses. J Comp Pathol 1998; 119(4):333-395. 32. Issel CJ, Coggins L. Equine infectious anemia: current knowledge. J Am Vet Med Assoc 1979; 174(7):727-733. 33. Montelaro RC, Cole KS, Hammond SA. Maturation of immune responses to lentivirus infection: Implications for AIDS vaccine development. AIDS Res Hum Retroviruses 1998; 14 Suppl 3:S255-S259. 34. Payne SL, Qi X-M, Shao H et al. Disease induction by virus derived from molecular clones of equine infectious anemia virus. J Virol 1998; 72:483-487. 35. Cook RF, Leroux C, Cook SJ et al. Development and characterization of an in vivo pathogenic molecular clone of equine infectious anemia virus. J Virol 1998; 72:1383-1393. 36. Belshan M, Baccam P, Oaks JL et al. Genetic and biological variation in equine infectious anemia virus Rev correlates with variable stages of clinical disease in an experimentally infected pony. Virology 2001; 279(1):185-200. 37. Olsen JC. Gene transfer vectors derived from equine infectious anemia virus. Gene Therapy 1998; 5:1481-1487. 38. Mitrophanous K, Yoon S, Rohll J et al. Stable gene transfer to the nervous system using a nonprimate lentiviral vector. Gene Ther 1999; 6(11):1808-18. 39. Maury W. Monocyte maturation controls expression of equine infectious anemia virus. J Virol 1994; 68(10):6270-6279. 40. Sellon DC, Walker KM, Russell KE et al. Equine infectious anemia virus replication is upregulated during differentiation of blood monocytes from acutely infected horses. J Virol 1996; 70(1):590-594. 41. Maury W, Oaks JL, Bradley S. Equine endothelial cells support productive infection of equine infectious anemia virus. J Virol 1998; 72(11):9291-9297. 42. Bowles NE, Eisensmith RC, Mohuiddin R et al. A simple and efficient method for the concentration and purification of recombinant retrovirus for increased hepatocyte transduction in vivo. Hum Gene Ther 1996; 7(14):1735-1742. 43. Johnson LG, Mewshaw JP, Ni H et al. Effect of host modification and age on airway epithelial gene transfer mediated by a murine leukemia virus-derived vector. J Virol 1998; 72:8861-8872. 44. Yamada K, Olsen JC, Patel M et al. Functional correction of Fanconi anemia group C (FANCC) hematopoietic cells by the use of a novel lentiviral vector. Mol Therapy 2001; 3(4):485-490. 45. Gush KA, Fu KL, Grompe M et al. Phenotypic correction of Fanconi anemia group C knockout mice. Blood 2000; 95(2):700-704. 46. Liu JM, Kim S, Read EJ et al. Engraftment of hematopoietic progenitor cells transduced with the Fanconi anemia group C gene (FANCC). Hum Gene Ther 1999; 10(14):2337-2346. 47. Pepin M, Vitu C, Russo P et al. Maedi-visna virus infection in sheep: A review. Vet Res 1998; 29(3-4):341-367. 48. Georgsson G. Neuropathologic aspects of lentiviral infections. Ann NY Acad Sci 1994; 724:50-67. 49. Carey N, Dalziel RG. The biology of maedi-visna virus—An overview. Br Vet J 1993; 149(5):437-454. 50. Clements JE, Zink MC, Narayan O et al. Lentivirus infection of macrophages. Immunol Ser 1994; 60:589-600.

EIAV, CAEV and Other Lentiviral Vectors

145

51. Zink MC, Narayan O, Kennedy PG et al. Pathogenesis of visna/maedi and caprine arthritis-encephalitis: New leads on the mechanism of restricted virus replication and persistent inflammation. Vet Immunol Immunopathol 1987; 15(1-2):167-80. 52. Gendelman HE, Narayan O, Kennedy-Stoskopf S et al. Tropism of sheep lentiviruses for monocytes: Susceptibility to infection and virus gene expression increase during maturation of monocytes to macrophages. J Virol 1986; 58(1):67-74. 53. Narayan O, Kennedy-Stoskopf S, Sheffer D et al. Activation of caprine arthritis-encephalitis virus expression during maturation of monocytes to macrophages. Infect Immun 1983; 41(1):67-73. 54. Mselli-Lakhal L, Favier C, Da Silva Teixeira MF et al. Defective RNA packaging is responsible for low transduction efficiency of CAEV-based vectors. Arch Virol 1998; 143(4):681-695. 55. Berkowitz RD, Ilves H, Plavec I et al. Gene transfer systems derived from Visna virus: Analysis of virus production and infectivity. Virology 2001; 279(1):116-129. 56. Straub OC, Levy D. Bovine immunodeficiency virus and analogies with human immunodeficiency virus. Leukemia 1999; 13 Suppl 1:S106-9. 57. Wilcox GE, Chadwick BJ, Kertayadnya G. Recent advances in the understanding of Jembrana disease. Vet Microbiol 1995; 46(1-3):249-255. 58. Hartaningsih N, Wilcox GE, Dharma DM et al. Distribution of Jembrana disease in cattle in Indonesia. Vet Microbiol 1993; 38(1-2):23-29. 59. Dharma DM, Budiantono A, Campbell RS et al. Studies on experimental Jembrana disease in Bali cattle. III. Pathology . J Comp Pathol 1991; 105(4):397-414. 60. Chadwick BJ, Desport M, Brownlie J et al. Detection of Jembrana disease virus in spleen, lymph nodes, bone marrow and other tissues by in situ hybridization of paraffin-embedded sections. J Gen Virol 1998 ;79(Pt 1):101-106. 61. Soeharsono S, Hartaningsih N, Soetrisno M et al. Studies of experimental Jembrana disease in Bali cattle. I. Transmission and persistence of the infectious agent in ruminants and pigs, and resistance of recovered cattle to re-infection. J Comp Pathol 1990; 103(1):49-59. 62. Metharom P, Takyar S, Xia HH et al. Novel bovine lentiviral vectors based on Jembrana disease virus. J Gene Med 2000; 2(3):176-185. 63. Mselli-Lakhal L, Favier C, Leung K et al. Lack of functional receptors is the only barrier that prevents caprine arthritis-encephalitis virus from infecting human cells. J Virol 2000; 74(18):8343-8348. 64. Olsen JC, Patel M, Rohll JB et al. An inducible first generation stable packaging cell line for equine lentiviral vectors. Mol Therapy 2000; 1:S315 (Abstract 882). 65. Mitrophanous K, Ellard F, Slingsby J et al. Generation of an equine infectious anemia virus (EIAV)based lentiviral producer cell line. Mol Therapy 2000; 1:S314 (Abstract 881). 66. Berkowitz R, Ilves H, Lin WY et al. Construction and molecular analysis of gene transfer systems derived from bovine immunodeficiency virus. J Virol 2001; 75:3371-82.

CHAPTER 8

Safety Considerations in Vector Development John C. Kappes and Xiaoyun Wu

Abstract

T

he inadvertent production of replication competent retrovirus (RCR) constitutes the principal safety concern for the use of lentiviral vectors in human clinical protocols. Because of limitations in animal models to evaluate lentiviral vectors for their potential to recombine and induce disease, the vector design itself should ensure against the emergence of RCR in vivo. Issues related to RCR generation and one approach to dealing with this problem are discussed in this chapter. To assess the risk of generating RCR, a highly sensitive biological assay was developed to specifically detect vector recombination in transduced cells. Analysis of lentiviral vector stocks has shown that recombination occurs during reverse transcription in primary target cells. Rejoining of viral protein-coding sequences of the packaging construct and cis-acting sequences of the vector was demonstrated to generate env-minus recombinants (LTR-gag-pol-LTR). Mobilization of recombinant lentiviral genomes was also demonstrated but was dependent on pseudotyping of the vector core with an exogenous envelope protein. 5' sequence analysis has demonstrated that recombinants consist of U3, R, U5, and the ψ packaging signal joined with an open gag coding region. Analysis of the 3’ end has mapped the point of vector recombination to the poly(A) tract of the packaging construct’s mRNA. The state-of-the-art third generation packaging construct and SIN vector also have been shown to generate env-minus proviral recombinants capable of mobilizing retroviral DNA when pseudotyped with an exogenous envelope protein. A new class of HIV-based vector (trans-vector) was recently developed that splits the gag-pol component of the packaging construct into two parts: one that expresses Gag/Gag-Pro and another that expresses Pol (RT and IN) fused with Vpr. Unlike other lentiviral vectors, the trans-vector has not been shown to form recombinants capable of DNA mobilization. These results indicate the trans-vector design prevents the generation of env-minus recombinant lentivirus containing a functional gagpol structure (LTR-gag-pol-LTR), which is absolutely required for retroviral DNA mobilization and the emergence of RCR. Quality assurance based on monitoring for RCR may have limitations as a predictor of safety in vivo, especially in the long term. The demonstration of lentivirus infection via alternative entry mechanisms supports this notion. Therefore, the approach of monitoring trans-vector stocks for env-minus recombinant virus in vitro as a surrogate marker for the possible emergence of RCR in vivo should represent a significant advancement in vector safety quality assurance.

Introduction Retroviruses are small in the genetic sense, having genomes that are about 10 kb with a relatively small complement of proteins. With such limited genetic information, they depend Lentiviral Vector Systems for Gene Transfer, edited by Gary L. Buchschacher, Jr. ©2003 Eurekah.com.

148

Lentiviral Vector Systems for Gene Transfer

heavily on their host for essential replication functions, all of which are intimately linked to the host cell. In its proviral form the retroviral genome mimics that of the cell, and gene expression involves extensive interactions with the host cell machinery. Retroviruses are unique among infectious agents in the way they interact with the host cell and in the consequences of this interaction. Retroviruses regularly integrate their genetic information into the host genome; they can acquire genes into their genome and they can infect the germ line of the host organism. Integrated proviral DNA is permanent and enables coevolution in both the short and long term. Retroviruses display a remarkable variety of pathogenic interactions with their host organism. Retroviral diseases may be acute, episodic, or chronic, appearing soon after infection or after a long asymptomatic period. The ability of retroviruses to stably integrate their genomes into chromosomes of target cells provides a strong incentive for the development of retroviral-based vectors for gene therapy. Over the last two decades, retroviral vectors, derived from oncoretroviruses such as Moloney murine leukemia virus (MoMLV), have been used widely for gene transfer, both experimentally and in clinical protocols.1,2 In clinical protocols, MoMLV has been associated with inefficient gene transfer. This is due, in part, to the inability of oncoretrovirus-based vectors to infect nondividing or slowly dividing cells.3,4 Although of limited clinical benefit, studies on MoMLV vectors have important implications for the development and safety design of current retroviralbased vectors, including lentiviral vectors. The principal consideration regarding safety is the possibility that administration of the vector may lead to the emergence of replication-competent retrovirus. Safety advancement in retroviral vector design includes the development of packaging systems that provide all of the retroviral proteins in trans to a replication-defective gene transfer vector. The packaging component expresses the viral structural proteins required for assembly of the virus particle and is stripped of cis-acting sequences necessary for the transfer of the viral genome.5 For a review see Chapters 1 and 4 and reference.6 These cis-acting sequences are retained within the gene transfer vector where they facilitate its encapsidation, reverse transcription, and integration. This design does not completely exclude encapsidation of other viral and cellular RNA strands, including that derived from the viral packaging construct. Copackaging of helper and vector genomes allows genetic recombination to occur during reverse transcription and thus rejoining of viral protein-coding sequences and cis-acting sequences of the vector. If recombinants are formed that contain reproductive functions, it is possible they may replicate and spread to infect other cells. Thus, the inadvertent production of replication competent retrovirus (RCR) constitutes the principal safety concern for the use of retroviral vectors in human clinical protocols.7-9 Several studies have reported the generation of RCR from retroviral packaging cell lines as a result of genetic recombination between the vector and helper sequences. The most notable example comes from a study of retrovirus-mediated gene transfer in rhesus monkeys. After autologous transplantation of bone marrow stem cells that had been transduced ex vivo with MoMLV vector, three of ten treated animals developed T-cell lymphomas10 and RCR was isolated from two of the three animals. In one animal the RCR resulted from recombination between the vector and packaging sequences introduced into the producer cell line used to generate the vector.11,12 The genome of a second RCR was identified to have arisen by recombination involving the genome of a vector/helper recombinant with that of an endogenous murine retrovirus present in the producer cell line.12,13 Although the packaging and vector components were designed to minimize recombination events, the emergence of RCR in producer cells was common in this early type of packaging cell line.14,15 To improve vector safety, third generation retroviral packaging cell lines were designed by separating the gag/gag-pol, env and vector functions onto different genetic fragments. The generation of replication competent retrovirus by the third generation GP+envAM12 packaging cell line has been documented, showing that even packaging cell lines with split viral

Safety Considerations in Vector Development

149

protein-coding regions can recombine to form RCR.16,17 In this case, the RCR was produced by recombination events at sites of partial homology between sequences in the vector, the packaging construct and an endogenous retroviral element in the producer cell.18 Vector producing cells that eliminate homology with endogenous sequences (such as HEK-293) and minimize the amount of overlapping sequence shared between the genomes of the vector and packaging construct have not been found to generate RCR.19 While the generation of RCR has been reduced markedly by improvements in vector design and vector producer cells, these results emphasize the potential for retroviral-mediated recombination. Because retroviral vectors can be permanently integrated in the host cell’s chromosomes and passed to daughter cells, integration itself raises safety concerns at two different levels: insertional mutagenesis and germ line transduction. All viruses and vectors that randomly insert their genetic information into the genome of target cells can cause insertional mutagenesis. Insertion of proviral DNA can disrupt normal gene function, cause the inactivation of tumor suppressor genes or the activation of oncogenes. For reviews see references.6,20 This risk is directly proportional to the number of insertion/integration events. Although the activation of a protooncogene by retroviral insertion is not itself usually sufficient to convert a normal cell into a tumor cell it can be a rate-limiting step. In the case of oncoretroviruses, insertional mutagenesis may be largely due to relatively high numbers of infected cell/insertion events. Thus, the probability of insertion near a protooncogene is relatively high. By analogy, if genetic recombination was to produce virus that could replicate and spread, the likelihood of inducing insertional mutagenesis would increase. The analysis of various tissues from a rhesus monkey that developed T-cell lymphoma from RCR revealed a common recombinant proviral insertion site, suggesting that chronic productive retroviral infection leads to insertional mutagenesis of critical growth control genes.12 There is no evidence to date that primary vector-transduced cells have had adverse effects in animals or humans. Provirus formation in a germ cell provides the means for retroviruses to colonize the germ line of the host. Extensive analysis in animal models and human trials has not demonstrated transmission of retroviral vectors into the germ line. Based on studies with retroviral vectors it is clear that RCR can be generated from retroviral packaging cell lines through genetic recombination between the gene transfer vector and transpackaging genetic elements. Moreover, endogenous retroviral sequences may contribute to the emergence of RCR. Therefore, our primary consideration for safety, and the principal focus of discussion in this Chapter, is a lentiviral vector design as an example of an approach that might minimize the risk that genetic recombination may lead to the emergence of RCR. Discussion of other safety considerations involving the use of lentiviral vectors may be found in Chapters 4 (HIV-1 Vectors), 9 (Prospects for Gene Therapy) and 10 (Ethical Considerations).

Current Safety Design Lentiviruses represent a genus of the retroviridae. Unlike the oncoretroviruses, lentiviruses can infect and replicate in non-mitotic cells because of karyophilic properties of the nucleoprotein preintegration complex.21,22 This has generated considerable interest in lentiviruses as vectors for gene therapy. See Chapters 4, 5, 6 and 7 for reviews on the different lentiviral-based vectors. Within the last four years tremendous progress has been made in the development of lentiviral-based vectors that are capable of efficient gene transfer and that incorporate important safety design features.23-27 Similar to oncoretroviral vectors, the principal safety concern for lentiviral vectors is the generation of RCR. To date, neither second nor third generation lentiviral vectors have been found to generate RCR.28,29 However, what we learned from MoMLV vectors indicates that genetic recombination may occur and contribute to the emergence of RCR. Unlike MoMLV, RCR derived from lenti-vectors would likely retain the ability to infect nondividing cells, which would raise additional safety concerns. Clearly, the lentiviral vectors will be held to rigorous standards to ensure their safety prior to use in the clinic. Because of the

150

Lentiviral Vector Systems for Gene Transfer

Fig. 1. Overview of method for monitoring the generation of env-minus recombinant lentivirus. (A) Through nonspecific encapsidation, other RNA molecules, including those derived from the packaging construct, may be co-packaged with the gene transfer vector RNA. Preparations of vector stocks are used to transduce cultures of 293T cells (MOI=2). (B) Genetic recombination between RNAs of the packaging construct and the vector may generate env-minus recombinant lentiviral genomes. If stably integrated into the chromosomes of transduced 293T cells, such recombinant proviral genomes may express viral proteins and produce env-minus progeny lentiviral particles that package the recombinant’s RNA genome. To detect these recombinants, the transduced 293T cells are cotransfected with a VSV-G expression plasmid to pseudotype progeny virions and a tat expression plasmid to up-regulate expression of recombinant proviral genomes. (C) Three days later, the culture supernatants are concentrated by ultracentrifugation. Progeny virions with encapsidated recombinant genomes are depicted. (D) The pseudotyped virions are then used to infect HeLa-puro cells and 30 hours later the cell monolayers are trypsinized and replated in medium containing puromycin. Pseudotyped particles confer resistance to puromycin if they contain a recombinant lentiviral genome that is reverse transcribed, integrated, and expresses Tat protein.

Safety Considerations in Vector Development

151

limitations of animal models to evaluate lentiviral vectors for their potential to induce disease, the vector design itself should ensure the greatest level of safety that is achievable. Principally, lentiviral vector systems are comprised of three separate parts: the packaging, gene transfer vector, and envelope (env) components. The most advanced third generation packaging construct includes many safety design features. For the HIV-1 based vector, all of the accessory genes (vif, vpr, vpu, nef) have been deleted. In addition, the two key regulatory genes, tat and rev, have been either deleted or separated from the packaging construct.30-33 See Chapter 4 for a review of HIV-1 based vectors. With only the gag and pol genes remaining, it is not possible for genetic recombination to generate a virus with pathogenic features like that of the parental virus. Importantly, the third generation packaging construct has not diminished vector titer. Safeguards have also been built into the gene transfer vector. The vector contains all of the cis-acting elements necessary for packaging, reverse transcription, and integration, while all viral open reading frames have been eliminated. The self-inactivating design (SIN) feature deletes the viral transcriptional promoter and enhancer elements. Self-inactivation relies on a deletion made in the U3 region of the 3' LTR of the DNA that produces the vector RNA. By reverse transcription, the 3' U3 deletion is duplicated in the 5' LTR. Consequently, transcription of full-length vector RNA is markedly reduced in cells transduced with a SIN vector.34,35 This further minimizes the risk for generating RCR and reduces the chance that cellular coding sequences located adjacent to the integrated vector will be aberrantly expressed due to promoter activity of the 3' LTR. With the deletion of the viral transcriptional elements from the vector, synthesis of vector RNA will largely depend on the site of vector integration and surrounding cellular transcription elements.34 Several in vitro assays have been used to evaluate the safety of HIV-based vectors, including marker rescue, tat-transfer and gag-transfer assays. These assays have not provided any evidence of either RCR or genetic recombination.28,29 To fully assess the safety risks, we thought it was necessary to understand whether recombination occurs between the genetic components of the lentiviral vector system, and if so, to characterize the nature of the recombinants at the molecular level. Then it would be feasible to design and evaluate new approaches that overcome the safety risks associated with genetic recombination. A highly sensitive biological assay was developed to specifically detect vector recombination in transduced cells, independently of the generation of RCR.36 This assay is based on a cell line containing a stably integrated copy of the puromycin resistance gene introduced by transduction with a lentiviral vector. Since the puromycin gene was placed under control of the HIV-1 LTR, it is inducible by Tat expression. An overview of this assay is depicted in Figure 1. Supernatants from cultures of 293T cells infected with a second generation lentiviral vector (107 infectious units [IU]) and then cotransfected with VSV-G (pMD.G) and tat (ptat) expression plasmids were analyzed for recombinant virus by incubation with the puromycin inducible cell line (Fig. 1C). The detection of puromycin resistant cell colonies suggested the formation of vector/helper recombinants (Fig. 2). The non-nucleoside reverse transcriptase inhibitor, Nevirapine, completely blocked colony formation, indicating that the formation of lentiviral vector recombinants was dependent on the function of the HIV-1 reverse transcriptase. The generation of resistant colonies was also completely dependent on pseudotyping with the VSV-G protein, demonstrating the recombinant virus was defective in env. Transfection of the VSV-G expression plasmid for pseudotyping virions generated from recombinant lentivirus is unique to this assay approach and importantly, it allows for the monitoring of env-minus recombinant lentivirus instead of RCR. The formation of recombinant provirus was confirmed by genetic sequence analysis.36 The 5’ sequence consisted of U3, R, U5, and the ψ packaging signal (derived from the vector) joined with the gag open reading frame (derived from the packaging construct). Analysis of 3’ sequences also revealed a physical linkage between the packaging construct and the gene transfer vector. Interestingly, the point of vector recombination occurred within the poly(A) tract of

152

Lentiviral Vector Systems for Gene Transfer

Fig. 2. Detection of env-minus recombinant lentivirus. Supernatants from cultures of 293T cells transduced with 107 infectious units (IU) of the lentiviral vector were used to infect cultures of HeLa-puro cells. One set was infected two hours after adding Nevirapine (5 uM) to the culture medium. The control culture contained uninfected HeLa-puro cells. After selection in medium containing puromycin, the cell monolayers were visualized by staining with crystal violet. The formation of resistant colonies was documented by microscopy.

the packaging construct’s mRNA (Fig. 3). This finding confirmed that the env-minus recombinant lentivirus was generated during reverse transcription in primary transduced cells. Such non-homologous genetic recombination suggests that a similar mechanism could occur in the more advanced third generation vectors. This assertion is strengthened by earlier studies that demonstrated avian retroviruses capture oncogenes through non-homologous recombination within the poly(A) tract.37,38 Using a modified version (Tat-independent) of the above DNA mobilization assay, preparations of third generation and SIN vectors were analyzed and found to generate env-minus proviral recombinants capable of mobilizing retroviral DNA when pseudotyped with an exogenous envelope protein (Fig. 4). Taken together, these data indicate the generation of env-minus recombinant provirus, expression of the recombinant’s genes, assembly of lentivirus-like particles, and encapsidation and mobilization of viral nucleic acids to new host cells when an exogenous Env is provided in trans. These findings underscore the significance for understanding lentiviral vector genetic recombination and reemphasize the importance for thorough examination of possible safety risks while advancing lentiviral vector toward human trials. A safety design was recently introduced that may largely overcome the risks that stem from genetic recombination. Based on the ability to rescue the infectivity of RT-IN deleted HIV-1 by providing the RT and IN proteins in trans,39-42 a new class of HIV-based vectors was designed that splits the gag-pol component of the packaging construct into two separate parts: one that expresses Gag/Gag-Pro and another that expresses Pol (RT and IN) fused with Vpr (Vpr-RT-IN) (Fig. 5). Removing RT and IN from the other components of the packaging construct disarms the retroviral replication machinery conserved within the gag-pol structure. This vector (termed trans-vector) was evaluated in side-by-side experiments with third generation and SIN vectors for its ability to recombine and mobilize DNA. While the third generation and SIN vectors transferred puromycin resistance to naive cells, the trans-vector did not (Fig. 5). This suggests that the trans-vector design prevents the generation of recombinant

Safety Considerations in Vector Development

153

Fig. 3. Genetic analysis of env-minus recombinant lentivirus. High molecular weight DNA prepared from puromycin resistant cells was subjected to PCR using primer pairs specific for sequences of the vector and packaging construct. (A) Sequence analysis of the 5’ end of the recombinant genome. Using a sense primer specific for the U3 region of the 5’ LTR of the gene transfer vector and an antisense primer specific for 3’ gag sequences of the packaging construct, a PCR fragment of approximately 2000 base pairs was amplified, cloned and sequenced. The nucleotide sequence at the junction of the vector and gag gene is illustrated. Of 10 clones analyzed, the sequence was identical. (B) Sequence analysis of the 3’ end of the recombinant genome. Using a sense primer specific for the first tat exon of the packaging construct and an antisense primer specific for the 3’ U3 region of the vector, a DNA fragment of approximately 2500 bp was derived for sequencing. In all of the clones that were analyzed, the 3’ end of the vector was found to be joined with the poly(A) tract of the packaging construct. The nucleotide sequence between the U3 region of the vector and the poly(A) tract of the packaging construct is illustrated. The genotype of four clones among nine that were analyzed is depicted. (C) Genetic recombination during reverse transcription. The diagram illustrates how the recombinants (depicted in B) were likely generated during synthesis of the negative-strand DNA.

154

Lentiviral Vector Systems for Gene Transfer

Fig. 4. Analysis of state-of-the-art lentiviral vector systems for the generation of env-minus recombinant lentivirus. Stocks of lentiviral vectors were prepared using a third generation packaging construct (A), a third generation packaging construct in combination with a SIN vector (B), and the trans-vector (C), respectively. 108 infectious units of A and B and 109 infectious units of C were used to infect cultures of 293T cells containing the puromycin resistance gene as depicted in Figure 1. Three days later the culture supernatants were collected, concentrated by ultracentrifugation and used to infect HeLa-tat cells. After selection in medium containing puromycin the resistant cell colonies were visualized by staining with crystal violet.

lentivirus containing a functional gag-pol structure (LTR-gag-pol-LTR), which is absolutely required for retroviral DNA mobilization and the emergence of RCR. Importantly, the trans-vector appears to retain all of the properties for efficient transduction of non-dividing cells. Terminally differentiated macrophages and CD34+ human bone marrow cells are efficiently transduced.36 Bone marrow-derived mouse stem cells transduced ex vivo with the trans-vector are able to mediate long term reconstitution of lethally irradiated mice and maintain normal hematopoiesis in vivo. In fully reconstituted animals, trans-gene (GFP) expression was observed for 20 weeks and GFP positive cells could be transferred longterm to secondary transplant recipients (unpublished), similar to that reported for the lentiviral vector.43 Other studies in mice indicate that the trans-vector can mediate efficient and longterm trans-gene expression in the retinal pigment epithelium (unpublished results). Since the trans-vector alters the normal process of virion assembly, its stability and titer have been carefully examined and found to be three to five fold reduced compared to that of the lentiviral vector.36 It is possible that this difference in titer may be overcome using a stable packaging cell line for vector production. In summary, the trans-RT-IN/Gag-Pro design of the trans-vector represents an important advancement in biosafety at two levels: it controls the regeneration of lentiviral recombinants containing functional gag-pol and it enables in vitro testing of trans-vector stocks to assess the risk of generating RCR in vivo. The trans-vector design appears to address important safety considerations that will likely help advance lentiviral vectors toward the clinic.

Advancing Lentiviral Vectors Toward the Clinic Great progress has been made in the design of lentiviral vectors with respect to both biosafety and performance in vivo.24,25,27,36,43-46 This appears to be especially true for targets where long term expression of the trans-gene is desired such as the central nervous system, hematopoietic stem cells and the eye. Based on this premise, it is likely that a lentivirus will soon be proposed for clinical evaluation. Because of the limitations of animal models to evaluate lentiviral vector designs for their potential to induce disease, safety will ultimately be determined in human hosts.

Safety Considerations in Vector Development

155

Fig. 5. Genetic components of the trans-lentiviral packaging system. The trans-lenti packaging construct is illustrated as pCMV-gag-pro. The pCMV-vpr-RT-IN construct encodes the Vpr-RT-IN fusion protein, which is packaged into the Gag/Gag-Pro particles, providing the reverse transcriptase and integrase function. Proteolytic processing by the viral protease liberates mature and enzymatically active RT (p51/ p66) and IN proteins.41

Quality assurance methods based on in vitro monitoring for RCR seem to be of limited value as a predictor of safety in vivo, especially in the long term. The detection of env-minus recombinant lentivirus among 107 infectious vector particles raises the possibility that in the large scale production required for gene therapy applications, the probability of generating recombinants containing functional env may be increased. The ability of HIV-1 cores to acquire and utilize cellular membrane proteins such as CD4/CCR5 or CXC4 for infection47-49 suggests the possibility that env-minus recombinant virus could infect dividing and nondividing cells through alternative mechanisms (Env independent). It has long since been thought that infection with xenotropic endogenous retroviruses may have occurred via alternative viral or cellular receptors provided in trans.50 This notion was recently supported by a report showing envminus HIV-1 can infect CD4-minus and CD4-positive cells through a pathway independent of the viral Env glycoprotein.51 Therefore, the ability to directly monitor env-minus recombinant virus in vitro as a surrogate marker for the possible emergence of RCR in vivo may represent a significant advancement in quality assurance. The combination of the trans-vector design with an in vitro assay that monitors for recombinants devoid of functional gag-pol as a means to quality assure lentiviral vector stocks is illustrated in Figure 6. The failure to detect env-minus recombinant lentivirus in the trans-vector system may suggest that either the assay is not sensitive enough or that recombinants are not formed at the titers tested. By comparison with lenti-vectors, the assay is at least three orders of magnitude beyond the limits of sensitivity. Since the trans-vector has been show to generate RT-IN minus recombinants,36 it may also be possible to regenerate recombinants containing functional gagpol. However, this would require three restricted steps: first, single virions (vector particles) must co-package three different mRNAs, two of which do not contain the ψ packaging signal; second, the three separate genetic elements (mRNAs) must recombine; and third, recombination must occur in a manner that restores a functional LTR-gag/gag-pol-LTR structure. The failure to detect recombinants with functional gag-pol in large-scale, high titer stocks (>109) of transvector support the idea that the trans-lenti design help control the risks associated with genetic recombination.

156

Lentiviral Vector Systems for Gene Transfer

Fig. 6. Monitoring for env-minus recombinant lentivirus. The diagram illustrates how the in vitro analysis of vector stocks for env-minus recombinant lentivirus may serve as a surrogate marker for the emergence of RCR in vivo. In the absence of detecting recombinants (LTR-gag-pol-LTR) there would be a greatly diminished chance of generating RCR.

Previous studies indicated that endogenous retroviral sequences may recombine with retroviral vectors and lead to the generation of RCR. This raises the issue of whether RT and IN derived from endogenous retroviruses could complement RT-IN minus trans-vector recombinants. Recently, the protease of the human endogenous retrovirus HERV-K was analyzed for its ability to process the HIV-1 Gag and Gag-Pol precursor polyproteins by targeting the HERV-K protease into immature HIV-1 virions. This analysis demonstrated that the HERVK protease was severely defective in the proteolytic processing of the Gag and Gag-Pol precursors.52 Even the closely related HIV-2 RT protein is not correctly processed by the HIV-1 PR (unpublished). These results suggest it is unlikely that endogenous RT and IN could rescue RT-IN defective recombinants (LTR-gag-pro-LTR). Non-human primate lentiviruses, including equine infectious anemia virus (EIAV), feline immunodeficiency virus (FIV), and bovine immunodeficiency virus (BIV) are being considered as gene therapy vectors,26,53,54 primarily because they are not associated with disease in man. Significant improvements have recently been made in these vector systems (see Chapters 6 and 7). By analogy with the HIV-1-base lentiviral vector, it is unlikely the current design of non-human primate vectors will prevent the generation of recombinant virus. Since the design of the trans-vector packaging construct (Gag-Pro) is based on supplying RT and IN in trans as fusion partners of Vpr/Vpx and similar virion-associated accessory proteins are not encoded by the non-human lentiviruses, the trans-vector design would not be directly applicable to other lenti-vectors. However, the development of alternative approaches to disarm the gag-pol structure of these viruses would seem feasible.

Safety Considerations in Vector Development

157

References 1. Miller A. Retroviral vectors. Cur Top Microbiol 1992; 158:1-24. 2. Vile RG, Russell SJ. Retroviruses as vectors. Br Med Bull 1995; 51:12-30. 3. Lewis P, Emerman M. Passage through mitosis is required for oncoretroviruses but not for the human immunodeficiency virus. J Virol 1994; 68:510-516. 4. Roe T, Reynolds T, Yu G et al. Integration of murine leukemia virus DNA depends on mitosis. EMBO J 1993; 12:2099-2108. 5. Mann R, Mulligan RC, Baltimore D. Construction of a retrovirus packaging mutant and its use to produce helper-free defective retrovirus. Cell 1983; 33:153-159. 6. Miller MD, Farnet CM, Bushman FD. Human immunodeficiency virus type 1 preintegration complexes: Studies of organization and composition. J Virol 1997; 71:5382-5390. 7. Anderson WF, McGarrity GJ, Moen RC. Report to the NIH recombinant advisory committee on murine replication-competent retrovirus (RCR) assays. Hum Gene ther 1993 ;4:311-321. 8. Wilson CA, Ng TH, Miller AE. Evaluation of recommendations for replication-competent retrovirus testing associated with use of retroviral vectors. Hum Gene ther 1997; 8:869-874. 9. Miller AD. Development and application of retroviral vectors. In: Coffin JM, Hughes SH, Varmus HE, eds. Retroviruses. Cold Spring Harbor: Cold Spring Harbor Laboratory, 1997:437-474. 10. Donahue RE, Kessler SW, Bodine D et al. Helper virus induced t cell lymphoma in nonhuman primates after retroviral mediated gene transfer. J Exp Med 1992; 176:1125-1135. 11. Otto E, Jones-Trower A, Vanin EF et al. Characterization of a replication-competent retrovirus resulting from recombination of packaging and vector sequences. Hum Gene Ther 1994; 5:567-575. 12. Vanin EF, Kaloss M, Broscius C et al. Characterization of replication-competent retroviruses from nonhuman primates with virus-induced t-cell lymphomas and observations regarding the mechanism of oncogenesis. J Virol 1994; 68:4241-4250. 13. Purcell DFJ, Broscius CM, Vanin E et al. An array of murine leukemia virus-related elements is transmitted and expressed in a primate recipient of retroviral gene transfer. J Virol. 1996; 70:887-897. 14. Bestwick RK, Kozak SL, Kabat D. Overcoming interference to retroviral superinfection results in amplified expression and transmission of cloned genes. Proc Natl Acad Sci USA 1988; 85:5404-5408. 15. Bodine DM, McDonagh KT, Brandt SJ et al. Development of a high-titer retrovirus product line capable of gene transfer into rhesus monkey hematopoietic stem cells. Proc Natl Acad Sci USA 1990; 87:3738-3742. 16. Chong H, Vile RG. Replication-competent retrovirus produced by a “split-function” third generation amphotropic packaging cell line. Gene Ther 1996 ;3:624-629. 17. Markowitz D, Goff S, Bank A. Construction and use of a safe and efficient amphotropic packaging cell line. Virology 1988; 167:400-406. 18. Chong H, Starkey W, Vile RG. A replication-competent retrovirus arising from a split-function packaging cell line was renerated by recombination events between the vector, one of the packaging constructs, and endogenous retroviral sequences. J Virol. 1998; 72:2663-2670. 19. Rigg RJ, Chen J, Dando JS et al. A novel human amphotropic packaging cell line: High titer, complement resistance, and improved safety. Virology 1996; 218:290-295. 20. Boeke JD, Stoye JP. Retrotransposons, endogenous retroviruses, and the evolution of retroelements. In: Coffin JM, Hughes SH, Varmus HE, eds. Retroviruses. Plainview: Cold Spring Harbor Laboratory Press, 1997:343-435. 21. Bukrinsky MI, Sharova N, Dempsey MP et al. Active nuclear import of human immunodeficiency virus type 1 preintegration complexes. Proc Natl Acad Sci USA 1992; 89:6580-6584. 22. Lewis PF, Hansel M, Emerman M. Human immunodeficiency virus infection of cells arrested in the cell cycle. EMBO J 1992; 11:3053-3058. 23. Amado RG, Chen ISY. Lentiviral vectors-the promise of gene therapy within reach? Science 1999; 285:674-676. 24. Kordower JH, Emborg ME, Bloch J et al. Neurodegeneration prevented by lentiviral vector delivery of GDNF in primate models of Parkinson’s disease. Science 2000; 290:767-773. 25. Miyoshi H, Smith KA, Mosier DE et al. Transduction of human CD34+ cells that mediate longterm engraftment of NOD/SCID mice by HIV vectors. Science 1999; 283:682-686. 26. Trono D. Lentiviral vectors: Turning a deadly foe into a therapeutic agent. Gene Ther 2000; 7:20-23. 27. May C, Rivella S, Callegari J et al. Therapeutic heamoglobin synthesis in β-thalassaemic mice expressing lentivirus-encoded human β-globin. Nature 2000; 406:82-86.

158

Lentiviral Vector Systems for Gene Transfer

28. Kafri T, VAN Praag H, Ouyang L et al. A packaging cell line for lentivirus vectors. J Virol 1999; 73:576-584. 29. Naldini L, Blomer U, Gallay P et al. In vivo gene delivery and stable gene transduction of nondividing cells by a lentivirus vector. Science 1996; 272:263-267. 30. Dull T, Zufferey R, Kelly M et al. A third-generation lentivirus vector with a conditional packaging system. J Virol 1998; 72:8463-8471. 31. Kim VN, Mitrophanous K, Kingsman SM et al. Minimal Requirement for a lentivirus vector based on human immunodeficiency virus type 1. J Virol 1998; 72:811-816. 32. Zufferey R, Nagy D, Mandel RJ et al. Multiply attenuated lentiviral vector achieves efficient gene delivery in vivo. Nat Biotechnology 1997; 15:871-875. 33. Gasmi M, Glynn J, Jin M-J et al. Requirements for efficient production and transduction of human immunodeficiency virus type 1-based vectors. J Virol 1999; 73(3):1828-1834. 34. Miyoshi H, Blomer U, Takahashi M et al. Development of a self-inactivating lentivirus vector. J Virol 1998; 72:8150-8157. 35. Zufferey R, Dull T, Mandel RJ et al. Self-inactivating lentivirus vector for safe and efficient in vivo Gene Therapy. J Virol 1998; 72:9873-9880. 36. Wu X, Wakefield KJ, Liu H et al. Development of a novel trans-lentiviral vector that affords predictable safety. Mol Ther 2000; 2:47-55. 37. Huang CC, Hay N, Bishop JM. The role of RNA molecules in transduction of the proto-oncogene c-fps. Cell 1986; 44:935-940. 38. Raines MA, Maihle NJ, Mascovici C et al. Mechanism of c-erbB transduction: Newly released transducing viruses retain poly(A) tracts of erbB transcripts and encode C-terminally intact erbB proteins. J Virol 1988; 62:2437-2443. 39. Liu H, Wu X, Xiao H et al. Incorporation of functional human immunodeficiency virus type 1 integrase into virions independent of the Gag-Pol precursor protein. J Virol 1997; 71:7701-7710. 40. Wu X, Liu H, Xiao H et al. Human immunodeficiency virus type 1 integrase protein promotes reverse transcription through specific interactions with the nucleoprotein reverse transcription complex. J Virol 1999; 73:2126-2135. 41. Wu X, Liu H, Xiao H et al. Functional RT and IN incorporated into HIV-1 particles independently of the Gag-Pol precursor protein. EMBO J 1997; 16:5113-5122. 42. Wu X, Liu Ho, Xiao H et al. Targeting foreign proteins to human immunodeficiency virus particles via fusion with Vpr and Vpx. J Virol 1995; 69:3389-3398. 43. Chen WY, Wu X, Levasseur DN et al. Lentiviral vector transduction of hematopoietic stem cells that mediate long-term reconstitution of lethally irradiate mice. Stem Cells 2000; 18:352-359. 44. Abonour R, Williams DA, Einhorn L et al. Efficient retrovius-mediated transfer of the multidrug resistance 1 gene into autologous human long-term repopulating hematopoietic stem cells. Nat Med 2000; 6(6):652-628. 45. Miyoshi H, Takahashi M, Gage FH et al. Stable and efficient gene transfer into the retina using an HIV-based lentiviral vector. Proc Natl Acad Sci USA 1997; 94:10319-10323. 46. Takahashi M, Miyoshi H, Verma IM et al. Rescue from photoreceptor degeneration in the rd mouse by human immunodeficiency virus vector-mediated gene transfer. J Virol 1999; 73:7812. 47. Endres MJ, Jaffer S, Haggarty B et al. Targeting of HIV- and SIV-infected cells by CD4-chemokine receptor pseudotypes. Science 1997; 278:1462-1464. 48. Mebatsion T, Finke S, Weiland F et al. A CXCR4/CD4 pseudotype rhabdovirus that selectively infects HIV-1 envelope protein-expressing cells. Cell 1997; 90:841-847. 49. Schnell MJ, Johnson JE, Buonocore L et al. Construction of a novel virus that targets HIV-1infected cells and controls HIV-1 infection. Cell 1997; 90:849-857. 50. Blobel CP, Wolfsberg TG, Turck CW et al. A potential fusion peptide and an integrin ligand domain in a protein active in sperm-egg fusion. Nat Med 1992; 356:248-252. 51. Pang S, Yu D, An D-S et al. Human immunodeficiency virus env-independent infection of human CD4- cells. J Virol 2000; 74:10994-11000. 52. Padow M, Lai L, Fisher RJ et al. Analysis of human immunodeficiency virus type 1 containing HERV-K protease. AIDS Res Hum Retroviruses 2000; 16(18):1973-1980. 53. Olsen JC. Gene transfer vectors derived from equine infectious anemia virus. Gene Ther 1998; 5:1481-1487. 54. Poeschla EF, Wong-Staal F, Looney DJ. Efficient transduction of nondividing human cells by feline immunodeficiency virus lentiviral vectors. Nat Med 1998; 4:354-357.

CHAPTER 9

Prospects for Gene Therapy Using HIV-Based Vectors Jiing-Kuan Yee and John A. Zaia

Abstract

R

ecombinant vectors derived from murine leukemia virus (MLV) have been widely used to introduce genes in human gene therapy clinical trials and have shown the potential for medical applications and the promise of significantly improving medical therapies. Yet, the demonstrated limitations of these vectors support the need for continued development of improved vectors. The intrinsic properties associated with the MLV genome and its life cycle do not favor the successful application of this vector system in certain human gene transfer applications. Since MLV integrates randomly into the host genome, transgene expression is frequently affected by the flanking host chromatin.1 MLV insertions can often result in silencing or position effect variation of gene expression either immediately after insertion or following cell expansion in culture or in vivo.2-6 Migration of the MLV pre-integration complex from the cytoplasm into the nucleus of infected cells requires mitosis for nuclear membrane breakdown.7 Since a majority of human cells exist in a quiescent state in vivo, it is unlikely that direct in vivo gene delivery into target tissues can be achieved with the MLV vector system. Finally, insertion of tissue-specific cis-regulatory sequences to direct transgene expression frequently results in either the rearrangement of the vector sequence or disruption of the cis-regulatory sequence functions.8-12 The long terminal repeat (LTR) of MLV, which contains a ubiquitously active enhancer/promoter element, may partially account for this problem. Together, these problems pose a major obstacle for the use of MLV vectors in the treatment of human diseases. This Chapter discusses some of the potential targets to which HIV vectors might be applied in clinical settings and some of the issues surrounding use of HIV vectors in gene transfer clinical trials.

Advantages of HIV Vectors Support Development Human immunodeficiency virus-1 (HIV-1)-based vectors have recently been demonstrated to efficiently deliver genes into mammalian cells.13-18 Like MLV, HIV integrates randomly into the host genome. It is most likely, therefore, that expression of the transgene delivered by an HIV vector is integration site-dependent. Unlike MLV, however, HIV is capable of transducing quiescent cells both in culture and in vivo.14,15,17,19-25 This property of HIV is mediated by several virion-associated proteins, including the matrix protein, the accessory protein Vpr and integrase.26-28 It is believed that the nuclear localization signals (NLS) in these proteins mediate migration of the viral pre-integration complex into the nucleus through the nuclear pore. However, the role played by each protein and the exact mechanisms that mediate pre-integration complex migration in different cell types remain controversial (see Chapter 3). An additional Lentiviral Vector Systems for Gene Transfer, edited by Gary L. Buchschacher, Jr. ©2003 Eurekah.com.

160

Lentiviral Vector Systems for Gene Transfer

advantage of using HIV vectors is the ability to accommodate inserted cis-regulatory sequences to direct transgene expression.15,29,30 This is in contrast to the observation that insertion of similar sequences into MLV vectors frequently leads to vector genome rearrangement and functional loss of the cis-regulatory sequences.8-12 Although the reasons for this difference between the two vector systems remain unclear, it is likely that sequence elements present in a MLV vector are not always compatible with an inserted cis-regulatory sequence. While several major obstacles exist that may impede the start of human gene transfer clinical trials, such as isolation of stable packaging cell lines for efficient scale-up of vector production and establishment of processes for bulk vector purification and of standardized and sensitive procedures for the detection of replication competent retrovirus (RCR), the advantages of HIV vectors open exciting perspectives for using this vector system to ultimately treat human diseases.

Attempts to Engineer Biosafety into HIV Vectors One major issue of using an HIV vector system in human gene transfer experiments is concern for its safety (discussed in further detail in Chapters 4, 8, and 10). RCR contamination in a vector preparation could lead to HIV infection and induce detrimental consequences in human patients. Thus, before the vector system can be applied in human patients, a major effort should therefore be made to minimize or completely eliminate the possibility of generating RCR during vector production. In addition, highly sensitive and standardized assays to detect low levels of RCR contamination in vector preparations need to be established. Currently, a variety of modifications have already been made in the vector production system to reduce the possibility of RCR production. First-generation HIV vectors were generated by transient transfection of human 293T cells with a plasmid containing the desired transgene inserted in an HIV vector, a packaging plasmid expressing all of the HIV genes except the envelope gene, and an expression plasmid for the glycoprotein (G) of vesicular stomatitis virus (VSV). Unlike the natural HIV cell surface CD4 receptor, the receptor for VSV is widely expressed in various cell types.31 More importantly, VSV-G pseudotyped MLV or HIV vectors can sustain the force of ultracentrifugation and be concentrated to extremely high titers (> 109 transduction units/ml).17,32 The availability of high-titer vector preparations significantly facilitates direct gene delivery into animal models in vivo.

Second-Generation HIV Vectors While the crude vector titer generated from 293T cells is generally in the range between 106 and 107 transduction units (TU)/ml, RCR can theoretically be generated through DNA recombination during transfection. Co-packaging of the RNAs derived from the packaging plasmid and the vector genome into the same virion can also give rise to RCR during the subsequent round of reverse transcription in the transduced cells. Second-generation HIV vectors were generated in the absence of the four HIV-encoded accessory proteins: Vif, Vpr, Vpu and Nef, through mutation or deletion of these genes from the packaging plasmid.18,33,34 Accumulating evidence suggests that these four accessory proteins are crucial determinants of HIV virulence.35 Human patients infected with Nef-deficient HIV strains generally have lower viral loads and develop slower declines in CD4+ T-lymphocyte counts.36 In adult macaques, replication of simian immunodeficiency virus (SIV) with deletion of the Vpr, Nef and a sequence in the U3 region was markedly attenuated, although development of AIDS in a minority of inoculated adults continued to be observed.37,38 Complete removal of SIV accessory protein genes and the sequence in the U3 region resulted in a severely attenuated virus that failed to replicate and induce AIDS in inoculated rhesus monkeys.39 The titer of the vector generated from transiently transfected 293T cells was not affected by the absence of the accessory proteins.18 33 34 More importantly, the ability of the accessory protein-deficient HIV vector to transduce various cell types was not compromised when compared with its accessory

Prospects for Gene Therapy Using HIV-Based Vectors

161

protein-replete counterpart. These observations, however, do not exclude the possibility that the accessory proteins may contribute in subtle ways to enhance the transduction efficiency of specific cell types. Kafri et al reported that the use of a Vpr(-)/Vif(-) HIV vector in vivo resulted in a significant decrease in its ability to transduce liver cells.14 Gasmi et al also observed that the complete absence of all four accessory proteins resulted in a decrease in the ability of an HIV vector to transduce quiescent primary human skin fibroblasts.33 Chinnasamy et al demonstrated that HIV vectors without the accessory proteins failed to integrate in resting lymphocytes.40 Thus, the accessory proteins may facilitate efficient transduction of HIV vectors under certain conditions. Nevertheless, complete elimination of accessory proteins from HIV vector preparations seems to have little effect on titers and should significantly enhance the biosafety of this vector system in human clinical trials.

Third-Generation HIV Vectors Third-generation HIV vectors include other features besides the removal of the genes encoding all the accessory proteins. In the vector construct, the U3 region in the 5’LTR was replaced with the strong enhancer derived from the immediate early gene of cytomegalovirus (CMV). Due to the presence of this enhancer, transcription initiation to generate vector RNA is no longer Tat-dependent.34,41,42 As Tat is an essential gene product for HIV replication, the absence of the Tat gene during vector production further reduces the possibility of generating replication competent HIV. In addition, the enhancer and the promoter sequences in the U3 region of the 3’LTR are also removed to generate so-called self-inactivating (SIN) vectors.42,43 Since the U3 region in the 3’LTR serves as the template to generate the U3 region in both LTRs during reverse transcription, neither LTR in a SIN vector is transcriptionally active. Expression of the transgene is therefore completely dependent on an inserted cis-regulatory sequence. Inactivation of the LTRs also renders replication-competent HIV incapable of rescuing a Sin vector and spreading it from transduced cells to other non-transduced cells.43 In addition, this modification minimizes fortuitous activation of the cellular genes flanking the vector integration site by the HIV LTR. Unlike MLV-based SIN vectors which generated extremely low titers,44,45 HIV-based SIN vectors have titers similar to their normal counterparts with intact LTRs. The reason for this difference in titers remains unclear.

Rev-Independent HIV Vector Production With the modifications mentioned above, only three HIV encoded proteins, Gag, Pol and Rev, are required for the vector production. More recently, Kotsopoulou et al have made further modifications in the packaging plasmid to eliminate the requirement of Rev for vector production.46 HIV gene expression is tightly regulated by the binding of Rev to the RRE sequence within the env gene. The RRE is present in all unspliced and singly spliced messages and Rev binding results in rapid nuclear export of these messages into cytoplasm. It was postulated that the presence of inhibitory sequences (INS) throughout the HIV genome was responsible for nuclear retention and instability of these HIV messages. Several segments with potential INS properties have been identified in the gag-pol and env genes and within the RRE sequence itself.47-51 However, the exact sequences and the mechanisms responsible for the inhibitory effect remain undefined. Efficient HIV protein expression is restricted not only by the presence of INS, but also by inefficient translation of HIV-specific mRNAs due to biased codon usage. The HIV genome is AU-rich, and cannot be translated efficiently in human cells when compared with highly expressed human genes.52 To optimize the codon usage of the HIV genome, Kotsopoulou et al have constructed a completely synthetic HIV-1 gag-pol gene using favored codon in human cells.46 The INS in the gag-pol genes, however, was eliminated in the synthetic genes due to the changes in primary nucleotide sequences, leading to Revindependent gag-pol expression. Since the altered sequence contains no substantial regions of

162

Lentiviral Vector Systems for Gene Transfer

homology with any naturally occurring HIV-1 gag-pol sequence, the possibility of RCR production through homologous recombination between the packaging plasmid and the vector is also minimized. While vector titers generated from Rev-independent packaging plasmid were somewhat lower than that from Rev-dependent packaging plasmid, this study clearly demonstrates the possibility of generating HIV vectors with only two of HIV-encoded proteins: Gag and Pol.46 Such a Rev-independent Gag-Pol expression system, coupled with the observation that multiple vector sequence modifications eliminated the requirement of RRE and Rev for the production of the vector genomic RNA,53 should significantly alleviate the concern of applying HIV-based vectors in the treatment of human diseases. The presence of only the gag and pol genes in the HIV vector production system makes it even less likely to produce RCR than the MLV vector production system since HIV replication is critically dependent on Tat and Rev. Complete removal of the HIV promoter and enhancer sequences in SIN vectors further reduces the likelihood of RCR production, either through recombination between the packaging plasmid and the vector construct or through fortuitous co-packaging of RNAs derived from the packaging construct and the vector construct into the same virion. In this regard, the MLV vector system may be more likely to generate RCR, since the MLV promoter and enhancer sequences do not depend on any viral-encoded proteins for function and the LTR is transcriptionally active in most cell types. Since only Gag, Pol and VSV-G are required for the packaging of the HIV vector genomic RNA, stable packaging cell lines might be established. The problem of cell toxicity caused by stable expression of Vpr and Nef is therefore avoided.54-59 The presence of the stable packaging cell lines might not only facilitate large-scale vector production, but also minimizes the possibility of RCR production through DNA recombination during transient transfection for vector production. While VSVG expression is toxic to mammalian cells, inducible expression of this protein in established stable cell lines has been demonstrated before.60,61 With the experience in MLV vector production, it is very likely that stable HIV packaging cell lines that generate reasonably high vector titers can ultimately be established and used in human gene therapy clinical trials.62-64

RCR Detection Given the limitations of available animal models for HIV infection, the biosafety of HIV vectors will ultimately be determined in human patients. Thus, sensitive assays need to be established to detect the presence of RCR in vector preparations. The HIV p24 ELISA assay to detect the viral capsid protein is an extremely sensitive assay. The amount of p24 as low as 10 picogram can readily be detected with this assay system. To increase the sensitivity, a certain portion of the vector preparation is first allowed to infect a permissive cell line for HIV replication, such as human T cell lines. The infected cells are serially passaged to allow sufficient time for the amplification of contaminating RCR that may be present in very low abundance. The culture supernatant is harvested and virions in the culture supernatant are concentrated by ultracentrifugation and detected by the p24 assay kit. An alternative approach to detect RCR is using the marker rescue assay. A human cell line containing an integrated HIV vector with a selectable marker gene such as that encoding neomycin phosphotransferase (neo) can first be established. This cell line is transduced at high multiplicity of infection (MOI) with the vector preparation containing the gene of interest. If RCR is present in the vector preparation, the Neo vector would be rescued and released into the culture medium. The presence of the rescued vector can be detected by transduction and selection in G418-containing medium for colony formation. The sensitivity of such an assay to detect RCR, however, remains to be established. Ultimately, both the p24 assay and the marker rescue assay may be combined together to increase the sensitivity of detecting RCR in vector preparations. Further discussion regarding detection of RCR can be found in Chapter 8.

Prospects for Gene Therapy Using HIV-Based Vectors

163

Potential Clinical Applications of HIV Vectors Gene Transfer into Neurons In this and the following sections, two specific cell types, the neuronal and hematopoietic progenitor cells, are used as examples to demonstrate the potential of applying HIV vectors in treating human diseases. To test the ability of HIV vectors to transduce terminal differentiated neuronal cells, concentrated HIV vector stocks encoding the β-galactosidase gene were injected bilaterally into the corpus striatum and hippocampus of adult rat brains.17 Cells expressing both β-galactosidase and the neuron marker NeuN were detected in both areas one month after injection, demonstrating the ability of HIV vectors to transduce genes into terminally differentiated neuronal cells. Hottinger et al tested HIV vector-mediated delivery of the neurotrophic factor GDNF (glial cell line-derived neurotrophic factor) for its ability to rescue motor neurons from apoptosis.65 GDNF was chosen because it was the most potent neurotrophic factor for cultured embryonic motor neurons described.66 In the experimental model, facial nerve lesions in Balb/C mice lead to a progressive and long-term loss of motor neurons.67,68 This relatively slow cell death resembles motor neuron degenerative diseases such as amyotrophic lateral sclerosis (ALS). This group demonstrated that expression of GDNF with an HIV vector near the motor neuron cell bodies of the facial nucleus led to sustained expression and diffusion of GDNF and protection of motor neurons against lesion-induced apoptosis and atrophy.65 HIV-mediated gene delivery of GDNF also has been tested for effects on degenerating nigrostriatal neurons in nonhuman primate models of Parkinson’s disease (PD).69 In this case, the vector was injected into the striatum and substantia nigra of non-lesioned aged rhesus monkeys or young adult rhesus monkeys treated with 1-methyl-4-phenyl-1, 2, 3, 6tetrahydropyridine (MPTP). Non-lesioned aged monkeys display slow progressive loss of dopamine within the striatum and of tyrosine hydroxylase (TH) within the substantia nigra without frank cellular degeneration.70 Young adult monkeys receiving MPTP treatment exhibit extensive nigrostriatal degeneration, resulting in a behavioral syndrome characterized by motor deficits. In the non-lesioned old monkey model, GDNF gene delivery resulted in a significant increase of TH-immunoreactive neurons within the injected striatum and substantia nigra. In the MPTP-treated young monkey model, GDNF gene delivery completely prevented nigrostriatal degeneration, resulting in improvement in the parkinsonian clinical rating scale and reversed motor deficits. Deficiency in lysosomal enzyme arylsulfatase A (ARSA) causes metachromatic leukodystrophy (MLD). ARSA catalyzes the first step in the degradation pathway of galactosyl-3-sulfate ceramide (sulfatide), a major sphingolipid of myelin. MLD is characterized by myelin degeneration in both the central and peripheral nervous systems, leading to progressive neurologic symptoms including ataxia, seizures, quadriplegia, and death with decerebration in infancy.71 Using a mouse model of MLD, Gonsiglio et al injected an HIV vector containing the ARSA cDNA into the hippocampal fimbria. Persistent expression of the active enzyme throughout most of the injected area was observed. Lipid deposits were significantly reduced, resulting in effective rescue of hippocampal neurons from degeneration. In addition, the therapeutic activity of the transgene spread to a progressively larger fraction of the brain over time, possibly a result of sustained release of ARSA from transduced cells and diffusion to areas containing non-transduced cells. The performance in both short-term and long-term memory tests of ARSA vector-treated MLD mice was better than the control vector-treated MLD mice and was indistinguishable from the performance of wild-type mice. Taken together, these results demonstrate that HIV vector-mediated gene delivery in vivo holds promise as a treatment for diseases in the central nervous system (CNS).

164

Lentiviral Vector Systems for Gene Transfer

Retinitis pigmentosa (RP), one of the most common forms of retinal dysfunction, is a result of photoreceptor cell degeneration. While various genes presumably are involved in the development of RP, mutations in the rod photoreceptor cGMP phosphodiesterase β subunit (PDEβ) gene are found in patients with autosomal recessive RP as well as in rd mice and rcd1 Irish setters.72-75 Miyoshi et al injected HIV vectors into the subretinal space of rat eyes and demonstrated efficient transduction of both photoreceptor cells and retinal pigment epithelium.15 Based on this result, Takahashi et al evaluated RP progression in rd mice by the delivery of the PDEβ cDNA via HIV vectors.76 Several rows of photoreceptor cells were detected in PDEβ vector-treated rd mice for at least 24 weeks post-injection, whereas no photoreceptor cells remained in the eyes of control animals 6 weeks post-injection. PDEβ expression was also detectable in the photoreceptor cells of PDEβ vector-treated rd mice. Thus, besides treating CNS diseases, HIV vector-mediated gene delivery may also be employed to treat inherited retinal degeneration.

Gene Transfer into Hematopoietic Progenitor Cells While MLV-based vectors are able to transduce mouse hematopoietic progenitor cells efficiently, the same success is not observed with human hematopoietic progenitor cells.77,78 This could be due to the quiescent nature of human cells and the requirement of cell division for successful MLV infection. Since HIV vectors are able to transduce quiescent or even terminally differentiated cells, their ability to transduce human hematopoietic progenitor cells was tested in several studies. Miyoshi et al compared the transduction efficiency of HIV and MLV vectors in human cord blood CD34+ cells.21 In the absence of exogenous cytokines, they demonstrated that an HIV vector containing the green fluorescent protein (GFP) gene was able to transduce these cells with significantly higher efficiency than a MLV vector in a colony-forming cell (CFC) assay in methylcellulose. Transplantation of the transduced CD34+ cells into sublethally irradiated NOD/SCID mice resulted in the detection of GFP(+) human cells in bone marrow, spleen and peripheral blood of engrafted mice. Both human myeloid and lymphoid lineages in the bone marrow and spleen of these mice demonstrated GFP(+) cells, and the proportion of GFP(+) cells remained roughly constant for up to 22 weeks, indicating sustained proliferation of the transduced human cells in vivo. In contrast, no GFP(+) cells were detected in mice transplanted with MLV vector-transduced CD34+ cells. Several other studies were performed using either CD34+ or CD34+ CD38- cells as the target for HIV vector transduction.79-84 Together, these studies demonstrate the ability of HIV vectors to transduce very primitive human hematopoietic progenitor cells with long-term engraftment ability in NOD/SCID mice. Efficient gene delivery into hematopoietic progenitor cells in conjunction with the fact that cis-regulatory elements remain relatively stable in the context of an HIV vector offers the possibility of using gene transfer approaches to treat human genetic diseases such as β thalassemia. May et al constructed a first-generation HIV vector containing the human β-globin gene controlled by large segments of the β-globin locus control region (LCR).30 Using this vector, they were able to demonstrate efficient gene delivery into murine hematopoietic progenitor cells. Normal mice transplanted with the transduced bone marrow cells exhibited tetramers of two murine β-globin and two human β-globin molecules that could account for up to 13% of total hemoglobin in mature red cells 24 weeks after transplantation. Strikingly, in βthalassemic heterozygous mice that had a clinical phenotype similar to human thalassemia intermedia and showed chronic anemia, efficient gene delivery and expression were sufficient to ameliorate anemia and red cell morphology. This example clearly demonstrates the advantage of using HIV vectors for stable gene delivery into hematopoietic progenitor cells.

Prospects for Gene Therapy Using HIV-Based Vectors

165

Potential Applications to HIV/AIDS Current state of the art treatment for HIV-1 infection, highly active anti-retroviral therapy (HAART), employs combinations of three or more drugs. Of these, two are usually inhibitors of reverse transcriptase (RT), while one is often an inhibitor of HIV-1 protease. This approach has proven to be a great improvement over single-agent chemotherapy. HAART decreases virus load in the peripheral blood, often to undetectable levels. It increases CD4+ T lymphocytes in the peripheral blood. Opportunistic infections are less frequent and often less severe. Most importantly, many parameters of immune function are improved; patients are healthier and live longer.118-122 As compelling as these results are, current HAART regimens have limitations. Many compliant patients do not respond completely to HAART, especially—but not limited only to— people who have had prior therapy.123-125 HAART regimens are costly and can have major unpleasant or toxic side effects.126,127 In addition, dosing regimens are complex and difficult to follow. With therapy, overall immunity improves, but the increased peripheral blood CD4+ T cells often do not reflect new naive CD4+ cells, but rather mobilization of committed CD4+ cells from reserves.128,129 Although some increases in naive cells may be seen over time, specific immune function may or may not improve,130-132 and, thus, despite continued objective improvement with therapy, HIV-1 persists,122,133-135 probably for life,136 and may still be transmitted to others.137 Most disturbingly, HAART-resistant strains of HIV-1 arise during therapy,138141 and, for any specific patient on HAART, it is likely that resistance will become increasingly important in the management over a lifetime. Thus, there is a need for additional therapeutic options in treating AIDS patients. Therapies that are less toxic, less expensive, and/or more specifically, would be desirable adjuncts to HAART. The extensive information available regarding the molecular biology of HIV infection and replication has presented multiple potential strategies for applying gene transfer approaches to new treatments of HIV/AIDS. A diverse array of transgenes has been developed to suppress HIV-1 functions or block the infectious cycle, and these can be categorized into two types: RNA elements and proteins. Among the RNA-based suppressors are various antisense molecules designed to target such critical HIV genes as tat, rev, and integrase.85-89 In addition, RNA decoys include RNA homologues, such as TAR and RRE, that recognize and bind viral proteins and compete with the native ligands necessary for replication.90-93 Another category is ribozymes, RNA molecules that can cleave RNA at specific sequences and that can be designed to target HIV at critical sites such as tat, rev, gag.94-97 The protein structures developed for targeting of HIV by gene transfer include transdominant negative mutants, intrakines, toxins, and single chain antibodies. RevM10, a protein that retains two Rev functions—the ability to bind RRE and the ability to form Rev multimers, was the first transdominant protein to be evaluated in human trials.98,99 Other examples include tat 100 and a fusion of tat and rev transdominant genes coding for a Tat/Rev fusion protein called Trev.101 Intracellular toxins or conditionally toxic proteins, such as herpes simplex thymidine kinase,102 HIV-dependent diphtheria toxin 103, and even modified lytic viruses have been designed for anti-HIV activity.104,105 Since HIV-1 uses the cellular CD4 receptor and a chemokine co-receptor to infect cells, systems utilizing intracellular expression of either SDF-1, the ligand for CXCR4; RANTES and MIP-1a, the ligands for CCR5; or CD4 itself have been shown to inhibit HIV-1 infection in vitro.106-109 Finally, intracellular HIV-specific single-chain antibodies (intrabodies; SFv) can target and redirect essential HIV proteins away from required subcellular compartments, and block the function or procession of essential proteins, such as HIVgp120,110 Rev,111 Gag, 112 reverse transcriptase113 and integrase.114

166

Lentiviral Vector Systems for Gene Transfer

The complexities of HIV-1 pathogenesis, the high mutation rate of the viral genome, and its ability to persist in lymphoid and other tissues, all allow HIV-1 to evade many therapies.87 115 HIV-1 integrates into the cellular genome, which facilitates persistence and acts as a reservoir for reactivation and replication. Cells non-productively infected with HIV-1 have been identified following infection in vitro and in vivo. These cells may shelter the virus from antiviral therapy.87,116,117 The problem of applying these genetic strategies using HIV vectors is that there is the possibility that the anti-HIV gene will interfere with the production of the vector. For example, if the gene targeted by the anti-HIV strategy is required for genomic RNA production and/or packaging, then there will be decreased vector titer. Without efficient vector titers, transduction of hematopoietic progenitor cells or T lymphocytes would be limited. It is even possible that targeting integrase with SFv will adversely effect not only vector production but also subsequent cellular integration. Therefore, targets such as Rev, RRE, or Gag-Pol would be problematic, but agents that target Tat, any accessory protein, or envelope should not interfere with vector production or subsequent transduction.

Regulatory Issues General Considerations The unrealistic expectations associated with gene transfer research in the early 1990's generated concern that this area of research needed to be assessed more realistically. Because of the perception that unrestrained public pronouncements regarding gene therapy studies had unrealistically influenced the field, there was concern about the overall quality of this research. An expert panel reviewed the field and made several recommendations [see http://www.nih.gov/ news/panelrep.html]. The panel suggested that this research be considered “human gene transfer” research since the terminology “gene therapy” suggested potential effects on disease that might be unlikely to occur at this stage of development. In addition, it was recommended that there be a better focus on the basic science aspects of this research, including more pathogenetic studies and better use of animal models. Furthermore, the panel suggested the need for an improvement in the quality of gene therapy protocols and in the training of those performing the clinical studies. In 1999, after there had been hundreds of human research subjects safely enrolled into human gene transfer research protocols, there were two events which changed the way this research is regulated. The first was the death of a young man of liver failure after being treated with a direct infusion of an adenovirus encoding an enzyme involved in ornithine metabolism. The second was a report of a protocol violation regarding the treatment of a patient with an adenovirus encoding a potential angiogenesis enhancement factor at a time when the subject had a lung cancer which was an exclusionary criterion. Investigation of these incidences led to the conclusion that correct oversight of these important clinical protocols might not have been adequate. Furthermore, because both studies were conducted at prominent medical centers, the reviews raised the question of whether the conduct of gene transfer research was adequate in the multiple other centers conducting such studies. The Food and Drug Administration (FDA) randomly inspected approximately 10% of all gene transfer studies and found no significant other problems. Nevertheless, the potential for serious malfeasance was established, and therefore in early 2000, new rules were developed at both the NIH and at FDA for human gene transfer research. Heretofore, phase I studies had not been as rigorously held to “good clinical practices” as phase III trials, but beginning in 2000, both the FDA and the NIH inaugurated changes for improved research subject protection.

Prospects for Gene Therapy Using HIV-Based Vectors

167

Review of Human Gene Transfer Research Overview of Regulatory Review The basis for the regulations regarding recombinant DNA is the NIH Guidelines, the purpose of which is to define practices for constructing and handling recombinant DNA molecules and organisms, including viruses, containing recombinant DNA molecules. Recombinant DNA molecules are defined as either molecules that are constructed outside living cells by joining natural or synthetic DNA segments to DNA molecules that can replicate in a living cell or that result from the replication of such molecules. As a condition of continued funding, the NIH required that an institution abide by the NIH Guidelines for all recombinant DNA research projects regardless of funding source. The Office of Biotechnology Activity (OBA) at NIH has responsibility for overseeing the adequate implementation of the NIH Guidelines. When recombinant DNA molecules are considered for administration to one or more human subjects, the information that must be provided for review is clearly outlined in Appendix M of the NIH Guidelines (see http://www4.od.nih.gov/oba/RAC/guidelines/ appendix_m.htm). This describes the “points to consider” for the design and submission of the clinical research protocol. This information is submitted to OBA and is reviewed by the Recombinant DNA Advisory Committee (RAC). Of note, the RAC review and its recommendations represent the only information that is not confidential and, therefore, this is the only public forum for review of human gene transfer research. The main purpose of the RAC review is to ensure that safe and ethical experiments are performed and that understanding of novel areas of biomedical research are available to the public.

Federal Review Review of human gene transfer clinical protocols involves both federal and local review, but for institutions receiving NIH support, the review must begin with the submission of a proposal to OBA. OBA is responsible for monitoring all scientific progress in human genetics research and for facilitating a public discussion regarding the ethical, legal, and social concerns for research involving recombinant DNA, genetic testing, and xenotransplantation. In this regard, OBA manages the operation of the RAC that in turn monitors scientific progress in basic and clinical research involving all recombinant DNA and human gene transfer. Recommendations of the RAC are required before the local review committees can approve any gene transfer protocol involving human subjects. The review by the Food and Drug Administration occurs with the Investigational New Drug (IND) application and with pre-IND meetings with the sponsor of the research. The IND is a request for waiver from laws that require that only approved drugs be used in treating patients. The IND provides a complete description of the source and production of the drug and the approved protocol under which it will be used. The description of specific recommendations for IND preparation can be obtained from the FDA website (http://www.fda.gov/ cber/guidelines.htm)

Local Committee Review The local review of protocols consists of review by the Institutional Biosafety Committee (IBC) and by the Institutional Review Board (IRB) of the institution at which the research is being performed. The IBC is responsible for review of research involving recombinant DNA and addresses safety issues related to the production of the genetic vector and its safe administration (see Section IV-B-2 of the NIH Guidelines). The IBC has responsibility for documenting the sources and nature of DNA used in the experiment, the host and vectors to be used, the nature of the transgene product, and the containment conditions that will be implemented as noted in the NIH Guidelines. The issue about level of containment is described in Appendices

168

Lentiviral Vector Systems for Gene Transfer

E and K of the NIH Guidelines. Thus, this committee provides a means for verifying that the federally mandated rules are being implemented correctly at the local level. The IRB is responsible for review of human subject research to assure that it is scientifically sound, ethical, and safe. The IRB mandate is defined in the code of federal regulation (CFR), e.g., see 45 CFR 46. In addition, many institutions have a scientific review committee that provides guidance on adequacy of the study design.

Review of HIV Vectors for Human Use The major question of safety relates to whether there is a significant chance for homologous recombination due to sequence overlap between the packaging plasmid and the HIV vector that might lead to the production of helper virus. The significant modifications in HIV leading to the production of a self-inactivating virus (see above) minimizes the possibility of generating broad host-range, replication competent HIV during scale-up production of vector. Currently, the vector is generated by transient transfection of human 293T cells with plasmids containing the genes required for vector packaging, the HIV vector itself, and the gene encoding the VSV-G protein.17 The removal of Tat and the accessory proteins during vector production, and the deletion of the HIV promoter and enhancer sequences from the U3 region of the long terminal repeats are major advances in the development of a safe HIV vector. These safeguards should reduce the possibility that infection with wild-type HIV could mobilize an integrated vector genome and the possibility, unlikely though it is, of insertional activation of cellular protoconcogenes. Since the VSV-G gene, instead of the HIV envelope gene, is the envelope of choice in HIV vector production, fortuitous incorporation of the VSV-G gene into an HIV genome could expand its host range extensively with potentially catastrophic results. To apply HIV vectors safely in human clinical trials, it may be important to further modify the vector to minimize the possibility of generating broad host-range, replication-competent virus by this means. This would require incorporation of an anti-VSV-G gene into the HIV vector, and such an approach has not been reported. A crucial part of safety preparation will be the development of reliable methods for detection of helper virus contamination. Sensitive assays to detect low-level helper virus contamination in a vector preparation need to be verified. To facilitate more efficient helper virus detection systems, it is important to identify the most suitable assays and cell lines for HIV amplication. There are 3 such assays: the standard p24 assay, a marker rescue assay, and a RT-PCR assay. These assays are currently available and there should be no reason to doubt the ability to develop adequate product release standards for HIV vectors.

Acknowledgements The authors are grateful for the technical assistance of Ms. Irene Tomeck in the preparation of this manuscript. This work was supported in part by USPHS Grants No. 5P01 A146030, and P01 30206-20, and by grant M01 RR-43 from the GCRC Branch of the National Center for Research Resources, NIH.

References 1. Hoeben RC, Migchielsen AA, van der Jagt RC et al. Inactivation of the Moloney murine leukemia virus long terminal repeat in murine fibroblast cell lines is associated with methylation and dependent on its chromosomal position. J Virol 1991; 65(2):904-912. 2. Flanagan JR, Becker KG, Ennist DL et al. Cloning of a negative transcription factor that binds to the upstream conserved region of Moloney murine leukemia virus. Mol Cell Biol 1992; 12(1):38-44. 3. Palmer TD, Rosman GJ, Osborne WR et al. Genetically modified skin fibroblasts persist long after transplantation but gradually inactivate introduced genes. Proc Natl Acad Sci USA 1991; 88(4):1330-1334.

Prospects for Gene Therapy Using HIV-Based Vectors

169

4. Rivella S, Sadelain M. Genetic treatment of severe hemoglobinopathies: the combat against transgene variegation and transgene silencing. Semin Hematol 1998; 35(2):112-25. 5. Williams DA, Orkin SH, Mulligan RC. Retrovirus-mediated transfer of human adenosine deaminase gene sequences into cells in culture and into murine hematopoietic cells in vivo. Proc Natl Acad Sci USA 1986; 83(8):2566-2570. 6. Xu L, Yee JK, Wolff JA et al. Factors affecting long-term stability of Moloney murine leukemia virus-based vectors. Virology 1989; 171(2):331-341. 7. Roe T, Reynolds TC, Yu G et al. Integration of murine leukemia virus DNA depends on mitosis. Embo J 1993; 12(5):2099-2108. 8. Chang JC, Liu D, Kan YW. A 36-base-pair core sequence of locus control region enhances retrovirally transferred human beta-globin gene expression. Proc Natl Acad Sci USA 1992; 89(7):3107-3110. 9. Gelinas R, Frazier A, Harris E. A normal level of beta-globin expression in erythroid cells after retroviral cells transfer. Bone Marrow Transplant 1992; 9(Suppl 1):154-157. 10. Kaptein LC, Breuer M, Valerio D et al. Expression pattern of CD2 locus control region containing retroviral vectors in hemopoietic cells in vitro and in vivo. Gene Ther 1998; 5(3):320-330. 11. Novak U, Harris EA, Forrester W et al. High-level beta-globin expression after retroviral transfer of locus activation region-containing human beta-globin gene derivatives into murine erythroleukemia cells. Proc Natl Acad Sci USA 1990; 87(9):3386-3390. 12. Plavec I, Papayannopoulou T, Maury C et al. A human beta-globin gene fused to the human betaglobin locus control region is expressed at high levels in erythroid cells of mice engrafted with retrovirus-transduced hematopoietic stem cells. Blood 1993; 81(5):1384-1392. 13. Blomer U, Naldini L, Kafri T et al. Highly efficient and sustained gene transfer in adult neurons with a lentivirus vector. J Virol 1997; 71(9):6641-6649. 14. Kafri T, Blomer U, Peterson DA et al. Sustained expression of genes delivered directly into liver and muscle by lentiviral vectors. Nat Genet 1997; 17(3):314-317. 15. Miyoshi H, Takahashi M, Gage FH et al. Stable and efficient gene transfer into the retina using an HIV-based lentiviral vector. Proc Natl Acad Sci USA 1997; 94(19):10319-10323. 16. Naldini L, Blomer U, Gage FH et al. Efficient transfer, integration, and sustained long-term expression of the transgene in adult rat brains injected with a lentiviral vector. Proc Natl Acad Sci USA 1996; 93(21):11382-11388. 17. Naldini L, Blomer U, Gallay P et al. In vivo gene delivery and stable transduction of nondividing cells by a lentiviral vector. Science 1996; 272(5259):263-267. 18. Zufferey R, Nagy D, Mandel RJ et al. Multiply attenuated lentiviral vector achieves efficient gene delivery in vivo. Nat Biotechnol 1997; 15(9):871-875. 19. Case SS, Price MA, Jordan CT et al. Stable transduction of quiescent CD34(+)CD38(-) human hematopoietic cells by HIV-1-based lentiviral vectors. Proc Natl Acad Sci USA 1999; 96(6):2988-2993. 20. Miyake K, Suzuki N, Matsuoka H et al. Stable integration of human immunodeficiency virusbased retroviral vectors into the chromosomes of nondividing cells. Hum Gene Ther 1998; 9(4):467-475. 21. Miyoshi H, Smith KA, Mosier DE et al. Transduction of human CD34+ cells that mediate longterm engraftment of NOD/SCID mice by HIV vectors. Science 1999; 283(5402):682-686. 22. Reiser J, Harmison G, Kluepfel-Stahl S et al. Transduction of nondividing cells using pseudotyped defective high-titer HIV type 1 particles. Proc Natl Acad Sci USA 1996; 93(26):15266-15271. 23. Sutton RE, Wu HT, Rigg R et al . Human immunodeficiency virus type 1 vectors efficiently transduce human hematopoietic stem cells. J Virol 1998; 72(7):5781-5788. 24. Uchida N, Sutton RE, Friera AM et al. HIV, but not murine leukemia virus, vectors mediate high efficiency gene transfer into freshly isolated G0/G1 human hematopoietic stem cells. Proc Natl Acad Sci USA 1998; 95(20):11939-11944. 25. Wang X, Appukuttan B, Ott S et al. Efficient and sustained transgene expression in human corneal cells mediated by a lentiviral vector. Gene Ther 2000; 7(3):196-200. 26. Bukrinsky MI, Haggerty S, Dempsey MP et al. A nuclear localization signal within HIV-1 matrix protein that governs infection of non-dividing cells. Nature 1993; 365(6447):666-669.

170

Lentiviral Vector Systems for Gene Transfer

27. Gallay P, Hope T, Chin D et al. HIV-1 infection of nondividing cells through the recognition of integrase by the importin/karyopherin pathway. Proc Natl Acad Sci USA 1997; 94(18):9825-9830. 28. Heinzinger NK, Bukinsky MI, Haggerty SA et al. The Vpr protein of human immunodeficiency virus type 1 influences nuclear localization of viral nucleic acids in nondividing host cells. Proc Natl Acad Sci USA 1994; 91(15):7311-7315. 29. Kowolik CM, Hu J, Yee JK. Locus Control Region of the Human CD2 Gene in a Lentivirus Vector Confers Position-Independent Transgene Expression. J Virol 2001; 75(10):4641-4648. 30. May C, Rivella S, Callegari J et al. Therapeutic haemoglobin synthesis in beta-thalassaemic mice expressing lentivirus-encoded human beta-globin. Nature 2000; 406(6791):82-86. 31. Marsh M, Helenius A. Virus entry into animal cells. Adv Virus Res 1989; 36:107-151. 32. Burns JC, Friedmann T, Driever W et al. Vesicular stomatitis virus G glycoprotein pseudotyped retroviral vectors: concentration to very high titer and efficient gene transfer into mammalian and nonmammalian cells. Proc Natl Acad Sci USA 1993; 90(17):8033-8037. 33. Gasmi M, Glynn J, Jin MJ et al. Requirements for efficient production and transduction of human immunodeficiency virus type 1-based vectors. J Virol 1999; 73(3):1828-1834. 34. Kim VN, Mitrophanous K, Kingsman SM et al. Minimal requirement for a lentivirus vector based on human immunodeficiency virus type 1. J Virol 1998; 72(1):811-816. 35. Trono D. HIV accessory proteins: Leading roles for the supporting cast. Cell 1995; 82(2):189-192. 36. Deacon NJ, Tsykin A, Solomon A et al. Genomic structure of an attenuated quasi species of HIV1 from a blood transfusion donor and recipients. Science 1995; 270(5238):988-991. 37. Baba TW, Liska V, Khimani AH et al. Live attenuated, multiply deleted simian immunodeficiency virus causes AIDS in infant and adult macaques. Nat Med 1999; 5(2):194-203. 38. Wyand MS, Manson KH, Garcia-Moll M et al. Vaccine protection by a triple deletion mutant of simian immunodeficiency virus. J Virol 1996; 70(6):3724-3733. 39. Desrosiers RC, Lifson JD, Gibbs JS et al. Identification of highly attenuated mutants of simian immunodeficiency virus. J Virol 1998; 72(2):1431-1437. 40. Chinnasamy D, Chinnasamy N, Enriquez MJ et al. Lentiviral-mediated gene transfer into human lymphocytes: role of HIV-1 accessory proteins. Blood 2000; 96(4):1309-1316. 41. Dull T, Zufferey R, Kelly M et al. A third-generation lentivirus vector with a conditional packaging system. J Virol 1998; 72(11):8463-8471. 42. Miyoshi H, Blomer U, Takahashi M et al. Development of a self-inactivating lentivirus vector. J Virol 1998; 72(10):8150-8157. 43. Zufferey R, Dull T, Mandel RJ et al. Self-inactivating lentivirus vector for safe and efficient in vivo gene delivery. J Virol 1998; 72(12):9873-80. 44. Yee JK, Moores JC, Jolly DJ et al. Gene expression from transcriptionally disabled retroviral vectors. Proc Natl Acad Sci USA 1987; 84(15):5197-201. 45. Yu SF, von Ruden T, Kantoff PW et al. Self-inactivating retroviral vectors designed for transfer of whole genes into mammalian cells. Proc Natl Acad Sci USA 1986; 83(10):3194-3198. 46. Kotsopoulou E, Kim VN, Kingsman AJ et al. A Rev-independent human immunodeficiency virus type 1 (HIV-1)-based vector that exploits a codon-optimized HIV-1 gag-pol gene. J Virol 2000; 74(10):4839-4852. 47. Maldarelli F, Martin MA, Strebel K. Identification of posttranscriptionally active inhibitory sequences in human immunodeficiency virus type 1 RNA: Novel level of gene regulation. J Virol 1991; 65(11):5732-5743. 48. Nasioulas G, Zolotukhin AS, Tabernero C et al. Elements distinct from human immunodeficiency virus type 1 splice sites are responsible for the Rev dependence of env mRNA. J Virol 1994; 68(5):2986-2993. 49. Olsen HS, Cochrane AW, Rosen C. Interaction of cellular factors with intragenic cis-acting repressive sequences within the HIV genome. Virology 1992; 191(2):709-715. 50. Schwartz S, Campbell M, Nasioulas G et al. Mutational inactivation of an inhibitory sequence in human immunodeficiency virus type 1 results in Rev-independent gag expression. J Virol 1992; 66(12):7176-7182. 51. Schwartz S, Felber BK, Pavlakis GN. Distinct RNA sequences in the gag region of human immunodeficiency virus type 1 decrease RNA stability and inhibit expression in the absence of Rev protein. J Virol 1992; 66(1):150-159.

Prospects for Gene Therapy Using HIV-Based Vectors

171

52. Haas J, Park EC, Seed B. Codon usage limitation in the expression of HIV-1 envelope glycoprotein. Curr Biol 1996; 6(3):315-324. 53. Cui Y, Iwakuma T, Chang LJ. Contributions of viral splice sites and cis-regulatory elements to lentivirus vector function. J Virol 1999; 73(7):6171-6176. 54. Baur AS, Sawai ET, Dazin P et al. HIV-1 Nef leads to inhibition or activation of T cells depending on its intracellular localization. Immunity 1994; 1(5):373-384. 55. Mahalingam S, MacDonald B, Ugen KE et al. In vitro and in vivo tumor growth suppression by HIV-1 Vpr. DNA Cell Biol 1997; 16(2):137-143. 56. Murphy KM, Sweet MJ, Ross IL et al. Effects of the tat and nef gene products of human immunodeficiency virus type 1 (HIV-1) on transcription controlled by the HIV-1 long terminal repeat and on cell growth in macrophages. J Virol 1993; 67(12):6956-6964. 57. Ryan-Graham MA, Peden KW. Both virus and host components are important for the manifestation of a Nef- phenotype in HIV-1 and HIV-2. Virology 1995; 213(1):158-168. 58. Stewart SA, Poon B, Jowett JB et al. Lentiviral delivery of HIV-1 Vpr protein induces apoptosis in transformed cells. Proc Natl Acad Sci USA 1999; 96(21):12039-12043. 59. Yao XJ, Mouland AJ, Subbramanian RA et al. Vpr stimulates viral expression and induces cell killing in human immunodeficiency virus type 1-infected dividing Jurkat T cells. J Virol 1998; 72(6):4686-4693. 60. Chen ST, Iida A, Guo L et al. Generation of packaging cell lines for pseudotyped retroviral vectors of the G protein of vesicular stomatitis virus by using a modified tetracycline inducible system. Proc Natl Acad Sci USA 1996; 93(19):10057-10062. 61. Ory DS, Neugeboren BA, Mulligan RC. A stable human-derived packaging cell line for production of high titer retrovirus/vesicular stomatitis virus G pseudotypes. Proc Natl Acad Sci U A 1996; 93(21):11400-11406. 62. Kafri T, van Praag H, Ouyang L et al. A packaging cell line for lentivirus vectors. J Virol 1999; 73(1):576-584. 63. Klages N, Zufferey R, Trono D. A stable system for the high-titer production of multiply attenuated lentiviral vectors. Mol Ther 2000; 2(2):170-176. 64. Xu K, Ma H, McCown TJ et al. Generation of a stable cell line producing high-titer self-inactivating lentiviral vectors. Mol Ther 2001; 3(1):97-104. 65. Hottinger AF, Azzouz M, Deglon N et al. Complete and long-term rescue of lesioned adult motoneurons by lentiviral-mediated expression of glial cell line-derived neurotrophic factor in the facial nucleus. J Neurosci 2000; 20(15):5587-5593. 66. Henderson CE, Phillips HS, Pollock RA et al. GDNF: A potent survival factor for motoneurons present in peripheral nerve and muscle. Science 1994; 266(5187):1062-1064. 67. Ferri CC, Moore FA, Bisby MA. Effects of facial nerve injury on mouse motoneurons lacking the p75 low-affinity neurotrophin receptor. J Neurobiol 1998; 34(1):1-9. 68. Kou SY, Chiu AY, Patterson PH. Differential regulation of motor neuron survival and choline acetyltransferase expression following axotomy. J Neurobiol 1995; 27(4):561-572. 69. Kordower JH, Emborg ME, Bloch J et al. Neurodegeneration prevented by lentiviral vector delivery of GDNF in primate models of Parkinson’s disease. Science 2000; 290(5492):767-773. 70. Emborg ME, Ma SY, Mufson EJ et al. Age-related declines in nigral neuronal function correlate with motor impairments in rhesus monkeys. J Comp Neurol 1998; 401(2):253-265. 71. Neufeld EF. Lysosomal storage diseases. Annu Rev Biochem 1991; 60:257-280. 72. Danciger M, Blaney J, Gao YQ et al. Mutations in the PDE6B gene in autosomal recessive retinitis pigmentosa. Genomics 1995; 30(1):1-7. 73. McLaughlin ME, Ehrhart TL, Berson EL et al. Mutation spectrum of the gene encoding the beta subunit of rod phosphodiesterase among patients with autosomal recessive retinitis pigmentosa. Proc Natl Acad Sci USA 1995; 92(8):3249-3253. 74. Pittler SJ, Baehr W. Identification of a nonsense mutation in the rod photoreceptor cGMP phosphodiesterase beta-subunit gene of the rd mouse. Proc Natl Acad Sci USA 1991; 88(19):8322-8326. 75. Suber ML, Pittler SJ, Qin N et al. Irish setter dogs affected with rod/cone dysplasia contain a nonsense mutation in the rod cGMP phosphodiesterase beta-subunit gene. Proc Natl Acad Sci USA 1993; 90(9):3968-3972.

172

Lentiviral Vector Systems for Gene Transfer

76. Takahashi M, Miyoshi H, Verma IM et al. Rescue from photoreceptor degeneration in the rd mouse by human immunodeficiency virus vector-mediated gene transfer. J Virol 1999; 73(9):78127816. 77. Havenga M, Hoogerbrugge P, Valerio D et al. Retroviral stem cell gene therapy. Stem Cells 1997; 15(3):162-179. 78. Kiem HP, Andrews RG, Morris J et al. Improved gene transfer into baboon marrow repopulating cells using recombinant human fibronectin fragment CH-296 in combination with interleukin-6, stem cell factor, FLT-3 ligand, and megakaryocyte growth and development factor. Blood 1998; 92(6):1878-1886. 79. Guenechea G, Gan OI, Inamitsu T et al. Transduction of human CD34+ CD38- bone marrow and cord blood-derived SCID-repopulating cells with third-generation lentiviral vectors. Mol Ther 2000; 1(6):566-573. 80. Haas DL, Case SS, Crooks GM et al. Critical factors influencing stable transduction of human CD34(+) cells with HIV-1-derived lentiviral vectors. Mol Ther 2000; 2(1):71-80. 81. Ramezani A, Hawley TS, Hawley RG. Lentiviral vectors for enhanced gene expression in human hematopoietic cells. Mol Ther 2000; 2(5):458-469. 82. Salmon P, Kindler V, Ducrey O et al. High-level transgene expression in human hematopoietic progenitors and differentiated blood lineages after transduction with improved lentiviral vectors. Blood 2000; 96(10):3392-3398. 83. Sirven A, Pflumio F, Zennou V et al. The human immunodeficiency virus type-1 central DNA flap is a crucial determinant for lentiviral vector nuclear import and gene transduction of human hematopoietic stem cells. Blood 2000; 96(13):4103-4110. 84. Woods NB, Fahlman C, Mikkola H et al. Lentiviral gene transfer into primary and secondary NOD/SCID repopulating cells. Blood 2000; 96(12):3725-3733. 85. Chang HK, Gendelman R, Lisziewicz J et al. Block of HIV-1 infection by a combination of antisense tat RNA and TAR decoys: A strategy for control of HIV-1. Gene Ther 1994; 1:208-216. 86. Kim JH, McLinden RJ, Mosca JD et al. Inhibition of HIV-1 replication by sense and antisense rev response elements in HIV-based retroviral vectors. J Acquir Immune Defic Syndr Hum Retrovirol 1996; 12:343-351. 87. Ho DD, Pomerantz RJ, Kaplan JC. Pathogenesis of infection with human immunodeficiency virus. N Engl J Med 1987; 317:278-286. 88. Buck HM, Koole LH, van Genderen MHP et al. Phosphate-methylated DNA aimed at HIV-1 RNA loops and integrated DNA inhibits viral infectivity. Science 1990; 248:208-212. 89. Lisziewicz J, Sun D, Klotman M et al. Specific inhibition of human immunodeficiency virus type 1 replication by antisense oligonucleotides. Proc Natl Acad Sci USA 1988; 89:11209-11213. 90. Sullenger BA, Gallardo HF, Ungers GE et al. Overexpression of TAR sequences renders cells resistant to human immunodeficiency virus replication. Cell 1990; 63:601-608. 91. Lisziewicz J, Peng B, Ensoli B et al. An autoregulated dual-function antitat gene for human immunodeficiency virus type 1 gene therapy. J Virol 1995; 69:206-212. 92. Rosenzweig M, Johnson RP, Lisziewicz J et al. Transduction of CD34+ hematopoietic progenitor cells with an anti-tat gene protects T-cell and macrophage progeny from AIDS virus infection. J Virol 1997; 71:2740-2746. 93. Morgan RA, Walker R. Gene therapy for AIDS using retroviral mediated gene transfer to deliver HIV-1 antisense TAR and transdominant Rev protein genes to syngeneic lymphocytes in HIV-1 infected identical twins. Hum Gene Ther 1996; 7:1281-1306. 94. Sarver N, Cantin EM, Chang PS et al. Ribozymes as potential anti-HIV-1 therapeutic agents. Science 1990; 247:1222-1225. 95. Ojwang JO , Hampel A , Looney DJ et al. Inhibition of human immunodeficiency virus type 1 expression by a hairpin ribozyme. Proc Natl Acad Sci USA 1992; 89:10802-10806. 96. Bai J, Gorantla S, Banda N et al. Characterization of anti-CCR5 ribozyme-transduced CD34+ hematopoietic progenitor cells in vivo. Molec Ther 2000; 1:244-254. 97. Bauer G, Kohn DB, Zaia JA et al. Inhibition of human immunodeficiency virus-1 (HIV-1) replication after transduction of granulocyte colony-stimulating factor-mobilized CD34+ cells from HIV1 infected donors using retroviral vectors containing anti-HIV-1 genes. Blood 1997; 89:2259.

Prospects for Gene Therapy Using HIV-Based Vectors

173

98. Woffendin C, Ranga U, Yang Z et al. Expression of a protective gene prolongs survival of Tlymphocytes in human immunodeficiency virus-infected patients. Proc Natl Acad Sci USA 1996; 93:2889-2894. 99. Ranga U, Woffendin C, Verma S et al. Enhanced T cell engraftment after retroviral delivery of an antiviral gene in HIV-infected individuals. Proc Natl Acad Sci (USA) 1998; 95:1201-1206. 100. Caputo A, Rossi C, Bozzini R et al. Studies on the effect of the combined expression of anti-tat and anti-rev genes on HIV-1 replication. Gene Ther 1997; 4:288-295. 101. Chinen J, Aguilar-Cordova E, Ng-Tang D et al. Protection of primary human T-cells from HIV1 infection by Trev: A transdominant fusion gene. Hum Gene Ther 1997; 8:861-868. 102. Smith SM, Markham RB, Jeang KT. Conditional reduction of human immunodeficiency virus type 1 replication by a gain-of-herpes simplex virus 1 thymidine kinase function. Proc Natl Acad Sci USA 1996; 93:7955-7960. 103. Curiel TJ, Cook DR, Wang Y et al. Long-term inhibition of clinical and laboratory human immunodeficiency virus strains in human T-cell lines containing an HIV-regulated diphtheria toxin A chain gene. Hum Gene Ther 1993; 4:741-747. 104. Schnell MJ, Johnson JE, Buonocore L et al. Construction of a novel virus that targets HIV-1infected cells and controls HIV-1 infection. Cell 1997; 90:849-857. 105. Mebatsion T, Finke S, Weiland F et al. A CXCR4/CD4 pseudotype rhabdovirus that selectively infects HIV-1 envelope protein-expressing cells. Cell 1997; 90:841-847. 106. Yang AG, Chen S, Yao C et al. Phenotypic knockout of HIV-1 type 1 chemokine coreceptor CCR-5 by intrakines as potential therapeutic approach for HIV-1 infection. Proc Natl Acad Sci USA 1997; 94:11567-11572. 107. Chen JD, Bai X, Yang AG et al. Inactiviation of HIV-1 chemokine co-receptor CXCR-4 by a novel intrakine strategy. Nat Med 1997; 3:1110-1116. 108. Yang AG, Zhang X, Torti FJ et al. Anti-HIV type 1 activity of wild-type and functional defective RANTES intrakine in primary human lymphocytes. Hum Gene Ther 1998; 9:2005-2018. 109. Buonocore L, Rose JK. Blockade of human immunodeficiency virus type 1 production in CD4+ T cells by an intracellular CD4 expressed under control of the viral long terminal repeat. Proc Natl Acad Sci USA 1993; 90:2695-2699. 110. Marasco WA, Haseltine WA, Chen S-Y. Design, intracellular expression, and activity of a human anti-human immunodeficiency virus type 1 gp120 single-chain antibody. Proc Natl Acad Sci USA 1993; 90:7889-7893. 111. Duan L, Bagasra O, Laughlin MA et al. Potent inhibition of human immunodeficiency virus type 1 replication by an intracellular anti-Rev single-chain antibody. Proc Natl Acad Sci (USA) 1994; 91:5075-5079. 112. Levin R, Mhashilkar AM, Dorfman T et al. Inhibition of early and late events of the HIV-1 replication cycle by cytoplasmic Fab intrabodies against the matrix protein, p17. Mol Med 1997; 3:96-110. 113. Rondon IJ, Marasco WA. Intracellular antibodies (intrabodies) for gene therapy of infectious diseases. Ann Rev Microbiol 1997; 51:257-283. 114. Levy-Mintz P, Duan L, Zhang H et al. Intracellular expression of single-chain variable fragments to inhibit early stages of the viral life cycle by targeting human immunodeficiency virus type 1 integrase. J Virol 1996; 70:8821-8832. 115. Mitsuya H, Yarchoan R, Broder S. Molecular targets for AIDS therapy. Science 1990; 249:1533-1543. 116. Embretson J, Zupancic M, Ribas JS et al. Massive covert infection of helper T lymphocytes and macrophages by HIV during the incubation period of AIDS. Nature 1993; 362:357-362. 117. Michael N, Morrow P, Mosca J et al. Induction of HIV-1 expression is associated with a shift in RNA splicing patterns. J Virol 1991; 65:1291-1303. 118. Palella JrF, Delaney KM, Moorman AC eal. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. N Engl J Med 1998; 338:853-860. 119. Gulick Rm, Mellors JW, Havlir D et al. Treatment with indinavir, zidovudine, and lamivudine in adults with human immunodeficiency virus infection and prior antiretroviral therapy. N Engl J Med 1997; 337:734-739.

120. Cavert W, Notermans DW, Staskus K et al. Kinetics of response in lymphoid tissues to antiretroviral therapy of HIV-1 infection. Science 1997; 276:960-964. 121. Wong JK, Gunthard HF, Havlir DV et al. Reduction of HIV-1 in blood and lymph nodes following potent antiretroviral therapy and the virologic correlates of treatment failure. Proc Natl Acad Sci USA 1997; 94:12574-12579. 122. Ho DD. Toward HIV eradication or remission: The tasks ahead. Science 1998; 280:1866-1867. 123. D’Arminio Monforte A, Testa L, Adorni E et al. Clinical outcome and predictive factors of failure of highly active antiretroviral therapy in antiretroviral-experienced patients in advanced stages of HIV-1 infection. AIDS 1998; 12:1631-1637. 124. Ledergerber B, Egger M, Opravil M et al. Clinical progression and virological failure on highly active antiretroviral therapy in HIV-1 patients: A prospective cohort study. Swiss Cohort Study. Lancet 1999; 353:863-868. 125. Aukrust P, Muller F, Lien E et al. Tumor necrosis factor (TNF) system levels in human immunodeficiency virus-infected patients during highly active antiretroviral therapy: Persistent TNF activation is associated with virologic and immunologic treatment failurehuman immunodeficiency virusinfected patients during highly . J Infect Dis 1999; 179:74-82. 126. Lucas GM, Chaisson RE, Moore RD. Highly active antiretroviral therapy in a large urban clinic: Risk factors for viroogic failure and adverse drug reactions. Ann Intern Med 1999; 131:81-87. 127. Wit FW, van Leeuwen R, Weverling GJ et al. Outcome and predictors of failure of highly active retroviral therapy: One year follow-up of a cohort of human immunodeficiency virus type 1-infected persons. J Infect Dis 1999; 179:790-798. 128. Evans TG, Bonnez W, Soucier HR et al. Highly active antiretroviral therapy results in a decrease in CD8+ T cell activation and preferential reconstitution of the peripheral CD4+ T cell population with memory rather than naive cells. Antiviral Res 1998; 39:163-173. 129. Bucy RP, Hockett RD, Derdeyn CA et al. Initial increase in blood CD4+ lymphocytes after HIV antiretroviral therapy reflects redistribution from lymphoid tissues. J Clin Invest 1999; 103:1391-1398. 130. Haase AT. Population biology of HIV-1 infection: viral and CD4+ T cell demographics and dynamics in lymphatic tissue. Ann Rev Immunol 1999; 17:625-656. 131. Zanussi S, Simonelli C, Bartolin MT et al. Immunological changes in peripheral blood and in lymphoid tissue after treatment of HIV-infected subjects with highly active anti-retroviral therapy (HAART) or HAART + IL-2. Cliin Exp Immunol 1999; 116:486-492. 132. Martinon F, Michelet C, Peguillet I et al. Persistent alterations in T-cell repertoire, cytokine and chemokine receptor gene expression after 1 year of highly active antiretroviral therapy. AIDS 1999; 13:185-194. 133. Finzi D, Hermankova M, Pierson T et al. Identification of a reservoir for HIV-1 in patients on highly active antiretroviral therapy. Science 1997; 278:1295-1300. 134. Natarajan V, Bosche M, Metcalf JA et al. HIV-1 replication in patients with undetectable plasma virus receiving HAART, HIghly active antiretroviral therapy. Lancet 1999; 353:119-120. 135. Chun TW, Engel D, Berrey MM et al. Early establishment of a pool of latently infected, resting CD4(+) T cells during primaryHIV-1 infection. Proc Natl Acad Sci USA 1998; 95:8869-8873. 136. Finzi D, Blankson J, Siliciano JD et al. Latent infection of CD4+ T cells provides a mechanism for lifelong persistence of HIV-1, even in patients on effective combination therapy. Nature Med 1999; 5:512-517. 137. Zhang H, Dornadula G, Beaumont M et al. Human immunodeficiency virus type 1 in the semen of men receiving highly active antiretorviral therapy. N Engl J Med 1998; 339:1803-1809. 138. Schapiro JM, Winters MA, Lawrence J et al. Clinical cross-resistance between the HIV-1 protease inhibitors saquinavir and indinavir and correlations with genotypic mutations. AIDS 1999; 13:359-365. 139. Loveday C, Devereux H, Huckett L et al. High prevalence of multiple drug resistance mutations in a UK HIV/AIDS patient population. AIDS 1999; 13:627-628. 140. Vandamme AM, Van Laethem K, DeClerq E. Managing resistance to anti-HIV drugs: An important consideration for effective disease management. Drugs 199; 57:337-361. 141. Jimenez-Nacher I, Rodriguez-Rosado R, Anton P et al. Virological failure and adherence to antiretroviral therapy in HIV-infected patients. Int Conf AIDS 1998; 12:591, 32350a

CHAPTER 10

Ethical Considerations in the Use of Lentiviral Vectors for Genetic Transfer Ina Roy

Abstract

T

his chapter will outline the various concerns which have been raised in scientific, bioethics, and lay communities about the use of lentiviral vectors for purposes of gene therapy. Many of these concerns are ranged around gene therapy itself; others are concerns particular to using this sort of vector for genetic modification of human cells. These concerns are outlined within the chapter, and arguments are given in favor and against various approaches to these concerns. Lastly, it is noted throughout that at this stage of research into gene therapy, the most practical approach to these dilemmas is to maintain awareness of the ethical problems and provide information to those concerned with all aspects of the development of this set of technologies.

Introduction The purpose of this Chapter is to discuss the various ethical considerations that are raised by the use of lentiviral vectors in gene therapy. As will become clear through the course of the Chapter, there are many more questions than can be answered; the role of any writings on the bioethical implications of new technologies is to raise such questions so that we are aware of them and give them due attention. Solutions to such dilemmas will grow and develop over time, as the technology advances; in the meantime, the bare outlines of certain solutions will be presented with an eye toward future work. There are a number of axes along which the ethical issues associated with clinical lentiviral vector use can be divided. For the purposes of this Chapter, the ethical issues associated with this technology (or rather, set of technologies) will be divided into two types: ethical concerns about the nature of the application of these and concerns regarding the results of any use of the above uses of exogenous genetic introduction. Within each category, the ethical considerations about these technologies will reflect the values that might be held with respect to such fundamental human concerns as individual health, public health, scientific knowledge, and privacy. The conflict between the values we hold regarding each of these is the source for many of these dilemmas, which will become apparent through the course of this Chapter. For the purposes of this discussion, we will limit the uses considered to those in which genetic transfer is involved. There are three categories of cell types to which genetic transfer may be done: 1. Genetic transfer may be targeted to specific cell types within an adult organism. For example, experimentation has been done for transfer of genes exclusively to liver cells, leukocytes,

Lentiviral Vector Systems for Gene Transfer, edited by Gary L. Buchschacher, Jr. ©2003 Eurekah.com.

Lentiviral Vector Systems for Gene Transfer

176

or retinal cells. In each case, the genetic introduction is intended to replace or enhance existing function of the particular cell in question. 2. Genetic transfer may be widespread, so that a wide variety of cells within the adult body receive or are targeted to receive the exogenous gene carried by the vector. 3. Exogenous genes may be introduced into cells of an embryo, in a manner that results in will acquire a copy of this exogenous gene. Of note, this will include cells which will form the gametes possessed by the future adult, thus allowing for the possibility of transfer of the new genetic material to future generations.

Transfer types 1 and 2 are sometimes known as Somatic Cell Gene Therapy, SCGT from here on. SCGT has been used as an acronym for a number of things, including “GT Sports Cars,” in professional auto racing. More relevant to our purposes here is the use of SCGT as an acronym for Somatic Cell Gene Transfer, which describes the process by which Dolly the sheep was created, since the entire genome of a somatic cell is transferred into an ovum. This acronym is used1 in reference to cloning; please note the cloning process is not to be confused with the process we are discussing in this Chapter. Type 3 is sometimes known as Germ Line Gene Therapy (GLGT). The distinction between these two different processes, SCGT and GLGT, has been explored by a number of different authors, including Philip Kitcher.2 Please note that both the ethical relevance of this distinction and the distinction itself have been called into question, by ethicists including Roy Moseley;3 we will not therefore use it as anything more than a method for organizing the discussion in this Chapter.

Gene Therapy: Potentially Problematic Types of Application Somatic Cell Gene Therapy(SCGT) Somatic Cell Gene Therapy, Problem 1: Treatment Versus Enhancement Let us begin by looking at potential uses for SCGT. Many of the uses currently being tested in the research context involve the treatment of genetic diseases, including hemoglobinopathies, cystic fibrosis, and macular degeneration. Other potential uses include targeting cancer or tumor cells with genes which would either “turn off” tumorogenic genes within the cell or themselves are translated into cytotoxic products. Such treatments are currently being tested on gliomas and a number of gastrointestinal carcinomas. On the face of it, such treatments would seem ethically unproblematic. Many of the ethical concerns regarding SCGT, however, arise in connection with the possibility that genetic material could be introduced into somatic cells for the purposes of enhancement rather than therapy. Enhancements might be understood as genetic modifications used not to treat a deficiency or raise a patient’s level of health, but rather for the purposes of intensifying some quality in an already healthy individual. In surveying a population, it became clear that while the public is in general in favor of treatment of genetic diseases through gene therapy, it is by no means comfortable with the idea of using these same technologies for enhancement purposes; for example, a piece included in the Washington Post4 expresses public fears about the “thorny” problem of the possibility of genetic enhancement as an outgrowth of technology used for genetic therapy. The nightmarish visions that arise are legion—scenarios in which we enhance the traits of children so that they grow taller than they would be expected to “naturally” so that they can be trained to be basketball players, or genes that are inserted specifically into the liver so that someone can enjoy a larger amount of liquor without long term side-effects—and have no comparable voice within the realm of cosmetic surgery.

Ethical Considerations in the Use of Lentiviral Vectors for Genetic Transfer

177

Before we can address the concerns about genetic enhancement, however, we must address two foundational questions, the second dependent upon the answer to the first. The first: is it even possible to separate enhancements in principle from therapies? How would we delineate or define the demarcation between these two practices? If this can in fact be done, then we need to consider a second question which has been posed by a number of ethicists and scientists:5,6 what, if anything, justifies our considering genetic enhancements but not treatments morally problematic? An analogous division between therapy and enhancement might be helpful in determining whether we can differentiate the two. One possible analogy can be made to the division between medically necessary surgery and surgery that is considered “cosmetic.” Some surgeries (e.g., surgery to stop an esophageal hemorrhage) are medically necessary; others are cosmetic (e.g., blephoroplasty), and still others may fall within a grey zone, being viewed as both cosmetic and medically valuable (e.g., breast reduction). This analogy has its benefits, but it cannot do all the necessary ethical work for us. In part, this is because the distinction between cosmesis and medically necessary surgery is often unclear, as indicated by the fact that a grey area is acknowledged; where the divide falls is sometimes a matter of the particular case of the patient. Blephoroplasty for example could be considered medically necessary, when eyelids interfere with vision. As we will see, this same problem of definition appears when trying to distinguish enhancement from therapy. On the face of it, it seems that genetic diseases would be easily defined, and thus, anything that eliminates or reduces disease processes might be considered to be a therapy, where other genetic modifications could be considered to be enhancements and therefore available for moral critique. “Disease” and “illness” have negative conotations, compared to the more neutral “condition”; if a healthy organism is one that is functioning to its full capacity, as an ideal of members of its species, then diseases are those conditions which prevent it from so functioning. Diseases are connected with the values of human beings, in that disease can be defined as a condition which detracts from one of the human goods which is considered to be primary: health. We could say that a person has a disease if she has a condition that detracts from her good health. Using this definition, genetic diseases are easily distinguishable from other genetically based non-disease traits. Sickle cell anemia is a condition which detracts from the health of the organism; if a healthy human being should be able to walk, run, and breathe without difficulty in a variety of different situations then clearly, sickle cell anemia is a disease because people who have this condition cannot meet these basic criteria for health. Given its genetic basis, sickle cell anemia is a genetic disease. And therefore the introduction of exogenous genetic material that reduces or bypasses the problematic genes carried by someone with sickle cell anemia should be viewed as a cure or therapy for this disease. But what is it exactly for a human being to be healthy? Is it a mere biological or physical quality? The WHO defines health as “a complete state of physical, mental, and social wellbeing and not merely the absence of disease or infirmity.” Moreover, while we might be tempted to separate the physical part of health as the most important portion where the psychosocial health of a person to be secondary, and therefore of lesser consideration than the physical health of the person, it is not clear that this move is justified. First, the mind is the product of a physical entity namely the brain. Moreover, it is clear that psychosocial health has a very direct impact on the physiological functioning of a given human organism, and vice versa. For example, the context in which someone lives—for example, in a city without adequate means of public health protections such as clean water supplies and sanitation systems—may influence how susceptible someone is to a variety of diseases ranging from tuberculosis to measles to sickling disease. And diseases we once thought of as purely psycho-socially derived and extended (e.g., Down syndrome and schizophrenia) have subsequently been found to have biological bases.

178

Lentiviral Vector Systems for Gene Transfer

How does the problematic attempt to divide physical from mental and social health affect the consideration of SCGT as an ethical entity? Without being able to distinguish these two, it’s not clear that things that seem on the face of it to be “enhancements” can in fact be easily identified (for a detailed discussion of many of the issues here, see ref. 8). We may see someone who chooses to have a treatment with the HGH (human growth hormone) gene as being vain. The modification would seem to fall in the realm of enhancement, but it could be argued that such a person is actually contributing to his or her own health by ensuring that the success associated with taller persons in certain western societies—in turn increasing the odds of economic success and security that are correlated with higher level of health, lower stress and longer life. Given the fact that we cannot easily or clearly separate therapeutic projects from those which are primarily intended for enhancement, it behooves us to question what grounds we can have for condemning genetic enhancements as a class. The root issue underlying the concerns over genetic enhancements may be the value that is placed on personal identity. One may identify oneself with one’s illness, but psychological acceptance of an illness is not felt to preclude removal of that illness without harm to one’s self-identity and individuality. Treatment of long-term chronic illnesses requires for example that someone rebuild a core identity that is not centered around illness, on the assumption that having such an independent source of selfhood provides a healthier and more fruitful backdrop to human existence. Public perception of genetics includes a very close tie between genetic and self-identity; enhancement may appear to be changing the very nature or essence of the person, those traits which make him unique (through the color of his hands or ability to play piano). This is a rather undefined notion of selfhood, but one that nonetheless pervades day-to-day thinking about genetics. An appropriate ethical approach to concerns over enhancement requires us to acknowledge this underlying concern about identity without giving credence to simple equations of personal and genetic identity. One possibility for grappling with this issue is to question the usefulness of creating a division between therapy and enhancement. Perhaps rather we can make some decisions as to what constitutes a healthy life —understanding that this may be subject to change as our understanding of human biology and of cultural variability change—and try to use genetic therapies to bring the level of persons to the minimum standard of health. Of course, this would require a rather broader revamping of our entire health care system, as well our current system for providing contributions to the health and welfare of other nations, a project the discussion of which is well beyond the scope of this Chapter.

Somatic Cell Gene Therapy, Problem 2: Eugenics Projects Eugenics projects have, in a variety of formal and less formal ways, been part of human groupings throughout recorded history. Literally “eugenic” means “good life” from the Greek “eugenes.” Plato’s Republic asks for the ideal state to engage explicitly in breeding projects such that those persons with the most noble or socially valuable traits would have more children, while those with less valuable or more problematic traits would not breed—ensuring an eventual production of a highly functional society with healthy, intelligent and talented citizens. Possibly the most well known example is the explicitly eugenicist project that formed a central part of the Hitler-Nazi party agenda. It is important to note that eugenics projects have not been limited to Nazi Germany; the US made an explicit commitment in the 1930s to sterilization programs for those deemed “unfit” or “imbecilic”.9 In modern interpretation, a eugenics project is one which certain traits are removed from a species through some combination selective breeding and population reduction, with an understanding that the traits upon which we focus are those that are heritable by genetic means. These projects have been criticized on two fundamentally different but equally important ethical grounds. First, these projects do not take into account the inherent value of all persons in

Ethical Considerations in the Use of Lentiviral Vectors for Genetic Transfer

179

that they are persons. As long as one is wedded to some sort of democratic ideal or idea of human rights, it is likely that eugenics projects will be found morally or ethically repugnant, simply because the wants and desires (to have a family, to live a full life-span) of certain persons (for example those who are deemed insufficiently intelligent) will be overridden by the needs of society as a whole. A corollary complaint has been made by, among others, feminist and post-modern scholars: that the determination of what traits are worthy of being continued is made from a particular perspective – often that of those in power or those in the majority – but this perspective may not be the sole or the best view of what is best for society. This corollary provides a nice segue into the second type of criticism of eugenics projects, namely that these projects are merely vehicles for the perpetuation of prejudiced and unfairly determined existing beliefs: the Nazis for example associated purity with non-Jewish characteristics, and created evidence suggesting that persons of the Jewish religion and culture were inferior in terms of intelligence, character, and ability. The “extermination” of the Jewish persons of Germany was a project considered on par with the extermination of persons who had Downs’ syndrome and other biological forms of mental retardation. Hidden underneath the agenda that accompanies Nazi-style eugenics project is a directed and biasing search for evidence that the traits purported to belong to the denigrated group have a biological basis. In other words, the process by which eugenics projects are found to be morally egregious is two-fold: first, there is a created correlation between certain undesirable traits and certain groups of persons previously held to be undesirable; second, there is a created correlation between those undesirable traits and a biological or heritable basis for these traits (again, S.J. Gould has an excellent discussion of the historical events considered here). It is unsurprising that the connection has been made between genetic engineering and eugenics projects. It would seem that because SCGT is not directed toward long term changes in genetic lines (directed rather toward changing individuals, rather than whole ancestral lines) that the worries associated with eugenics projects would not arise. Nonetheless, versions of both the concerns raised in the above two paragraphs have come up with respect to SCGT. First, concerns have been raised, by such groups as the American Disabled Person’s Association that large scale attempts to eliminate genetic diseases, not from the human gene pool, but from expression in the phenotype pool of human beings might a) create a genetic underclass of those who cannot afford such treatments b) will incur or exacerbate existing prejudices against those with certain sort of genetically based disabilities by making those persons with the disability apparent rarer and therefore more noticeable and c) as a result devalue the lives, activities and experiences with those who have and either cannot or choose not to have their illness treated. It might be possible that a universalized health care system could ensure that the first two concerns could be eliminated by making somatic cell treatments of genetically based diseases available to all persons. This should strike the reader as rather unlikely, since so few nations, even those with healthy economic structures, are willing or able to to provide such health care assurance for those within their borders. Even imagining this to be possible, however, it is not clear that concern underlying both of those two worries is actually being addressed. In some sense, it is inevitable that the labeling of a condition as a disease—even the implicit labeling that is hidden under the fact that something has a “treatment “ available—will result in some amount of negative feeling about the condition itself. And while most of us cannot imagine life with the more severe genetic conditions, let alone imagine choosing to live such a life, the pressure not to live with such an illness is heightened by the knowledge that a treatment is readily available, especially if it eliminates all traces of the disease, fitting a person into the “norm” of existence without needing special medications or other technical supports. Is it possible for us to have continued concern for those with genetically based disabilities while simultaneously trying to eliminate those genotypes from the human population?

180

Lentiviral Vector Systems for Gene Transfer

This question gives rise to the last potential concern: that widespread use of SCGT will devalue the lives of those who will not or cannot receive SCGT treatments. Religious or cultural prohibitions of such treatment may be frowned upon, and the resulting devaluation of the persons who continue to have the disease prevent acknowledging the value that this disease may have for the persons or persons involved. Moreover, it is unclear, say the objectors, that the devaluing of persons who do not or cannot be treated will only be expressed through derision or social ostracism; the devaluing of persons with certain traits began, in SS Germany, with a simple project which entailed the removal of those felt to be genetically unfit to another location (in particular, an attempt was made to purchase the island of Madagascar for this purpose), soon moved into the systematized and brutal killing and torture of those same individuals with little or no institutional respect for the humanity or the simple life process of those persons whom the system had devalued. Moreover, what begun as an attempt to eliminate only those with clearly genetic conditions (which in and of itself can be seen as morally repugnant), the attitude soon progressed until anyone could be included under the “unfit” umbrella— Slavs, Catholics, those who were gay, those with even minor mental illnesses. We should not approach the overhangs of these slippery-slopes—from disease to devaluing, from those with a disease to all of those who are perceived as different—for fear that we will be unable to stop a headlong and rapid descent to a hierarchical, inhumane society. It can and has been argued that in fact no such project could even be contemplated in this day and age of great scientific knowledge and increasing reign of democratic ideals, that previous repugnant eugenics projects were enacted in scientifically questionable times and difficult political climates, and that these sorts of eugenics projects must be separated from the benign attempt to give every person the opportunity to live a healthy and pain-free life. But even acknowledging the hope that we have learned our lessons and learned them well, we must recognize that the worries raised represent a number of values that are being called into question by the practices of SCGT. First one must consider the value of an individual qua human being or person, regardless of the particular form that that life may take. Second that of the value of choice, that a person should not feel pressured or obligated to live a life other than what they choose, provided that it does not harm another. Those values are being weighed, in these considerations, against others—the value of health, the economic burden placed on health care systems and the public at large in the care of those with genetic illnesses, the non-economic burdens of time and other non-tangible resources for the adaptation of culture to those with differences. How to weigh these values against one another—or to change the weights of these various factors through institutional and legal means—is a question to which we will return at the end of this Chapter.

Somatic Cell Gene Therapy, Problem 3: Unintended Germ Line Gene Transfers The possibility exists, especially in the early stages of development of lentiviral gene transfers, that cells other than those that are the intended target will receive copies of the gene. Of particular concern is the possibility that germ cells, not targeted by the vector, will accidentally become the recipients of exogenous genetic material, thereby allowing for the transfer of said genetic materials beyond the intended recipient to future generations. In such cases, the questions discussed in the next section will arise as well in the case of unintended GLGT.

Germ Line Gene Therapy (GLGT) While SCGT raises concerns in the general public, this outcry is quite limited relative to the concerns raised by the prospect of GLGT. There are a number of potential sources of concern with regard to GLGT that should be explored, and their ramifications outlined in detail.

Ethical Considerations in the Use of Lentiviral Vectors for Genetic Transfer

181

Germ Line Gene Therapy, Problem 1: Eugenics Projects, Continued If SGCT might be considered to be a means to enacting eugenics projects, how much more of a concern will be raised by the prospect of GLGT? The nature of what might be considered to be exemplar eugenics projects involves two related subprojects: first, the reduction of a certain traits in a population (through either reduction of breeding of those carrying the traits, or the elimination of those persons carrying those genes) and second, increasing the presence of those traits considered to be “better” or more desirable traits through increased breeding of those who carry the trait, increasing the number of persons who might breed carry the trait, or selective creation of embryos (say, through selection of IVF created embryos) with that trait. GLGT would be far more efficient and direct mechanism for promotion of a eugenics project than the sorts of increases that would be promoted through the use of SCGT. Widescale implementation of SCGT programs would contribute to eugenics projects through indirect means: creating or supporting the beliefs that certain traits are not just problematic for the persons carrying them, but in some general sense undesirable for the population at large. GLGT will allow many of those beliefs to be made concrete and to be played out through genetic modification. It might be a further matter for concern that the determination of what is healthy or ideal in a human being is made by persons who are imposing their own standards upon others, and may in particular hold a set of racist, sexist, or classist beliefs. Moreover, some new concerns appear which—while general concerns about eugenics projects – become particularly acute in the fact of the ability to create human beings who have certain genes from the moment of birth. Among these concerns: the question of what happens to populations should they be streamlined genetically. The potential narrowing of bio(genetic) diversity in the population continues to be a troublesome one. In other animal populations which have had a drastic reduction of genetic diversity—the problems of the diminished cheetah population being a prime example—an increase in apparent recessive traits (often of a harmful kind) and a reduction of population size as a result is a worrisome one. Not only might genetic drift of this kind increase the appearance of certain traits but also the genes eliminated may later have an unknown benefit. These benefits may not be apparent at a given time either because we have not looked for benefits of that kind or because those benefits will not be apparent in our current environment but in other environmental conditions which we may encounter in the future. In the former category falls the banner case of the sickling allele of the hemoglobin β gene. It was believed that to be a carrier (rather than a homozygote recessive) for that variant was not only neutral, but even potentially harmful (persons had been thought to be more susceptible to sickling in high altitudes though there is no substantive evidence to support this concern). It was discovered however that in fact not only does the trait not cause harm, but those who are heterozygous for this trait have some resistance to infection by the set of parasites which can cause malaria by preventing residence of those parasites within red blood cells.10 With the recent publication of the results of the human genome project and the surprising revelation that humans have far fewer genes (apparently) than had previously thought (less than 1/3 of the original expectation of 100,000), we must be more aware than ever that the simple gene-protein equation under which we have been laboring is not the correct one. Genes are likely to play a more complex role in the final determination of organismal traits that we had previously thought; moreover, the interactions between genes and environment is probably much more complex than we had previously imagined. Eugenics projects—particularly those with a streamlining effect—may have long-term effects on the human population’s ability to adapt to unknown conditions of the future.

182

Lentiviral Vector Systems for Gene Transfer

Germ Line Gene Therapy, Problem 2: Rights of Future Generations Another issue of concern is that of the limitation of choices of future generations. By performing GLGT upon current members of our population, we limit the choices available to those of future generations Several philosophers and ethicists11 have argued that we have to consider the rights of those individuals—including the rights of free choice. Given this, should not each person be allowed to decide for him or herself whether he or she would prefer to live with a particular trait? This argument however is subject to several criticisms. First, it can be countered that in fact we constantly make decisions for and on behalf of future generations—by doing things as simple as using some technologies over others, using the environment in particular ways, building and planning countries, governments and other institutions with the hope that these things will be of value to those in the future. Why biology, and more particularly genetics, should be given some premier status of concern is unclear. One would have to say that genetic traits in particular are more important to what we are and more limiting than the choices, say, that we make about protecting certain types of land, or certain types of environmental resources. Even if we were to accept the argument that in fact this information is more valuable, a secondary objection arises. If in fact we should not make decisions for future generations, why is it that we should not be compelled to engage in completely random breeding? As it stands, we are allowed to breed with whom we will and often our choices are far from arbitrary. The stereotypical discussion of a young unmarried couple in which the possible eye and hair color of future children is discussed exemplifies the choices that we make in favor of our future progeny. Much of the unwillingness to allow interbreeding between races is argued in terms of not mixing traits that have formerly been associated with a particular racial group. This is not an argument which necessarily condones these sorts of discriminations; however, it does require us to think again about whether or not our arguments against GLGT are based on an arbitrary piece of line-drawing between certain kinds of “acceptable” selection on behalf or our offspring and future generations and those that are considered “unacceptable.” The end result is that we cannot argue consistently that future generations should be allowed to decide what traits they will and will not have. We make decisions about what future generations will retain as their genetic heritage regularly; what we must do now is to make these decisions carefully in light of our increasing capacity to regulate technologically the outcome of those decisions.

General Ethical Considerations: Research and Clinical Settings Previous Chapters have discussed in detail both the biological bases behind the use of lentiviral vectors for transfer or transmission of exogenous genetic material into a cell and potential uses of these technologies. These technologies are currently in their infancy, so in order to fully assess their use, we must consider them as technologies under assessment. The denotation of SCGT trials as “gene therapy” is for this reason ethically problematic. In presenting clinical trial information to potential research subjects, we must be concerned with the possibility that people misunderstand the nature of a trial, particularly when it is called a “therapy”.12 In cases where SCGT has been shown to have benefits to humans, rather than merely not harming them, calling the process “therapy” seems less problematic (though not completely unproblematic). But in those cases where the procedure has not been shown to be beneficial, or has not even been shown to be benign with respect to human health, calling the practice “gene therapy” is misleading. The term implicitly promises the possibility of help from this procedure, above and beyond the benefits of additional disease monitoring and placebo effect options. Misleading patients for the sake of research is a contentious issue within most of medical practice. On the one hand, there are a number of temptations to even the most ethical researcher to present incomplete of misleading information regarding the nature of the trial to be

Ethical Considerations in the Use of Lentiviral Vectors for Genetic Transfer

183

attempted. One may wish or need to increase enrollment in trials so that a potentially valuable treatment will have a chance to be approved for eventual. Giving hope to a patient may increase his lifespan independent of the actual value of the treatment. In addition, researchers may tell themselves (as well as the subjects) that this procedure must be helpful to reduce the cognitive dissonance of putting patients through difficult protocols with little benefit to either patient or the researcher herself. And it might be thought that in the context of something as minor as a terminological difficulty, such linguistic difficulties could be overcome with careful explanations on the part of researchers in the course of providing information during the consent process. But does the use of such preventative measures justify the “therapeutic misconception” (the term first used by Churchill; see ref.12) with respect to SCGT? Raised expectations give rise to inevitable disappointed hopes—moreover, US research process acknowledges the high value of personal choice when in engaging in any project and particularly where engaging in the project might result in a harms to the person agreeing to a role in the project. We cannot of course guarantee that a person will come to fully accept the possibility of risk or the understanding (particularly with phase I trials) that there is no expected benefit to this trial, but we can ensure that we can provide information such that it is at least possible for someone to come to this understanding. Provided that such information is available, and that the terminology of “gene therapy” is either discarded or guarded in a way that mitigates the potentially misleading effect of this term, we can then turn to the question of what ethical issues are involved in SCGT in the research context itself.

Research Considerations One approach to appreciating and identifying ethical dilemmas within the research context is to look at the risks and benefits involved in being a research subject in an SCGT trial. The safety considerations mentioned in previous Chapters include those that are valuable to the subject enrolling in a trial. In the case of early trials, the benefits include increased monitoring, and possibly access to testing and maintenance that might not otherwise be available to the patient. It is not expected that the subject will receive benefit that might accrue as a result of treatment, because the treatment is not being given at a therapeutic level. In later phases of clinical trials, of course, these benefits are added. In addition to the particular benefits that accrue to the subject, there are potential benefits to society at large. Treatment trials may eventually lead to discoveries that will improve the quality of life for a number of persons and possibly provide treatments for future generations, which will increase the public health and decrease public expenditure for long-term health care. But these benefits must be weighed against the potential harms accrue to research subjects. One of the most difficult parts of weighing these harms is that when a practice is still in its infancy, still in the process of being developed, we are unsure of what those harms will be. While we can extend biological theory to predict where these harms might appear—in what form and through what causes—and suggest some preventative measures to prevent such harms from occurring, such prediction is not guaranteed to be accurate; we must therefore keep in mind that some of the harms are going to be of necessity unknowns. We can—and many persons, both scientists and lay persons persons have already done so—speculate on the sorts of harms that are likely to result in the course of research into exogenous genetic introductions. Lentiviral vector present particular, though not all are warranted, concerns. As discussed in previous Chapters, lentiviral vectors are modified versions of or recreated mimics of lentiviruses. The advantages of using these sorts of vectors (as opposed to artificial means of genetic introduction) include the adaptation of these entities to finding selected cell types within the human body and moreover to introduce foreign material into human cells.

184

Lentiviral Vector Systems for Gene Transfer

This very adaptation, however, makes concerns about lentiviral vector use come to the fore. Therefore, the viruses are modified in any of a number of ways to reduce the likelihood that an infection (such as an HIV infection) will occur. These modifications include: • Removal of the genetic elements that code for proteins required for replication from the genetic code of the virus, thus rendering the virus avirulent (“crippling”) • Combining outer coat proteins (to retain the specific targeting ability of retroviruses) with more benign viral genetic material which codes for proteins that will still assist in the integration of the exogenous DNA into the cell’s genome. • Replacing certain replicatory elements of the genome of the lentivirus with elements from other viral types.

These techniques are felt to reduce the chances that an otherwise potentially deadly virus will retain the potential to infect host cells. But a number of troubling scenarios remain at least in imagination if not in actuality; when looking at a benefits-harms analysis of the consequences as an aid to determination of the morality of such research and practice, these harms will have small weight based on probability of occurrence, but cannot be ignored because the consequences of such an unlikely event could be devastating. What are these scenarios? The most common is the possibility that an HIV-1 or -2 vector, even crippled, could retain sufficient capacity for infection to create an iatrogenic case of AIDS. Historically, a similar concern was raised about “killed” or “crippled” viral and bacterial vaccines. Most of these concerns have been shown to be without foundation, once the technologies involved in disarming the organisms used in such vaccines were perfected. This particular concern comes up with respect to modified HIV vectors which might be introduced into a therapy recipient who is already infected with a wild-type HIV. The “crippled” (attenuated) virus is nonfunctional because some of its replicatory material is missing; should the genome be placed in a cell which contains the genome of a complete provirus, the possibility that it will replicate within the person concerned is a serious one. It could be argued that it would not be a concern, given that the person is already infected with HIV; however, there would still remain two problems. First, the newly formed virus may be more virulent than either of its genomic progenitors because of the addition of the exogeneous DNA. Second, if the vector is derived from a different strain of virus than the wild-type being encountered, the resulting virus may be a more virulent or infectious strain than either progenitor, for example, one that has a shorter latent period, or a strain in which new modes of transmission are introduced. The possibility of such accidental modifications does seem unlikely, but is always possible. If the vector is introduced into the body, it may encounter other naturally occurring viruses and proviruses, leaving open the possibility of such modification through recombination or transcomplementation. This set of problems is not limited to the moment of introduction into the host for whom the gene therapy is targeted. Prior to either in or ex vivo introduction of the vector, the vector must be grown in sufficient quantity to allow for use in a host cell (the more vector that is introduced, the better the likelihood of uptake by target cells). Vectors generally have “packaging” genes removed, to better eliminate the risk that the virus could reproduce as a wild-type. The packaging gene is often added as a plasmid to the cells which play “host” during the massive reproduction of these vectors. There is however evidence that reversion to wildtype—in particular the acquisition of the packaging gene—may occur in the vector production process.13 Any of these risks could be greatly reduced by carefully checking modified vector sources and screening persons who will receive the vector, particularly if and when these vectors are being produced in commercial volumes for therapeutic rather than clinical trial use. A concern essentially similar to those raised in the previous paragraph stems from the use of CMV or other virus (e.g., Rous sarcoma virus) promotor sequences in conjunction with the exogenous genes which have been introduced into the vector. These are used to promote the continued production of the introduced protein once the exogenous genetic material has been

Ethical Considerations in the Use of Lentiviral Vectors for Genetic Transfer

185

introduced into the human cell in question, whether in or ex vivo. These promotor regions are valuable in that they allow for continuous expression of the gene in question. Potentially, however, this gene could be inserted into a point in the host chromosome that increases production of a protein product that can be harmful in large amounts, sometimes called “overexpression.” Accumulation of certain types of protein products can be very dangerous for an organism that naturally produces only limited amounts of the protein, or large amounts only at a particular point in its lifecycle. Furthermore, introduction of exogenous DNA into the promotor region of an endogenous gene might interrupt (and thus disrupt) the activity of a tumor suppressor gene. There are a number potential problems that are similar to those raised above with respect to the introduction of genetic material itself once it has been introduced into the trial subject or patient. Even without the use of specialized promotor regions from non-lentiviruses, there is the possibility of introduction of the exogenous genomic material into the regulatory sites of other genes, resulting in unintended increased or decreased production of other naturally occurring proteins in some cells or for decreased cell growth regulation and cancer. In addition, exogenous DNA may be inserted into unintended cell types where the products of the introduced gene may interact unfavorably with products produced in that particular cell. These concerns might be reduced through technologies being developed to introduce not viral vectors into the host body, but cells which have been modified ex vivo using viral vectors (work of this type has been begun with retroviruses, including HIV and Moloney murine leukemia virus, for transduction of new genetic material ex vivo into CD4 cells; reviewed in ref. 14). These cells, which will produce the protein that is currently missing or non-functional in the patient, can be separated from any free viral particles, screened for appropriate protein production, and then be introduced into the recipient’s body. This procedure is one similar to the one currently used to treat children with adenosine deaminase (ADA) deficiency using stem cells obtained from umbilical cord blood, and in general would be ideal treatments for single-gene diseases (for example, hemophilia). This treatment however comes with risks as well; it may be the case that cells with these modifications might cause a severe rejection reaction, limiting their potential use. There is also the possibility of inflammation or immunological/allergic reaction, in particular with repeated introduction of the viral vector. The last concern which should be mentioned in passing if only because of the nebulous nature of the problems it raises. The further concern is that the non-human vectors (such as FIV) might have the potential for species crossover. The possibility exists that these vectors, when modified, will—in addition to being better purveyors of the exogenous genetic material—be modified to allow for better infection of human beings. Given the fact that these vectors are also chosen for their generic ability to cross over to human cells, and given the increasing awareness of receptor similarity between persons and other animals, the possibilities exist that these will also prove devastating to human animals with modification. One of the great difficulties in engaging in risk-benefit analysis in the ethical context is that we often do not know what harms we may be giving subjects when they are part of research projects – in fact, one of the reasons for engaging in research is to determine those risks. While animal studies have value in this regard, those studies are often not applicable to the human being. Moreover, in this particular case, research has to be considered to be human because diseases that are particular to human beings may or may not have naturally existing non-human counterparts. Induced artificial versions of these illnesses in other organisms may not provide an adequate parallel to those illnesses that are naturally occurring in human beings. So, many of the risks that may be taken on by those treated by gene therapies are still unknown. How are we to understand the relative value of the unknown risk? One possibility is to weigh the risks of worst-case scenarios given our current knowledge against the benefits of treatment. Those worst-case scenarios would be created on the basis of existing knowledge of human physiology, viral function, and the disease in question. This method however does not

186

Lentiviral Vector Systems for Gene Transfer

accommodate certain core values that surround medicine. First, protection of autonomy and freedom of choice are primary values in the health-care context; it cannot be assumed that a decision can be made for individuals on the basis of worst-case scenarios. In other words, we must acknowledge the possibility that each individual will weigh the risks identified in a worstcase scenario differently and thus make different decisions about enrolling in clinical trials and embarking on therapy; while some persons may be unwilling to take such risks, especially at the clinical trial stage, others may feel that the risk of death through the treatment is preferable to risk of death from disease. Second, we value altruism for its own sake; in the research context, it might be said that such risks should be available for the taking, if someone fully understands those risks and wishes to engage in an undertaking with such risks in order to help others. These considerations will also come into play in any other schema by which one might attempt to formulate unknown risks. For example, the current knowledge base from which potential risks might be gleaned is determined to be insufficient for reasonable prediction, one might decide that unknowns of this sort should be sufficient to put an end to any clinical trials. But again, different persons weight the possibility of encountering unknown risks (with thus unknown severity) in different ways. These values however should not lead us to believe that there are no considerations to be placed on the side of restricting or preventing research in gene therapy. We might first raise the issue of fair distributions of harms. It might be argued that persons have a right to take on risks, but only if those risks are limited to themselves. We are not justified in taking on risks on behalf of other autonomous individuals. With respect to lentiviral use in gene therapy, the possibility has been raised of transfer of a virus to the non-infected population—where it could, even with modification, have negative effects. The question is whether we should allow individuals to take on the risk to themselves knowing that they may unintentionally put others at risk in the process. How are we to understand these risks to society in researching and perhaps using these technologies and how are we to weigh them relative to risks to individuals if we choose not to attempt the use of these therapies? Again, a history of vaccination may provide us with an illustration, though as will be seen, it probably should not be considered to be the standard for ethical trial of a new medical technology. Edward Jenner is often mistakenly credited for the first attempted vaccination for small pox. In fact, this is not the case; exposing a child or young adult to smallpox in controlled circumstances was common practice in hopes of preventing a more severe case of the disease later in life. This practice was formalized somewhat (in a number of countries) to a procedure in which a small sample of infected tissue (a scab or pus) was taken from a person with smallpox and introduced under the skin of a healthy individual. This practice raised a great outcry; the idea of deliberately introducing a deadly organism into a human being to ward off subsequent infections with that organism had the aesthetic and appeal of black magic. This process, called “variolation,” left scars and had what we would consider to be a horrifyingly high fatality rate (approximately 15%), but compared to the mortality rate for naturally occurring smallpox, this was hailed to be a great success. The practice was popularized in England when a number of the aristocracy undertook the risks themselves or had their children vaccinated [interestingly, one of the first Western novelists, Fanny Burney, based her 1796 novel Camilla15 on the horrifying results of the common fears of the day of crude vaccination]; a more functional vaccine was finally introduced with quite variable results for general use in England in 1798. The risk to the public, of course, was great and largely unknown except through speculation, since introduction of smallpox to any uninfected population raised the possibility of infecting not just one person but all of his contacts, and even quarantine could not guarantee that infection would not spread. Jenner’s subsequent insight into the use of a much less virulent pox virus as a vaccinating agent reduced greatly the risk incurred by the populace at large.

Ethical Considerations in the Use of Lentiviral Vectors for Genetic Transfer

187

While there are a number of possible examples of potential public harm being undertaken for the good of individuals, the vaccination case should provide us some interesting material to contrast with concerns about lentiviral vector use. First, it is to be noted that the cases are similar in that there is a potential for creating a deadly infectious disease, though the limits of that potential are unknown. Unknown too, in both cases, are not just the levels of risk taken but the types of risk; the biological facts about the immune system that would subsequently explain the effectiveness of cowpox vaccination were unknown, and similarly, many of the details of vector function are still being researched, with new vectors being produced or existing vectors being modified in response to new data (interestingly, the condemnation of gene therapy by certain branches of the Protestant church echoes the Church of England’s condemnation of the practices of variolation and innoculation as practices which flew in the face of God’s chosen path for individuals who were stricken with disease). What might seem at first glance significantly different between the two cases is the relative levels of benefit and harms expected by the public. In the case of smallpox, the disease was so ravenous and so infectious that reducing the number of people who might carry the disease was of obvious benefit to the public at large. This is not the case with gene therapy; the benefits to the public (with the exception of gene therapy meant to treat HIV infections) are indirect. Moreover, the harms that might be incurred by the public are different in the two scenarios. In the case of vaccination, the harm is a direct result of the object that also creates the benefit; gene therapy, however, does not inflict harm directly through the gene which gives the benefit, but rather through the method of application of the treatment. These differences are suggestive of how we might begin to assess public harms. First, while public harm sometimes is overridden by the general good, consent on some public level is required. With harms incurred by war, for example, even when that war is for the good of a few individuals (imagine for example that the US starts a war in order to release a limited number of POWs from another country, putting the entire nation at risk for retaliation), there are institutionalized systems in place by which persons can agree to take those harms (for example, through the representative branch of the government). In the case of vaccination, the public directly began to clamour for the vaccine, especially after the efforts of a number of the aristocracy, most notably Lady Mary Montagu and the then Prince of Wales. In the case of gene therapy, the public has not been directly engaged on the subject of these risks, in part because they are so small, but in part because the treatment is still in the research phase rather than potentially in wide-spread clinical use. An appropriate branch of study therefore is potential public risks so that the possible clinical use of these technologies can be preceded with a frank public discussion of undergoing these harms. Second, appropriate behavior with respect to the public good requires a re-examination of a knee-jerk weighting of indirect benefits as less significant than direct ones. Though it seems that the vaccination program provided a direct benefit to the public, it is wise to keep in mind that this tells us only how people think, not how they ought to. The possibility of protection of the public was apparent in the case of vaccination, but we should not take this to be constitutive of an actual difference in potential public benefit. Economic analyses along with long-term non-economic benefits from reducing the number of persons with chronic illnesses must be weighed against potential harms, despite the fact that these benefits might be less obvious. A second set of concerns having not to do with the public good, but with individual rights has been raised in a number of contexts within the literature of research ethics. The question, crudely put, is whether someone who is very ill, perhaps dying, is capable of making truly informed, rational, autonomous decisions, particularly one in which he or she will be involved in research with unknown risks. To create a rather blunt analogy: if someone agreed to enroll in a clinical trial only when a gun was pointed at his head, we would be inclined—rightly I think—to say that this person did not make an autonomous decision. They are not taking on

188

Lentiviral Vector Systems for Gene Transfer

the risks, both known and unknown, willingly and in fact they may be acting contrary to their own best judgment. We can imagine that perhaps having a potentially fatal illness puts undue pressure on the research candidate; they cannot be expected to behave rationally or even authentically (“like themselves”) in these situations. It has even been argued that no one with a fatal illness should be allowed to enroll in trials on these grounds. This is particularly true of situations in which the only trials available are Phase I trials, where there is absolutely no guarantee of future benefit. Studies have suggested that many persons who enroll in Phase I trials do not understand that the trial will not provide any benefit (barring that from placebo effect and other psychological mechanisms for an experienced improvement in quality of life). Parents of minors with such illnesses are in an even worse situation with regard to making a rational and well thought out decision on behalf of their children. For these reasons, costbenefit analyses are incomplete or simply misleading in terms of determination of course of action within research trials. What else should be considered in the course of clinical trials, particularly with respect to lentivirus use? First, we will have to consider what it is for someone to truly consent to such research. We can reduce the sense of coercion that is potentially associated with being ill by ensuring, for example, that anyone who might wish to enroll in a trial has access to alternative treatments, has tried other treatments, and continues to receive better than adequate palliative care for his or her condition and emotional support for his or her illness. If any of these are not present, we might want to consider the person to be a candidate for other possibilities first. Furthermore, bare information is not sufficient in lentiviral vector trials (if it ever is); by “bare information” I mean stating the biological facts and the purpose of the trial, even in simple language. The reason that this is not sufficient is that, in the case of lentiviral vectors, the very language used in biological descriptions come equipped with societal stories and images that reach far beyond their technical meaning. “Leukemia” conjures up stories of dying wan children for whom foundations are created; “HIV” suggests AIDS, a disease which has a 20 year emotional and symbolic history which far outstrips its medical description. In order for a patient to understand fully what she would be undertaking in such a trial, it is important to address some potential assumptions as fully and honestly as possible. Moreover, it will be important to provide any research subject with continuing support and with guidance in how to deal with potential concerns from those around her as she undergoes this experimental procedure. Research considerations for GLGT are similar to those noted for SCGT above. However, GLGT requires two additional considerations: one, at the level of entertaining risks and benefits, and another at a more general level of concern. With regard to the risk-benefit analysis in which one might engage: there are further risks to be considered in such research in addition to those above. SCGT poses some potential risks that may be taken on the population at large, in a given period of time. However, GLGT adds another risk—that of the future generations that will be effected by any such research. The harms to be considered come in two forms: 1). harms that are apparent upon introduction of new genetic material into the research subject and that will be passed onto to any progeny of that subject and 2). harms resulting from GLGT that will not become apparent, or do not occur, until a generation has passed (for example, because of unexpected interactions between exogenous genetic material and materials within a newly formed zygote). These risks are often brought forward when making ethical distinctions between SCGT and GLGT. These risks however are balanced with some benefits: for one thing, the potential benefits for future generations of such research is that when perfected, these techniques can prevent certain diseases from occurring, thus sparing future generations the cost of dealing with these diseases, both financial and emotional. In addition, such research might point to ways in which early (prenatal) protection from such diseases might be more beneficial than even immediate post-natal gene therapies, preventing untold harm to many persons.

Ethical Considerations in the Use of Lentiviral Vectors for Genetic Transfer

189

The more general concern requires us to look at who the research subject is. It is clear, at least, that for SCGT research we can only research upon organisms that are at a state of development such that their somatic and germ cells can be distinguished from one another. GLGT however can take place either within gametes under the assumption that these gametes will be used in the production of new zygotes, within zygotes, or within early (toti- or pluri-potent) embryos. This raises troubling issues about who, if anyone, is the subject of such research and who (if anyone) is therefore in the position to give permission for such work. Let us take the case of the zygote and embryo in research together, before returning to the case of gametes. The moral and legal status of zygotes and embryos is currently an extraordinarily contentious issue, as is made clear by the emotional appeals that undercut the fetal tissue use and abortion debates. While it might be tempting to make an analogy between the use of fetal tissue for research and GLGT performed experimentally upon embryos, there is a slight but significant difference. In the former case, the beneficiaries of the research are other persons, in the fullest sense of that term; in the latter case, however, the potential beneficiaries of the research are (depending upon one’s stance on the moral status of fetuses) the embryos themselves or the infants and adults that those embryos will become. Therefore, we cannot expect that any direct benefit will be incurred by trying to apply the arguments ranged around the fetal tissue debates to the subject of GLGT. Nor can we easily bring to bear arguments surrounding the use of children as research subjects on this topic. For one thing, presuming an analogy between children and embryos is taking as given part of what is under contention—namely the status of the embryo or zygote as a research subject. We cannot simply say therefore that parents should have—as they do with children—the final power of decision-making about research to be done on embryos. Further, it’s unclear that the power parents have over decisions regarding trial enrollment for their children should in fact be taken for granted. As greater scrutiny is applied to the means and methods by which parents make such decisions, the particulars of decision-making for children has become more and more complex. The status of the embryo remains in doubt, and as long as it does so, there are no clear paths for determination of whether or not the embryo should be the subject of such research. The further question remains, however, of the availability of gametes for such research. Whether such research should be done on gametes seems an easier question than the same question applied to the embryo. Of course, this makes sense since there is little concern about whether gametes should be viewed as research subjects in their own right. They can instead be viewed as property of the producer, or as their body parts, in either (or other) cases, allowing for the use as the person involved sees fit. This would suggest that research into GLGT should be allowed if on gametes. However, there will be a point in the course of research in which the gametes will have to be used experimentally in production of a zygote; the research, after all, is intended to prevent harms by genetic disease be preventing disease entirely, from the beginning of life. Such research might be complex; the pressure of persons who donate gametes to have a child might be immense, because they have enrolled in a trial. Moreover, there is the possibility that any pregnancy resulting from the use of new technologies will result in defects in the embryo and spontaneous abortion. If miscarriage does not occur, there remains the question of whether researchers are in any way responsible for the fetus’ future well being. Related questions include whether it would be acceptable in any way to request that a woman have an abortion should the fetus prove to have developmental problems as the result of use of altered gametes and whether the woman could choose to have an abortion if the fetus is developing normally. The underlying questions come down to whether any fetus created from such altered cells is in fact a research subject and therefore deserves special consideration. While it could be argued that only the gametes are under research scrutiny, the intention in creating such gametes in later trials would surely be to test their use in the formation of a child who is disease-

190

Lentiviral Vector Systems for Gene Transfer

free despite its parentage. One might look for guidance to the possibility of human cloning; one of the most troubling concerns raised about such experimentation is that prior to Dolly’s successful birth, more than 10 malformed fetuses were produced.1 Should we be willing to entertain the possibility of creating malformed or damaged humans in the process of creating the first human clone? Similar troubling questions surrounding who the research subject is and what consideration is owed to that subject are magnified by the possibility that perhaps hundreds of anomalous fetuses would be developed prior to a successful pregnancy following alteration of gametes.

Clinical Considerations Assuming that these technologies are researched and proven to be appropriate for clinical use, a number of questions about particular types of use are similar to those raised above under the section on types of applications. However, there are some concerns about clinical applications that are independent of whether the technology is used for enhancement, or eugenics projects, or what have you. Concerns have been raised about the possible injustices of genetic therapy being available to people as a medical procedure. These concerns by and large are raised with respect to access to these technologies. The potential development of an underclass raised above with respect to genetic enhancement can be raised more generally with respect to any new medical technology, regardless of the type or tenor. Gene therapy may encounter problems with questions of just distribution in particular because genetics seem to be so close to the basic identity of any given person. Regardless of whether or not that identification of genes and person is appropriate or reasonable, the fact that such possibilities exist will suggest that persons who do not have access to genetic modification technologies will feel more fundamentally inferior than those without other sorts of medical care, because the possibility of this sort of “self improvement” will be unavailable. We could of course provide such technologies to all persons at reduced or no cost. The initial cost to the government—if this were done through a federal agency—would be immense. Alternatives, such as lotteries or waiting lists for medical treatments at low-cost also raise issues of fairness. And general programs of genetic improvement raise all the questions about reduced biodiversity and eugenics raised above.

Summary In sum, there are a vast number of ethical dilemmas that face us as we enter into the realm of gene therapy. Many of these are particular to the use of lentiviral vectors, and many of these dilemmas have appeared because the state of our knowledge about genetic modification is still quite limited. As with any new technology, a pro-active stance towards potential ethical problems seems the appropriate one to take. Ignoring potential problems does not allow for rapid response to the problems as they arise; over-preparation however can result in overuse of already strained medical resources. An awareness of potential problems and a wide-spread availability of information can in combination increase the probability that risks are minimized while the benefits of the new technologies are allowed to develop. Certain regulatory bodies both within research institutions and external to them can also be created or charged with monitoring trials of new medical technologies such that potential problems can be dealt with in an even-handed manner. By addressing these issues now, we can prevent ethical crises in the future.

Ethical Considerations in the Use of Lentiviral Vectors for Genetic Transfer

191

References 1. Pence G. Recreating Medicine. Lanham: Rowman and Littlefield Publishers Inc., 2000. 2. Kitcher P. The Lives To Come. New York: Simon and Schuster, 1993. 3. Moseley R. Maintaining the somatic/germ-line distinction: Some ethical drawbacks. J Med Philos 1991; 16(6):641-644. 4. Weiss R. Cosmetic gene therapy’s thorny traits. Washington Post 1997; Oct 12:A01. 5. Wivel NA, Walters L. Germ-line gene modification and disease prevention: Some medical and ethical perspectives. Science 1993; 262(5133):533-539. 6. Fletcher JC, Anderson WF. Germ-line therapy: A new stage of debate. Law Med Health Care 1992; 20(1-2):26-39. 7. WHO mission statement. http://www.who.int 8. Juengst ET. Can enhancement be distinguished from prevention in genetic medicine? J Med Philos 1997; 22:125-142. 9. Gould SJ. The Mismeasure of Man. New York: WWNorton, 1996. 10. Ashley-Koch A, Yang Q, Olney RS. Am J Epidemiol 2000; 151(9):839-845. 11. Persson, I. Genetic therapy, identity and the person-regarding reasons. Bioethics 1995; 9:16-31. 12. Churchill LR, Collins ML, King N et al. Genetic research as therapy: Implications of “gene therapy” for informed consent. J Law Med Ethics 1998; 26(1):38-47. 13. Markowitz D, Goff S, Bank, A. A safe packaging line for gene transfer: Separating viral genes on two different plasmids. J Virol 1988; 62:1120-1124. 14. Poeschla E et al. Development of HIV vectors for anti-HIV gene therapy. Proc Nat Acad Sci 1996; 93:11395-11399. 15. Burney, F. Camilla. Oxford: Oxford University Press, 1999

193

Index

Index A

G

Accessory proteins 13, 21, 29, 52, 59, 62, 99, 100, 108, 110, 112, 113, 115, 156, 160, 161, 168 AIDS 14, 62, 68, 72, 83-87, 99, 108, 160, 165 Assembly 5, 23-27, 29, 40, 45, 105, 148, 152, 154

G-to-A mutations 134 Gag 5, 7, 13, 14, 20, 22, 23, 25-30, 52, 53, 55-57, 59, 61, 63, 65, 66, 68, 70-72, 84, 90, 94, 104, 105, 107, 110, 114, 116, 135, 139, 140, 147, 152, 154-156, 161, 162 Gag-Pol 5, 52, 53, 55, 63, 66, 70, 135, 156, 162 Gag-Pro 65, 147, 152, 154-156 Gene therapy 13, 26, 60, 71, 83, 84, 92, 95, 99, 108, 112, 117, 118, 137, 148, 149, 155, 156, 159, 162, 166, 175, 176, 182-184, 186, 187, 190 Gene transfer vector 54-56, 58, 60-69, 71, 135, 136, 138, 142, 148-151, 153

C Caprine arthritis encephalitis virus (CAEV) 102, 106, 131, 133, 134, 138, 139, 142 CD4 14-18, 28, 29, 59, 85-87, 91, 92, 100, 101, 131, 155, 160, 165, 185 Central termination sequence (CTS) 19, 53, 55, 57, 68 Cis-acting sequences 9, 10, 56, 91-93, 109, 135, 136, 140, 147, 148 Constitutive transport element (CTE) 23, 63-66, 68, 69, 94, 109, 110, 114-116 Coreceptor 13, 16-18, 85, 86 cPPT 36, 43-46, 53, 55, 57, 68 Cytoskeleton 36, 43, 45

D DUTPase 133, 134, 136, 137

E Encapsidation signal 56, 95, 135, 140 Env 2, 3, 5-9, 13-15, 17, 22, 23, 25, 26, 28, 29, 52-54, 58-66, 69, 72, 88, 89, 92, 93, 102, 106, 107, 115, 132, 133, 135, 137-142, 147, 148, 150-155, 161 Equine infectious anemia virus (EIAV) 29, 41, 102, 105, 106, 131, 133-138, 142, 156

F Fanconi anemia 137 Fusion 3-5, 13-18, 25-27, 38, 40, 42, 53, 59, 65, 66, 85, 138, 155, 156, 165

H Helper cells 8 Helper virus 8, 92, 139, 168 Hematopoietic stem cells 58, 154 HIV 3, 13-30, 35, 36, 38-40, 42-46, 51-72, 83-95, 99-102, 104-111, 113, 114, 116, 131, 133, 136-138, 140, 141, 147, 149, 151, 152, 155, 156, 159-166, 168, 184, 185, 187, 188 HIV-1 3, 13-18, 20-30, 35, 40, 42, 45, 46, 51-72, 83-91, 94, 95, 100, 101, 104, 105, 107, 108, 113, 114, 131, 133, 136, 149, 151, 152, 155, 156, 159, 161, 162, 165, 166, 184 HIV-2 25, 27, 28, 30, 42, 65, 83-93, 95, 131, 156

I Integrase 3, 4, 13, 35, 36, 38, 39, 42, 44, 51, 53, 57, 70, 84, 87, 104, 105, 118, 155, 159, 165, 166 Integration 4-6, 9, 10, 18, 20, 21, 35, 36, 38-40, 42, 43, 45, 46, 51, 53, 55-57, 61, 66, 67, 70, 83, 84, 91-93, 108, 109, 135, 136, 148, 149, 151, 159, 161, 166, 184

Lentiviral Vector Systems for Gene Transfer

194

J Jembrana disease virus (JDV) 131, 133, 141, 142

L Lentivirus 1, 10, 20, 29, 30, 38, 40, 44-46, 51, 53, 55, 58, 62, 65-71, 83-95, 99, 102, 105-111, 114, 116-118, 131, 133-135, 137, 138, 140, 147-156, 175, 180, 182-184, 186-188, 190 Long terminal repeat (LTR) 2, 3, 5-8, 13, 21, 22, 28, 29, 40, 42, 51, 53-55, 57, 59-62, 65-67, 69-72, 87, 89, 90, 92-94, 99, 102, 103, 106, 108, 109, 112-114, 116, 133, 135, 136, 138-141, 147, 151, 153-156, 159, 161, 162

M Macrophage 15, 40, 42, 86, 90, 93, 94, 100, 131, 133, 134, 137 Matrix 3, 13, 36, 38, 39, 51, 53, 84, 90, 104, 159

N Nef 14, 22, 28, 29, 52, 53, 59, 60, 62, 88-90, 92-94, 133, 151, 160, 162 Neurons 40, 69, 93, 94, 101, 113, 136, 137, 163 Non-primate lentiviruses 42, 43, 84, 102, 104, 131, 133, 136 Nuclear localization signal (NLS) 21, 22, 36, 37, 39, 40, 42-45, 90, 159 Nuclear transport 4, 36, 44, 45, 90

O Oncoretrovirus 1, 148

P Packaging cells 8, 9 Packaging construct 55, 66, 91-94, 114, 115, 141, 147-156, 162 Packaging signal 5, 14, 24, 25, 55, 93, 108110, 115, 116, 147, 151, 155

Pathology 85, 86 Pol 5, 7, 13, 22, 23, 27, 52, 53, 55, 63, 66, 70, 71, 104, 135, 147, 152, 156, 161, 162 Pr160Gag-Pol 53 Pr55Gag 13, 23, 53 Preintegration complex (PIC) 4, 18-21, 23, 28, 36, 38-40, 42, 44-46, 51, 53, 55, 57, 59, 83, 84, 90, 149 Production 4-7, 9, 10, 23, 25, 43, 55, 58-65, 67, 69, 71, 72, 86, 91, 93-95, 108, 109, 112-118, 134-136, 138-142, 147, 148, 154, 155, 160-162, 166-168, 178, 184, 185, 189 Protease 3, 5, 13, 26, 53, 65, 70, 102, 104, 105, 155, 156, 165 Provirus 2-5, 8, 20, 21, 42, 52, 53, 63, 66, 68, 149, 151, 152, 184 Pseudotyping 9, 26, 58, 59, 61, 92, 93, 110, 111, 137, 142, 147, 151

R Recombination 7, 9, 20, 62-64, 66, 71, 85, 91, 93, 94, 108, 109, 114, 116, 135, 147-153, 155, 160, 162, 168, 184 Replication-competent virus (RCR) 9, 65, 71, 114, 147-149, 151, 154-156, 160, 162, 168 Retroviral genes 102 Retroviral vectors 6-9, 83, 92, 108, 110, 113, 148, 149, 156 Retrovirus classification 2 Retrovirus replication cycle 2, 4 Rev 13, 14, 21-23, 28, 52-55, 57-59, 61-66, 68, 69, 72, 88-90, 92, 94, 103, 104, 106, 107, 109, 110, 113-116, 133, 135, 139, 142, 151, 161, 162, 165, 166 Rev-independent 64, 65, 68, 69, 109, 161, 162 Reverse transcription (RT) 2-7, 10, 13, 14, 18-20, 28, 35, 36, 38-40, 43, 45, 46, 51, 53, 56, 59, 65-67, 71, 91-94, 102, 104, 105, 133, 134, 147, 148, 151-156, 160, 161, 165, 168 RNA transport 64, 66

195

Index

S

V

S2 gene 136 Safety 7, 9, 63, 66, 67, 71, 72, 93-95, 99, 108-111, 114, 116, 142, 147-149, 151, 152, 154, 155, 160, 167, 168, 183 Self-inactivating (SIN) vectors 7, 57, 66, 67, 70, 94, 95, 138, 147, 151, 154, 161 SIN 7, 57, 66, 67, 94, 95, 109, 138, 147, 151, 152, 154, 161, 162 SIV 14, 16, 27-29, 35, 66, 83-92, 94, 95, 101, 102, 105, 107, 131, 133, 160

Vector 1, 2, 6-10, 30, 45, 51, 53-72, 83, 84, 87, 90-95, 99, 108-118, 131, 134-142, 147-156, 159-164, 166-168, 175, 176, 180, 183-185, 187, 188 Vif 14, 22, 29, 52, 53, 59, 62, 100, 106, 107, 110, 113, 114, 133, 160, 161 Visna virus 131, 133, 134, 138-140, 142 Vpr 14, 18, 20, 22, 28, 30, 35, 36, 38, 39, 41-44, 51-53, 59, 62, 65, 66, 70, 84, 88, 90, 92-94, 113, 147, 151, 152, 155, 156, 159-162 Vpr-RT-IN 65, 66, 152, 155 Vpu 14, 22, 28, 29, 52, 53, 59, 62, 113, 160 Vpx 30, 41, 42, 66, 84, 88, 90, 92, 94, 156

T Tat 13, 14, 21, 22, 28, 51-59, 61-63, 65, 67, 69, 71, 72, 88-92, 106, 133, 135, 136, 138, 141, 150-154, 161, 162, 165, 166, 168 Tat-independent 59, 62, 63, 152 Tissue-specific promoter 7, 95 Trans-vector 147, 152, 154-156 Transgene 54-57, 60, 61, 67-72, 95, 109, 113, 114, 116, 117, 135, 159-161, 163, 167 Transport elements 64, 65

W Woodchuck Post-Transcriptional Regulatory Element (WPRE) 68, 109