Kuoting Wu

The appearance of defective interfering(DI)particle is a well-recognized consequence of passaging virus at high multiplicity of infection(2). The ubiquity of DI viruses was first clearly recognized by Huang and Baltimore(1970). A defective virus genome is any viral genome lacking adequate function in one or more of the essential genes required for autonomous viral replication. All viruses could be defective in one sense because they require living host cells and cell factors in order to replicate(3).

DI presence is not always easily detected. It depends on the virus assay system used, the type of virus, the DI genome class, etc. There is no sensitive reliable assay system suitable for DI genome types of all virus , so wide variety of DI genome assays have been developed. These methods includes electron microscopy of particles, particle purification, visualization on sucrose velocity gradients, assays for interference with virus-directed nucleic acid synthesis of virus particle synthesis, assays for subgenomic-sized nucleic acids in particles, cells, or tissues, and various cell protection assays. Recently polymerase chain reaction(PCR) techniques for amplifying DNA and RNA segments have been proven useful(1).

Defective viruses containing only some portions of the infectious virus genomes, require homologous parental viruses as helpers for replication. Helper viruses support virus structural proteins and antigens and DI particles replicate preferentially at the expense of infectious helper viruses in coinfected cells(4). DI particles most often interfere at the level of virus replication, but interference at the competition for encapsidation or maturation, for enzyme function, or for transcription /translation steps has also been reported(3). DI particles might frequently exhibit a cell-sparing effect(4), facilitate production of interferon(5) and apparently modulate immune responses in some circumstances(6,7).

Viruses as diverse as the specialized transducing phages, the defective RNA tumor viruses carry onc genes. Papovaviruses and adenoviruses are capable of carrying host-cell genes from previous hosts to subsequently infected cells. The generalized transducing phages or pseudovirions may not require to contain any viral sequences but only host-cell DNA encapsidated within the virus coat for transfer to other cells.

Both DNA and RNA tumor viruses cause oncogenic transformation. Most of these viruses contain defective genomes. Retroviruses and retrotransposons have provided powerful models for the study of animal genetics from oncogenes to transgenic animals. Defective retroviruses can not only capture, mutate, and transduce exons of cellular genes, but by insertional mutagenesis they can activate, inactivate, and rearrange host genes on chromosomes. Human genomes contain a number of defective retrovirus genomes and can co-express viral and cellular sequences. Defective human retroviruses might carry genes with disease-inducing potential. The presence of transactivating genes, as for example the tat gene of human T-lymphotropic virus type 1, might cause cell transformation or exert subtle effects on expression of certain host genes. Hepatitis B virus, a major factor in most human hepatocellular carcinoma, exhibits integrated defective viral genomes in most of these carcinomas, often associated with mixed viral-cell DNA rearrangements, hybrid cell-virus RNA transcripts, and even extensive host chromosome DNA deletions around viral integration sites. Whether integrated or not, defective viral genomes may be involved in many chronic diseases to malignancy. Since defective virus genomes are often overlooked or difficult to detect in chronic diseases, their roles have generally been inadequately explored in these diseases

DI particles of RNA viruses exert a modulating effect on viral disease processes in both plants and animals. DI particles have been reported to play a major role in attenuating helper virus virulence and in allowing cell survival and prolonged persistent injections by a large number of RNA viruses. SSPE(subacute sclerosing panencephalitis) is a reasonable model for studying chronic diseases of suspected RNA viruses etiology because the extreme rates of mutation and evolution of RNA virus genomes can readily produce nonmaturing defective replicons.

Botstein and Susskind proposed a generalized modular theory of virus evolution which holds that evolution acts primarily not at the level of intact viruses, but at the level of individual functional units (modules). According to this theory, each intact virus in nature is the product not of a single line of evolution but a favorable combination of gene modules that successfully occupy a particular niche. Obviously, this scenario of virus evolution places defective virus genomes within the mainstream of evolution. During persistent infections mediated by DI particle interference, infectious virus mutants often appear that are much more resistant to interference by the original DI particles than the original virus used to establish persistence. DI genomes can represent an intrinsic destablizing force to generate virus population diversity and rapid evolution(3).

Many classes of animal RNA virus accumulate defective variants when propagated serially at high multiplicities of infection. This was first observed by von Magnus in influenza virus. Such defective viruses frequently interfere with the replication of non-defective genomes and they may play a role in pathogenesis with persistent infection. Defective retroviruses do not typically interfere with their helper viruses and are not simply truncated forms of the competent virus: they contain recombinant viral or cellular sequences that can directly influence the pathogenesis of the virus population. Nearly all retroviral oncogenes are carried in defective viruses requiring helper viruses for transmission. The mixture of defective oncoviruses and the helper viruses required for their generation and transmission represents a special case of quasispecies genome dynamics(29). No doubt defective forms of HIV will be carefully scrutinized at the molecular level.

An important genetic polymorphism has been recognized for the human retrovirus at the origin of the acquired immune deficiency syndrome (AIDS) and other diseases, like lymphadenopathy syndrome (LAS), AIDS-related complex (ARC), and probably some encephalophathies.(21) Human immunodeficiency virus type 1 (HIV-1) displays both interstrain and intrastrain genetic variability. Virus populations with extensive micro heterogeneity have been defined as swarms or quasispecies.(8) Quasispecises are populations of closely related genomes which are unique and in consequence to describe HIV isolates (23). Many of the genomes within HIV-1 swarms appear to be defective in one or more genes required for viral replication. Heterogeneity of HIV-1 can lead to substantial biological differences, has adverse consequences with respect to an effective immune response, poses problems in designing a vaccine, and by allowing selection of natural virus mutant resistant to anti viral therapy(8). It is important to investigate the genetic variability of the AIDS virus, in particular its range, the underlying mechanisms, and the apparent existence of hypervariable or well conserved domains in the viral proteins. Such conserved domains are likely to be associated with important biological functions, and their delineation would be a step towards understanding the molecular mechanisms of viral pathogenicity(22). Mutation frequencies have been estimated from the nucleotide sequences of HIV-1 isolates taken from a single patient at different times after infection(8). Minimum estimates for the variation rate in the env gene range from 0.001 to 0.01 nucleotide substitutions per site per year, whereas mutation frequencies for the gag gene are about one-tenth as high. The most remarkable feature of HIV-1 variability is the presence of hypervariable regions within specific portions of the env and nef genes(9). Some general features of the AIDS virus genetic variability are now emerging. First, it is principally caused by point mutations, often resulting in amino acid substitutions, more frequently in the 3' end of the genome (ORF S, env,, and ORF F). Another source of genetic diversity is insertions-deletions. Analyzing these insertions, they have been observed that most often represent one of the two copies of a direct repeat. Defective tat and vpu genes have been found in up to 15% of viral genomes both in PBMC (peripheral blood monocytes) and in cultured viral isolates. 15% of gag and env sequences were defective in viral isolates(38). The identification of conserved domains in the highly immunogenic envelope glycoprotein and the core structural proteins (gag) are very important (22). The gp120 gene is also the most variable of the HIV-1 genes, and the degree of sequence variation in distinct virus isolates is growing at a rate of up to about 1% per year(39).A strong bias in the nucleotide substitution frequencies was evident within the gag and env sequence data. By far the most frequent nucleotide misincorporation was G to A. The G to A bias may account for the high A content of HIV(34%). It also means that codons particularly rich in G (like glycine, GGN or tryptophan, TGG) will be particularly vulnerable to change(26).

Retroviruses are subject to these changes in part because of the infidelity and lack of proofreading mechanisms of reverse transcriptase (RTase). Genetic recombination also increases the variability of the retroviral genome.

Genetic variability in HIV-1 may involve one or all of several steps in the viral life cycle (Fig.1), including reverse transcription, integration into the host chromosome to establish the provial state and provirus replication by RNA polymerase II into RNAs to be packaged into virions. Preston, et al used three different methods to evaluate the fidelity of DNA synthesis by HIV-1 RTase: first, minus sequencing gel assay for nucleotide misincorporation; second, primer extension for site-specific nucleotide misincorporation; third, reversion of the amber codon to wild type to measure reversion of a single base mutation. To compare with avian myeloblastosis virus, HIV-1 RTase introduced base substitution errors at estimated frequencies of 1/2000 to 1/4000; catalyzed nucleotide mismatches with a specificity of A:T>>A:C>>A:G>A:A. The high error rate of in vitro translation approximately five to ten errors per HIV-1 genome per round of replication in vivo(10). Roberts, et al used M13mp2-based fidelity assays indicated HIV-1 RTase was exceptionally inaccurate, having an average error rate per detectable nucleotide incorporated of 1/1700. It was the least accurate reverse transcriptase described to date: one tenth as accurate as the polymerases isolated from avian myeloblastosis or murine Leukemia viruses, which have average error rates of ~1/17,000 and ~1/30,000, respectively(9). The rate of nucleotide misincorporation by the RTase is of the order of > 10-4 per base per cycle. From these results are consistent with the possibility that the infidelity of reverse transcription is responsible for the sequence diversity among different HIV-1 isolates.

Homologous recombination makes it possible to (i)increase the variation in a population by combining different variants, and (ii)repair damaged genes. Two retroviral RNA molecules are packaged in one virion. One consequence of this packaging of two RNA molecules in one retroviral particle is a high rate of homologous recombination(11). Recombination between two genetically marked retroviruses is not observed after coinfection with these two viruses, but only after infection with viruses. The rate of retroviral recombination is about 2% per kilobase per replication cycle. Polygenetic analysis of HIV-1 genomes indicate that some individual viral isolates contain both "ancient" and "contemporary" sequences and suggest that genetic recombination could have played a role in their generation(34). Two models have been proposed for the mechanism of retroviral recombination(12):

It proposes that the genomic RNA in the virion is damaged. When reverse transcriptase encounters a break in the viral RNA, it switches to the other copy of genomic RNA to salvage the encoded genetic information. This model can be generalized to include all recombination that occurs during the synthesis of minus strand DNA.

It proposes that both copies of viral RNA are intact and that two minus strand DNA's are made by one virion. The strand displacement-assimilation model predicts that recombination occurs during plus strand DNA synthesis.

Recombinants with more than one template switch were observed frequently. This result indicates that the genetic information in two copackaged retroviral RNA's can be shuffled rapidly, and recombinants with mosaic patterns can form frequently in one round of replication. Thus, the diversity of a retroviral population increases quickly, especially with a hypermutanted virus that has multiple mutations in one viral genome(12). Repeated recombinations could also lead to the emergence of infectious particles with novel pathogenic potential(35). This could herald the end of the asymptomatic phase of an HIV infection and result in disease progression.

Genetic variants might become less virulent. Several mechanisms could compensate for attenuated virulence, including recombination, phenotypic mixing, and complementation in trans(15).However, as with the feline leukemia virus, a defective murine leukaemia virus plays the role in inducing severe immunodeficiency disease(13). The evaluation of proviral latency and defective viral genomes in vivo remains an important area in the study of the complex pathogenesis of HIV-1 infection(16).

Latent infection involves the integration of viral or proviral DNA into host genetic material concomitant with no viral expression in the period of time (31).The characteristics of latency are the following modifications of complete viral expression: (1) lack of detectable structural proteins;(2) expression of regulatory proteins from only doubly spliced mRNA;(3) cellular activation controlling per versus post-viral integration events;(4)intracellular budding and reservoir effects;(5)defective integrated virus expressing no, some, or all structural proteins;(6)wild-type integrated proviral DNA with no RNA expression. The persistently HIV-1 infected cells obtained after HIV-1 infection were shown to greatly reduce the expression of surface CD4 antigen(40). When multiple defective integrated copies exist in patient PBL (peripheral blood leukemia), then cellular mixing with concomitant recombination or complementation might permit a new wave of wild-type infectious virus. Such a recombination would give rise to more defectives. The levels of wild-type infectious virus would subside and an appearance of latency would again persist(31).If a defective or latent provirus is found within a cell, its potential role in the pathogenesis of AIDS should not be dismissed since these cells can presumably be infected with replication-competent virus and undergo recombination to generate new intragenic viral assortments. The latency of HIV may be controlled by expressed cellular factors which are influenced by the state of differentiation and by susceptibility to physiologic stimuli. Several models of HIV transcriptional modulation have been proposed which might explain how HIV can be induced by physicochemical stimuli to be expressed as infectious virus, when HIV harbored in a latent form. These models include transcriptional repression of integrated provial DNA by (i) cellular or viral repressor DNA-binding proteins ;(ii) chromatin conformation change;(iii)DNA hyper-methylation;(iv)pre-integrative viral species;(v) mediation of transcriptional transaction by cellular nuclear transcription factors;(vi) HIV regulatory proteins(31).(Fig.4) The potential role of defective genomes in the pathogenesis of AIDS remains to be elucidated.

In attempts to determine which region of the viral genome controls cell tropism, biologically diverse molecular variants have been generated from infectious clones of HIV-1 by substitution of the env gene portion. The result suggested that the viral envelope determines at least in part, biological diversity(17). Takeuchi and Hoshino have described a mutant virus that is infectious not only for T cells, but also for fibroblasts-like cells derived from human brain tumors. The reciprocal help of two defective infectious molecular clones of HIV-1 is at least partially mediated by complementation in trans resulted in broadened cell tropism and increased virulence(15). Several epidemiological and clinical studies suggest that patients coinfected with HIV and other viruses, such as cytomegalovirus or HTLV, have a more sever clinical course than those infected with HIV alone. It is possible that cross-transactivation, phenotypic mixing of HIV with other viruses plays a role in pathogenesis. Mixing of envelope glycoproteins in doubly infected cells to form pseudotype virions has been demonstrated for several of the enveloped viruses(32). Landau et al established methods for the generation of HIV-1 and HTLV-II pseudotype viruses. It can test the ability of these viruses to incorporate heterologous envelope glycoprotein in the presence or absence of the homologous envelope glycoprotein and in the absence of other heterologous viral components. Their results showed that HIV-1 can be readily pseudotyped by the envelope glycoprotein of HTLV-1, resulting in expanded cell tropism and host range. The mechanism by which viruses incorporate homologous or heterologous envelope glycoproteins is poorly understood. It has been suggested that heterologous envelope glycoproteins are incorporated into the assembling virion by associating with homologous envelope glycoprotein molecules on the cell surface. The homologous envelope glycoprotein could then serve to nucleate the budding reaction (33). Pseudotypes may have an important role in the pathogenesis of HIV-related syndromes, such as those involving in the central nervous system (32).

Defective RNA genomes may modulate viral characteristics in vitro and immune responses and pathogenesis in vivo(19,20). While natural selection driven by viral requirements for infection, replication, and regulatory controls limit the extent of viral variation in many important function, immune selection can allow for propagation of replication-competent mutants which have escaped immunologic control. The extensive variability of the viral envelope provides a basis for immune selection. Viral variants may arise during the course of infection and are not neutralized by the host immune system. A single, major, immunodominant neutralizing epitope has been identified within the third variable(V3)region of HIV-1 (14).V3 encodes the type-specific principal neutralizing domain of the virus, and divergence within this region has been shown to result in escape from neutralization(23). Wilson et al show that the 582 region (V3) does not by itself constitute a neutralization epitope. The basis for neutralization resistance of the immune-selected variant is more likely a conformational change which alters a neutralization epitope at a distant site. The tat protein of HIV-1 is known to be secreted by cells infected with HIV-1. It can penetrate and act as a trans-activator in uninfected cells and was thought to be a growth factor for Kaposi's sarcoma(KS) cell lines in vitro(27). A sizeable proportion (10-15%) of each quasispecies, whether in vivo or in vitro, was made up of functionally defective tat gene variants(26). Huang Y.Q.et al showed the HIV-negative KS patients who are also at high risk for HIV disease have been infected with a defective HIV-1 virus that retained the ability to express tat(27). The use of site-directed mutagenesis to disrupt the putative HIV-1 fusion domains supports the role of this domain in syncytium formation. The mutant envelope proteins were observed to be defective in their ability to form syncytia. A loss of syncytium formation results in an attenuation of infectivity and the syncytial cytopathic effect loss, but single-cell lysis is still maintained. The decreased infectivity of the mutant viruses also resulted in an altered tropism for different CD4+ cell lines(28). The differences between the temporal patterns in vivo and in vitro are striking. Monocytoid cultures may be extremely useful for the future identification and propagation of slow-replicating HIV-1 strains. But it will be considerable importance to clarify how accurately in vitro adapted HIV-1 strains reflect the properties of in vivo strains and whether defective or slow replicating viruses have significance in disease progression(24). There are a number of possible explanations for the differences between the quasispecies in vivo and in vitro. First, it could represent simply the selection of the form best adapted to tissue culture conditions. A second possibility is that in vitro culture using conventional techniques selects against viruses harbored in circulating monocytes. A third is that in vitro culture expands both CD4+ helper and CD8+ cytotoxic and suppressor cells. It is possible that cells harboring the most abundant viral forms are either suppressed or killed during early culture(25). The biological diversity, present in vivo, is presumably lost in vitro by the selective outgrowth of HIV-1 strains with rapid replication capability (24).

Defective viruses might affect the severity of infection because they compromise the normal immune responses of the host(18). Aziz et al have identified a severe immunodeficiency disease induced by a defective murine leukaemia virus. Molecular cloning and sequencing of this 4.8Kb genome showed that the pol and env genes have been deleted, but the complete gag region has been conserved and has a novel sequence encoding the unique p12 protein. Weiss, R.A. suspected this product that is responsible for inducing AIDS. The studies of structure of this defective viral genome revealed several large deletions in pol, env and a relatively well conserved region in gag. Because this p12 related protein is unique and because of the absence of any other unique long open reading frame, this gag region is probably important in the pathogenesis of the disease. Their results emphasize the need to search for similar pathogenic replication-defective variants in immunodeficiency disease, including AIDS. Their results suggested that the pathogenic AIDS virus may also be a defective retrovirus. Poss et al indicated that the property of the env protein of the defective feline AIDS virus, which makes it cytopathic for T cells, lies in its post-translational glycosylation pattern(30). These posttranslational function modifications include both the protein primary structure and the repertoire of host cell processing enzymes, either or both of which may influence the biological behavior of the viral envelope proteins. Poss et al also demonstrate that the Feline Leukemia Virus genomic changes in the env gene result in antigenic molecular weight and processing differences in the pathogenic variant envelope glycoprotein compared with that of its apathogenic putative parent genome.

Different algorithms for analyzing sequence relationships among viruses yield evolutionary tree relating one isolate to another within families of viruses. Use of mathematical modeling based on the predator-prey relationship described for a defective interfering particle and its helper standard virus. There was chaotic unpredictability even in a deterministic model. Unlike most evolution studies, here is immediate and important practical value to understanding the evolution of viruses. The key is that altered conditions permit efficient spread of virus in its new host rather than the initial infection events themselves(37). Recent studies of HIV-1 infected chimpanzees also demonstrated that neutralization-resistant variants arise naturally in vivo during the course of infection. The possibility that HIV-1 may thus elude host immunologic control mechanisms must be carefully considered in designing putative vaccines(14). African isolates of the AIDS virus are expected to differ significantly from the USA isolates previously sequenced. The nucleotide sequence analysis will probably be an important source of information on the genetic polymorphism of the virus. This work may also give some insight into the possibility that the AIDS virus originated in Africa had some evolutionary relationship of LAV to other human and simian retroviruses more recently identified in Africa(36). Genetic variation represents a general selective advantage for lentiviruses by allowing an adaptation to different environment by modifying their tissue or host tropisms. In the particular case of the AIDS virus, rapid genetic variation are tolerated, especially in the envelope; they could allow the virus to adapt to different "micro environments" of the membrane of their principal target cells, namely the T4 lymphocytes. These different "micro environments" could result from the proximity of the virus receptor to polymorphous surface proteins, differing either between individuals or between clones of lymphocytes. Identification of the AIDS virus envelope domains responsible for this interaction (receptor-binding domains) appears as fundamental for understanding the host-viral interactions. It is useful for designing a protective vaccine, since an immune response against these epitopes could possibly elicit neutralizing antibodies. The conservation of the tropism for T4 lymphocytes probably represents a major evolutionary advantage of the retroviruses(22).

1. Roux, L., Simon, A.E., and Holland, J.J. (1991) effects of defective interfering viruses on virus replication and pathogensis Advances in virus research Vol 40 pp.181-211. Academic Press

2. Schlesinger, S. (1988) In "RNA Genetics"(E. Domingo, J.J. Holland, and P. ahlquist, eds), Vol 2 pp. 167-185. CRC Press

3. Field, B.N., Knipe, D.M., (1990) Field Virology 2nd Edition Vol 1 pp151-166 Raven Press, New York

4. Huang, A.S. (1988) In "RNA Genetics"(E. Domingo, J.J. Holland, and P. ahlquist, eds), Vol 3 pp. 195-208. CRC Press

5. Johnston M.D.(1981) The characteristics required for a Sendai virus preparation to induce high levels of interferon in human lymphoblastoid cells. J Gen. Virol. 56:175-84

6. Dimmock N.J., Beck, S. McLain, L.(1986) Protection of mice from lethal influenza: evidence that defective interfering virus modulates the immune response and not virus multiplication. J.Gen. Virol. 67:839-50

7. Browning, M.J., Huneycutt, B.S., Huang, A.S. and Reiss, C.S. (1991)Replication-defective viruses modulate immune responses. J. Immunol. 147(8):2685-2691

8. Hahn, B. H. et al,(1986) Science 232 :1548

9. Roberts, J.D., Bebenek, K., and Kunkel, T.A.(1988) The accuracy of reverse transcriptase from HIV-1 Science 242:1171-1173

10. Prestion, B.D., Poiesz, B.J. and Loeb, L.A.(1988) fidelity of HIV-1 Reverse transcriptase Science 242:1168-1170.

11. Hu, W.S. and Temin, H.M. (1990) Genetic consequences of packaging two RNA genomes in one retroviral particle:Pseudodiploidy and high rate of genetic recombination Proc.Natl.Acad.Sci.USA 87:1556-60

12. Hu, W.S. and Temin, H.M. (1990) Science 250:1227-1238

13. Aziz, D.C., Hanna, Z. and Jolicoeur P. (1989) Severe immunodeficiency disease induced by a defective murine leukaemia virus Nature 338:505-508

14. Wilson, C., Reitz, M.S.Jr.,Aldrich, K., et al (1990) J. Virology 64(7):3240-3248

15. Lori, F., Hall, L., etal (1992) Effect of reciprocal complmentation of two defective human immunodeficiency virus type 1 molecular clones on HIV-1 cell tropism and virulence J. Virology 66(9):5553-5560

16. Bagasra, O. et al (1992)N. Engl J. Med. 326:1385-91

17. Fenyo, E.M., Albert, J. and Asjo, B.(1989) AIDS 3(suppl 1):S5-S12

18. Takeuchi, Y.,Hoshino, H. A host-range mutant of HIV which can infect fibroblast-like cells derived form human brain. V International conference on AIDS. Montreal, June 1989 (Abstract Th C.P.134)

18. Browning, M.J., Huneycutt, B.S., Huang, A.S. and Reiss, C.S.(1991) J.Immunology 147(8):2685-2691

19. Barrett, A.D.T. and Dimmock, N.J. (1986) Defective interfering viruses and infections of animals. Cur. Topics Microbiol. Immunol. 128:55-84.

20. Perrault, J. (1981) Origin and replication of defective interfering particles. Cur. Topics Microbiol. Immunol. 93:152-209

21. Weiss, R.A (1984) Human T-cell retroviruses. In Molecular Biology of the Tumor Viruses: RNA Tumor Viruses, vol II,Supplement, pp. 405-485.

22. Alizon, M., Wain-Hobson, S., Montagnier, L. and Sonigo, P. (1986) Genetic variability of the AIDS virus: nucleotide sequence analysis of two isolates from African patients. Cell 46:63-74.

23. Mckeating, J.A.,J. Gow, J.Goudsmit, L.H. Peral, C.Mulder, and R.A. Weiss. (1989) Characterization of HIV-1 neutralization escape mutants. AIDS 3:777-784.

24. Fisher, A.G. et al (1988) Biologically diverse molecular variants within a single HIV-1 isolate

25. Meyerhans, A. et al (1989) Temporal fluctuations in HIV quasispecies in vivo are not reflected by sequential HIV isolations. Cell 58:901-910

26. Wain-Hobson, S (1989) HIV genome variability in vivo AIDS 3:S13-s18.

27. Huang, Y.Q. et al (1992) The absence of tat sequences in tissues of HIV-negative patients with epidemic Kaposi's sarcoma. AIDS 6:1139-1142

28. Dedera, D. and Ratner, L. Demonsratin of two distinct cytopathic effects with syncytium formation-defective human immunodeficiency virus type 1 mutants. J. Virology 65(11):6129-6136

29. Weiss, R.A. AIDS defective viruses to blame? (1989) Nature 338:458

30. Poss. M.L. Mullins, J.I. and Hoover, E.A. (1989) Posttranslatinal odifications distinguish the envelope glycopretein of the immunodeficiency disease-inducing Felin Leukemia Virus Retrovirus. J. Virology 63(1):189-195.

31. Bednarik, D.P. and Folks, T.M. (1992)Mechanisms of HIV-1 latency AIDS 6:3-16

32. Landau, N.R., Page, K.A. and Littman, D.R.(1991) J. Virology 65(1):162-169

33. Witte, O.N. and Baltimore, D.(1977) Mechanism of formation of pweudotypes between vesicular stomatitis virus and murine leukemia virus. Cell 11:505-511

34. Clavel,F. et al (1989) Genetic recombination of human immunodeficiency virus J.Virology 63(3):1455-1459

35. Tersmette, M. et al (1988) Differential syncytium-inducing capacity of human immunodeficiency virus isolates:frequent detection of syncytium-inducing isolates in patients with acquired immunodefieiency syndrome and AIDS-related complex. J. Virology. 62:2026-2030

36. Kanki, P.J. et al (1985) Isolation of T-lymphotropic retroviruses from wild-caught African Gren Monkeys. Science 230:951-954

37. Huang, A.S.and Coffin, J.M. (1992) How does variation count? Nature 359:107-108

38. Goodenow, M.,Huet, T et al (1989) HIV-1 isolates are rapidly evolving quasispecies:evidence for viral mixtures and perferred nucleotide substitution. J. Acquired Immune Defic.Syndr. 2:344-352.

39. Balfe, P. et al (1990) Concurrent evolution of human immunodeficiency virus type 1 in patients infected from the same source: rate of sequence change and low frequency of inactivationg mutations. J. Virology 64:6221-6233.

40. Yunoki, M. et al (1991) Production of infectious particles from defective human immunodeficiency virus type-producing cell clones by superinfection with infectious HIV-1 Arch. Virology 116:143-158.

Fig.1 Step in the life cycle of HIV.(1)Viral entry is dependent upon the presence of the CD4 molecule, which bonds a specific domain in the gp120 HIV envelope protein. (2)After viral uncoating, reverse transcriptase synthesizes a complimentary to the genomic viral RNA; the viral RNA copy is simultaneously degraded by the specific ribonuclease H activity. (3)The double-stranded proviral DNA may undergo a two base pair deletion at each end, followed by random integration, as a nucleoprotein complex, into the host cellular genome. (4)Alternatively, the proviral DNA may form circular, non-integrated intermediates as monomeric or multimeric concatomers. (5)Once integrated, the 5' long terminal repeat will bind both viral and cellular transcriptional proteins to initiate transcription. (6)The viral messenger RNA is translated as mature, spliced transcripts to yield viral proteins. (7)Viral proteins are orchestrated into an assembly mechanism which packages the genomic viral RNA. (8)Mature virions emerge via budding through the cell membrane.

Fig.2 Forced copy-choice model (minus strand recombination). Each virion contains two RNA's-RNA-a and RNA-b. (Step1)After transfer of minus strand strong stop DNA, the minus strand DNA synthesis continues with RNA-a as template until a break in RNA-a is encountered. (Step2)The growing point of the minus strand DNA switches to RNA-b and continues to copy the genetic information in RNA-b. (Step3)After the completion of synthesis of both DNA strands, a recombinant DNA molecule containing genetic information from both parents results.

Fig.3 Strand displacement-assimilation model (plus strand recombination) (Step1) Two minus strand DNA's are synthesized from on virion, each with one of the co-packed RNA's as template. Upper case A and B represent the DNA generated from RNA-a and RNA-b, respectively. (Step2)Transfer of the plus strand strong stop DNA occurs, and plus strand DNA synthesis is initially discontinuous, forming internally initiated fragments. (Step3)One of the internally initiated fragments in plus strand DNA-B is displaced by the growing point of DNA elongation from the transferred plus strand strong stop DNA. This displaced DNA-B fragment begins to hybridize to the minus strand DNA-A. (Step4)The displaced DNA-B fragment is assimilated into the DNA-A structure. (Step5)A complete viral DNA molecule is formed. A region of this DNA molecule contains genetic information from the two parents, the minus strand from parent a and plus strand from parent b. (Step6)Mismatch repair, probably before cell replication, corrects the different sequences in this region. When the genetic information from parent b remains, a recombinant is formed.

Fig.4 Possible mechanism of HIV latency (a)Wild-type -expressing form of HIV showing the integrated provirus with elongating messenger RNA transcripts. (b)Five potential scenarios whereby mRNA initiation or elongation would not occur. (1)Hypermethylation. (2)Dense chromatin formation. (3)Cellular or viral repressor bound to initiation sites in long terminal repeats. Such structural restraints might render enhancer or initiation sites unavailable for transcriptional factors. (4)Cellular signal(X) might be missing, resulting in lost activation of transcriptional initiation. (5)Similar to (4), except here activation signals to convert incoming HIV genome-(a) into a completely reverse transcribed proviral DNA-(b), resulting in non-integrated state are missing.