Human immunodeficiency virus

I. GENERAL FEATURES 

A. History The infectious origin of acquired immunodeficiency syndrome (AIDS) was recognized soon after the identification of the disease itself in 1981. The first AIDS cases diagnosed in 1982 in hemophiliacs indicated that filtered Factor VIII concentrates could transmit the disease, suggesting that a virus or a small bacterium which could pass through bacteriological filters was the etiological agent. Indeed a retrovirus was isolated in early 1983 by L. Montagnier, F. Barre-Sinoussi, J. C. Chermann and their colleagues from a culture of activated T-lymphocytes derived from a lymph node biopsy of a homosexual patient with lymphadenopathy. Other similar isolates from full-blown AIDS patients were made by the same group in 1983, who showed, along with D. Klatzmann and J. C. Gluckman, their tropism for CD4_ lymphocytes. The viruses were found not to be antigenically related to human T cell leukemia viruses (HTLVs) but to be more closely related to the animal lentiviruses. This was despite the fact that some other studies (R. C. Gallo) produced data in favor of an HTLV-related variant. In 1984, R. C. Gallo, M. Popovic and their coworkers described, under the name of HTLV-III, a virus which proved to be identical to the virus described by the Pasteur group. The essential role of the virus, now renamed human immunodeficiency virus (HIV), in AIDS was demonstrated by epidemiological studies, particularly in cases of transmission by blood, where HIV was the only detected common factor in the donor and the receiver. A second type of virus, named HIV-2, was isolated in 1986 from West African patients with AIDS. 

The discovery of HIV as the etiological agent of AIDS has led to the development of rational therapeutic strategies such as reverse transcriptase inhibitors, protease inhibitors and vaccines. B. Taxonomy and classification By its morphology, genetic structure, nucleotide sequence, the virus belongs to the second subfamily of retrovirus, the lentiviruses, which includes viruses causing slow pathologies in animals, such as Visna, equine infectious anemia (EIA) and feline immunodeficiency virus. Retrolentiviruses of primates (SIVs, HIVs) are characterized by a tropism to CD4_ lymphocytes, a property not shared by lentiviruses of ungulates. C. Virion structure Mature virions (diameter 100–120 nm) have a characteristic spherical morphology with a dense coneshaped core surrounded by a bilayered phospholipid envelope in which knobs are inserted (Fig. 51.1). There are approximately 80 knobs covering the viral sphere. Each knob is made of several molecules of external glycoprotein, gp 120, possibly as trimers or tetramers linked noncovalently to preformed oligomers of the integral membrane protein, gp41. In the case of HIV-2, dimers or tetramers of the equivalent transmembrane proteins (gp36 or gp41) are so tightly bound that they appear as such (gp80 and gp160) in electrophoretic gels under classical denaturating conditions (heating at 100 _C in 1% SDS). D. Morphogenesis Virus assembly takes place at the surface of the plasma membrane of lymphocytes or lymphoid cells and in the membrane of intracytoplasmic vacuoles in the case of macrophages. Gag polypeptides are necessary to induce budding, and the subsequent morphogenesis of the virion requires the presence of the precursor glycoprotein, gp 160. Released immature particles with uncondensed cores are transformed into mature particles with condensed cores by proteolytic cleavage of the p17/p18 protein from the gag precursor, which binds to the inner layer of the viral envelope. As this occurs, the p24 protein wraps around the nucleocapsid, formed by the two RNA molecules and the basic proteins p6 and p9. The correct morphogenesis and infectivity of the viral particles involves cleavage by a cellular protease of the glycoprotein precursor, gp160 into gp120 and gp41. E. Physical properties Virions are sensitive to acidic pH (total inactivation at pH 2), ethanol (20–70%), heat (6 log inactivation at 60 _C 30 min), detergents (ionic and nonionic) and chlorine. Virions are sensitive to radiation including UV light at 260 nm, and X- and _-rays at a D10 value of over 500,000 rad (50 Gy). 

F. Genome structure The genome is a single-stranded RNA molecule of 9400 base pairs; there are two molecules per virion linked by noncovalent bonds. Lysine tRNA is the primer of the reverse transcriptase, which is Mg2_ dependent (optimal concentrations of Mg 10mM). The structure of the HIV genome is reflective of the complex nature of the limitations that are exerted during the replication of these viruses. Such limitations affect mainly the expression of viral genes, in an overall process that seems to optimize and synchronize expression of the viral proteins during acute replication, but could also in some instances control the expression of silent proviruses in chronically infected cells. The genome of HIV-1 and HIV-2 (Fig. 51.2), as well as of the different members of the simian immunodeficiency virus (SIV) group, contains several small genes in addition to the classical retroviral structural genes gag, pol, and env. Some of these genes have a critical regulatory function, others appear not to be absolutely required for in vitro replication. In addition to these coding elements, the HIV genome is rich in cis-acting sequences active at different steps of replication, some of which are target sites for proteins regulating viral gene expression. G. DNA synthesis and integration Following entry into the cell, the viral genome, represented by a dimer of two identical genomic RNA molecules, is reverse transcribed into linear, doublestranded DNA. The HIV reverse transcription process does not fundamentally differ from what is known for other retroviruses. One interesting difference, however, is that the second (plus) strand of the viral DNA has two distinct origin sites for its synthesis instead of only one. The plus-strand origin is determined, in all retroviruses, by a polypurine tract (PPT) located at the 5_ boundary of the U3 region of the long terminal repeats (LTR). HIVs and other lentiviruses have a second PPT at the center of their genome, which defines a central discontinuity in the plus strand, reflecting its use as an additional origin site. This second origin appears to improve the efficiency of the reverse transcription process, as shown by a reduced replicative capacity of HIV mutants lacking the central PPT. This feature appears to be common to all lentiviruses and spumaviruses. 

During the course of an acute infection, high copy numbers of unintegrated viral DNA molecules, either linear, circular with one LTR, or circular with two LTRs, can be observed in infected cells. Such accumulation of viral DNA appears to be the result of multiple events of infection in individual cells. Following its synthesis in the cytoplasm, viral DNA is transported to the nucleus and integrated into the host cell genome. This transport is an active process, occurring through nuclear pores, mediated by the interactions of the preintegration complex with cellular proteins. It can occur in nondividing cells, unlike the case of oncoviruses. The integration reaction, which can be reproduced in vitro, is mediated by the viral integrase, encoded by the C-terminal region of the pol gene. The integrase reacts with the termini of the linear viral DNA molecule and with apparently random sites of the host genome. Integration is regarded as being required for expression of the viral genes. II. REGULATION OF GENE EXPRESSION A. Transcription The 5_ LTR of the HIV genome contains the active viral promoter. The transcriptional start site defines the boundary between the U3 and R regions of the LTR. Transcription from the HIV promoter requires the presence of cellular transcriptional activators. Upstream of the transcription start site, surrounded by the usual transcriptional complex recognition signals, two important sets of sequences have been described: these are the three copies of the SP1 transcriptional activator binding site, and two copies of an enhancer element, which reacts with the transcriptional activator NF_B. Initially described as able to control transcription of the gene coding for the kappa chain of the immunoglobulins, NF_B is a ubiquitous transactivator whose components belong to a larger gene family that include the rel and dorsal protooncogenes. This factor can be activated in a number of cell types of the immune system, in particular following activation of the protein kinase C pathway. Further upstream in the U3 region of the LTR, is also found a variety of potential regulatory signals; a large segment of this region, termed NRE (negative regulatory element), has been described as exerting a negative regulatory action on transcription. B. 

Viral regulation of transcription: the Tat system Transcription of viral RNAs starts at the viral LTR and is mediated by cellular RNA polymerase II. But efficient synthesis of viral mRNAs only occurs in the presence of the viral protein Tat. This protein is produced from multiply spliced RNA, in which the two coding exons of the tat gene are linked together. The Tat protein acts by binding to a bulged stem–loop structure present at the 5_ end of all viral mRNAs, called TAR (Tat activation region). This binding is itself mediated by a cellular protein, cyclin T. Like other cyclins, cyclin T is associated with a protein kinase (CDK9) which will phosphorylate the C-terminus of RNA-polymerase II. This phosphorylation will allow efficient elongation of viral RNAs. These Tat cellular cofactors are only present in activated cells. Their absence will allow repression of transcription of proviral DNA which could thus enter a period of latency in T lymphocytes. This is another operative regulation that can allow either high virus production or quasi latency, in addition to the upstream control operated by transcription activators (NF_B) on the LTR promoter. There is a good crosstransactivation of the HIV-2 LTR by HIV-1 Tat, but HIV-2 Tat is a poor HIV-1 transactivator: this could be consecutive to the presence of a second stem–loop structure on the HIV-2 TAR, required for efficient HIV-2 Tat binding. C. Regulation of mRNA processing: the Rev system In eukaryotic cells, only fully spliced mRNAs are exported to the cytoplasm and are translated. Unspliced RNAs are retained in the nucleus and eventually degraded. Since HIV contains a single LTR promoter element, it encodes only a single, genome-length primary transcript. Selective expression of different viral genes is determined by differential splicing of this transcript. Splice sites (recognized by cellular proteins) are relatively inefficient, so that a pool of incompletely spliced RNAs could accumulate in the nucleus. During acute infection, three major classes of viral transcripts can be observed: the 9kb genomic RNAmolecules, producing Gag and Pol proteins, that can also be encapsidated into viral particles; the singly spliced 4.3kb RNAmolecules, coding for the env glycoproteins; and the 1.8–2 kb, multiply spliced RNA molecules, coding for the regulatory proteins. 

The cytoplasmic export and stability of the unspliced and singly spliced HIV transcripts, coding for the structural proteins, is controlled by the Rev protein. In the absence of Rev, only multiply spliced messages are seen in the cytoplasm of infected cells. When subgenomic Env constructs are studied, the effect of Rev appears to involve mostly cytoplasmic transport of unspliced RNAs. This restriction, however, seems to be dependent on the presence of splice sites. It is believed that Rev is able to oppose nuclear retention of RNA molecules by elements of the splicing machinery. It has also been suggested that Rev could favor cytoplasmic stability of RNA, association to polysomes, and even translation of unspliced or singly spliced transcripts. The target site for Rev is a complex multi stem-loop structure termed RRE (Rev responsive element), located in the intron separating the tat and rev coding exon, in the env coding region. Rev binds to the RRE but needs to multimerize to be active, which results in a threshold effect in Rev activity. Rev binds to the RRE through its N-terminal domain, whereas its C-terminal domain behaves as a nuclear export signal (NES) which binds to cellular proteins involved in transport from nucleus to cytoplasm. This effect has important consequences: during the early phase of a synchronous acute infection, the first transcripts to accumulate are the short multiply spliced species, followed by singly spliced and unspliced species. This shift from an early to a late phase of HIV expression appears to be under the control of accumulation of sufficient amounts of Rev. Similarly, some cells containing integrated HIV genomes bearing low basal expression levels do not produce structural RNAs; these can accumulate, however, after cell stimulation, in a process that is believed to involve accumulation of Rev. Therefore, the Rev system seems to promote synchronization of particle production during acute infection, and to delay structural protein expression in chronically infected cells, a phenomenon that could favor persistence of the infection in the context of a vigilant immune system. 

III. STRUCTURAL PROTEINS Table 51.1 summarizes virion proteins, their functions, cellular location and compares HIV-1 with HIV-2 proteins.  A. Gag and Pol proteins The Gag and Pol proteins are expressed from the full length, unspliced RNA molecules. In 1 : 100 of translation events, approximately, translation of gag is followed by a frameshift into the pol ORF. The Gag or Gag–Pol polyprotein precursors are then packaged into particles where the protease domain of the Gag–Pol precursor will cleave the final Gag and Pol products. The Gag precursor is cleaved into four proteins. The first Gag protein, from the N-terminus of the precursor, is termed p17/18 for HIV-1 and p16 for HIV-2 (matrix protein); it carries a N-terminal myristic acid, which allows association of both the Gag and Gag–Pol precursors to the cell membrane. The largest Gag protein, p24 (p26 in HIV-2) is the principal constituent of the cone-shaped core characteristic of viruses of the HIV group (capsid protein). The next Gag protein, p7 or p9, is termed the nucleocapsid protein, and is tightly associated with the RNA genome: it has been found to promote both RNA dimerization and encapsidation. The most C-terminal of the Gag proteins is p6; although its function is not known, mutants lacking this protein exhibit a defect in particle budding. The Pol region can be divided into three main functional elements. The most N-terminal is the protease, which acts as a dimer and cleaves Gag and Pol products during particle maturation. The central element is the reverse transcriptase, which acts as a heterodimer: the smaller component has polymerase activity, and the larger component has both polymerase and RNaseH activities. The action of RNaseH is required to eliminate the RNA template, with the exception of selected sequences used as primers, and to allow strand transfer events. B. Env The envelope glycoproteins of HIV-1 and HIV-2 are translated from a singly spliced RNA species. The glycoprotein is synthesized as a large glycosylated precursor which is then subjected to a series of biochemical rearrangements of its polysaccharide moiety and to a single cleavage event at a specific site carried out by a cellular convertase. This cleaveage separates the larger, outer portion of the envelope glycoprotein (gp 120) from the smaller, transmembrane protein (gp 41). The gp 120 carries on a specific domain, in the form of a pocket, that binds with high affinity D1–D2 domain of the CD4 molecule present on susceptible cells. 

The gp 120 binding domain is conformational, involving several regions, which are conserved within HIV variants. Upon binding to CD4, the gp 120 molecule undergoes conformational changes, which exposes some other sites to cellular molecules favoring the tight association of gp 41 transmembrane protein with the plasma membrane. Unlike the CD4 binding site, these sites are located in variable regions or loops of gp 120, particularly the V3 loop. This explains why the virus can choose a variety of cell surface molecules for its entry generally in association but not always, with the CD4 main receptor. Among these surface molecules (called entry cofactors or coreceptors), two main species have been recognized, which are physiological receptors for chemokines: the CCR5 receptor for chemokines (RANTES, MIP-1, MIP-1) and the CXC4 receptor for a chemokine (SDF-1). Very interestingly, the use of each coreceptor corresponds to viruses with different biological properties and pathogenicity. Viruses isolated at the beginning of infection use in most cases the CCR5 co-receptor (R5 viruses). They are not cytopathic in vitro (nonsyncytium-inducing viruses) and replicate in macrophages and activated primary T lymphocytes expressing CCR5 and CD4. In full-blown AIDS cases, new viral species appear with high power of replication in vitro, cytopathic effect (syncytia and single cell lysis), poor ability to replicate in macrophages and ability to grow in tumor T cell lines. They use the coreceptor CXC4 (X4 viruses). With the characterization of more viral isolates, the situation appears more complex; there are dual-tropic viruses, using both CXC4 and CCR5 coreceptors, or using alternative chemokine coreceptors (CCR3, CCR2, CCR8). There is also evidence of binding of the V3 loop to nucleolin, a nuclear protein whose one isoform is expressed at the cell surface, and with a dipeptidase present on activated T cells, CD26. The fusion between the viral membrane and the cellular membrane involves a change in the conformation of gp 41, itself induced by the conformational changes of gp 120. It involves exposure of a hydrophobic fusion domain present at the N-terminus of gp41, which enables it to insert itself in the cellular phospholipid bilayer. The transmembrane protein also displays an unusually long intracytoplasmic domain. This domain is particularly interesting in HIV-2 and SIV; indeed, tissue culture propagation of these viruses appears to favor truncation of the intracytoplasmic tail. The correction of this truncation leads to reduced infectivity in vitro. C. An important regulatory protein: Nef This third early protein is expressed in far higher levels than Tat and Rev. Most of the protein is myristilated at its N-terminus and associated with the inner side of the plasma membrane. However, there is partial exposure of the C-terminus of the molecule at the outer cell surface and in experimental infections with Nef-expressing vectors, Nef is also in part excreted into the medium. Expression of the intact Nef protein is not required for in vitro infection of peripheral blood leukocytes (PBLs) or of T cell lines. However Nef-deleted mutants of HIV and SIV are much less pathogenic in vivo, leading to persistent infection without disease, at least for a long period of time. 

Studies in the SIV/macaque model have shown that the initial virus load in lymph nodes is reduced by a factor of 10–50, and that formation of germinal centers in response to infection occurs more quickly than with the pathogenic Nef_ strains. In the productive cycle of viral infection, several functions of Nef have been identified, which contribute to a higher or longer virus production. Nef has been shown to strongly downregulate the expression of cell surface CD4, by connecting CD4 with the cellular endocytic machinery. This prevents reinfection of the infected cell with newly produced virions, which could lead to its premature death. Nef also downregulates to a lesser extent surface expression of MHG-1, resulting in some inhibition of recognition of the infected cells by specific cytotoxic T lymphocytes (CTL). Nef seems also, by independent, yet unclear, mechanisms, to enhance virion infectivity, perhaps by increasing the stability of the nucleocapsid at its entrance in the cell. Perhaps the most relevant property of Nef for its role in AIDS pathogenesis is its capacity to activate CD4_ T lymphocytes by modulating signaling pathways. An extreme case is that of the PBJ14 mutant of SIV which can induce within a few weeks an acute disease in pigtail macaques, and can activate both resting CD8_ and CD4_ lymphocytes in vitro. These unusual properties are linked to the creation in Nef by directed mutagenesis (or selection by passage in animals) of a new SH2 (Sarc homologous region 2) binding domain (YXXL motif). The fact that CD8 cells which are not infectable by SIV are also stimulated, suggests some paracrine effect of Nef expressed or released at the surface of infected CD4_ cells. Nef from ‘normally’ pathogenic strains of HIV or SIV contains a proline-rich sequence which could bind to the SH3 domain of kinases involved in the activation pathway of T lymphocytes. When mutations are introduced in this region, SIV is no longer pathogenic, unless reversion to the wild-type occurs. 

In addition, recombinant Nef protein, in the 10–50 ng/ml range, can act as a costimulatory factor in T lymphocytes in which the T cell receptor was subliminally activated and permits HIV replication in such cells. Thus, Nef produced by HIV-infected cells could recruit for infection neighboring T cells which were on the verge of activation and could not otherwise be susceptible to HIV infection. This recruitment may be important at the beginning of infection when the number of activated T cells in lymphatic tissues is low and is a limiting factor for HIV expansion.  D. Other HIV accessory proteins: Vif, Vpr, Vpu, Vpx All these proteins are encoded by single spliced mRNAs and therefore are produced late after sufficient Rev expression. 1. Vif The Vif protein is expressed at high levels in the cytoplasm of infected cells. Deletion of this gene reduces infectivity of HIV-1 virions in primary T4 cells and some tumor cell lines CH9. Some Vif proteins are also present in mature virions. It is believed that Vif can stabilize the virions upon entry in the cells and favor the formation of the reverse transcription complex. 2. Vpr This small basic protein is packaged in the virion nucleocapsid in amounts equivalent to the Gag proteins and therefore can be considered as a virion structural protein. Packaging depends on its association with the p6 protein released by proteolytic cleavage from the C-terminus of the p55 Gag precursor. Two effects of Vpr have been recognized. One, which is controversial, is to contribute to the nuclear import of the preintegration complex, by interacting with nucleoporins, and other cellular proteins, especially in nondividing differentiated cells (macrophages). The other function is to induce in cycling cells an arrest in G2. This effect is similar to that induced by DNAalkylating agents, and could result in the induction of the DNA repair machinery, which could be useful for integration of the proviral DNA. Rhesus macaques inoculated with a mutant SIV having a deletion in Vpr can still evolve towards AIDS-like disease, although some individuals show more resistance to disease progression. 3. Vpu The Vpu gene is unique to HIV-1 and the related virus isolated from chimpanzees. The product, Vpu, translated from the same mRNA that produces env, is an integral membrane protein with a C-terminal intracytoplasmic tail. Vpu has two functions. The first is to allow selective degradation of CD4 when the latter associates with the viral envelop precursor in the endoplasmic reticulum. This will release the Env protein from this complex and allow its subsequent incorporation into new virions. The second function is to facilitate virion release from the cell plasma membrane. Vpu-defective mutants show accumulation of intracellular viral particles. 4. Vpx The vpx gene is found only in HIV-2 and SIVs. Like Vpr, the protein is present in mature virions, in amounts roughly equal to that of the major Gag protein p24. Inside the cell, Vpx is located at the inner side of the cell plasma membrane. Vpx deletion mutants have a reduced capacity to replicate in primary T lymphocytes and macrophages, although they grow normally in T cell lines. IV. GENETICS One striking feature of the HIVs arose as soon as the nucleotide sequence of several HIV isolates was known: HIVs display a high level of genetic variability. The most variable region of their genome is the outer region of the envelope glycoprotein, where divergence can reach 30%. The most conserved are the gag and pol genes. The leading cause of such a diversity is reverse transcriptase: this enzyme does not have any proof-correction activity, and appears to introduce errors every 104 nucleotides. Another cause is selection: having to replicate in the context of a vigilant immune system, HIV needs to escape the antiviral immune response. This could explain why the envelope glycoprotein is the most variable region of the genome. It has also been shown that a single infected individual does not carry a single isolate, but a heterogenous population of subisolates, sometimes referred to as “quasispecies,” and that changing the conditions of viral replication (by growing the virus in vitro, for example) will modify the balance of that viral population. In addition the biological properties of the virus—as determined in the in vitro isolated— change during the course of the disease: in general, isolates made from lymphocytes of asymptomatic patients grow slower and are less cytopathic than isolates made from patients with full blown AIDS. These changes seem to be related to mutations in the envelope protein gp120, but no particular mutation can be assigned to the changes in biological properties. 

Within the HIV-1 type, genetic diversity appears to be more striking when African isolates are compared with one another or with Western isolates. This is an indirect indication that African isolates may have evolved longer than their Western counterparts. Based on nucleotide sequence of the variable regions of the envelope protein gp120 (particularly V3), a cladistic classification of HIV-1 isolates is currently being used, using capital alphabetic letters, A to J. They all belong to the group M of HIV-1s (M for major). Since the early 1990s another group of HIV-1 has been described in Central Africa (Cameroon) with some sporadic cases in Europe. This group, which differs from the M group by around 50% of nucleotides sequences in the env gene, has been termed O (for outlayer). All HIV-1s are highly pathogenic but display various geographic locations (see below). More radically divergent is the HIV-2 group, which was prominent in West Africa in the 1980s, but now tends to be less prevalent than HIV-1. By serology and nucleotide sequences, HIV-2, which also displays several subtypes, is closer to SIVs than to HIV-1. In fact, it probably originates from a SIV naturally infecting sooty mangabeys living in West Africa. HIV-2 was initially found in some West African patients with AIDS, with symptoms comparable to HIV-1 disease. However, its pathogenic potential seems to be lower than HIV-1, displaying a lower viral load in chronically infected patients and a lower capacity of transmission by sexual contacts. V. ORIGIN AND GEOGRAPHIC DISTRIBUTION HIV-1 seems to have originated in Central Africa. This is suggested by the larger extent of variability found in strains isolated from African patients. Retrospective serological studies also indicate that the virus was present in the early 1970s in 0.25% of a population of young women living in Kinshasa (Zaire). However, only retrospective sporadic cases could be detected in Africa before 1980 as well as in Europe and the United States. The HIV-1 directed epidemic started at about the same time in larger cities of Central Africa and of the United States. Molecular analysis of the virus present in a serum taken from a Zairian patient in 1959 indicates that it could be an ancestor of several identified HIV-1 clades: B, which is now prominent in North America and Western Europe, D and F present in Central Africa. In the phylogenetic trees, two isolates made from chimpanzees are clearly closer to the M group than from the O group, the origin of which cannot be traced to any simian lentivirus. This group may have been circulating in some human populations for a longer time. In the 1990s HIV-1 infection has become pandemic, rapidly developing in new regions such as India, Southeast Asia and South America. 

In Africa, the infection has spread to the East (Uganda, Kenya, Tanzania, Rwanda, Burundi) and to the North (Ivory Coast). Concerning HIV-2, its spread has been more limited from the areas where it was first detected: Guinea-Bissau, Cap Verde Islands, Senegal. HIV-2 is also present in other countries of the Western part of Africa (Ivory Coast, Burkina Faso, Cameroon, Liberia etc.) and Portuguese-speaking countries (Angola, Mozambique). An outbreak of HIV-2 infection has been detected in prostitutes of the Bombay area, India, suggesting that the virus has also reached Asia. New rapidly spreading foci of HIV-1 infection have occurred in South Africa, Zimbabwe, Cambodia, and China. Some subtypes are becoming prominent, according to the geographic location, such as Ain Ivory Coast, C and D in Central and South Africa, E in Thailand, F in Brazil. As the epidemic progresses, infection of the same individual by two variants becomes frequent, leading to the emergence of recombinants (particularly in the env gene) between subtypes. As of 2002, the number of HIV-infected persons is estimated to be about 40 million worldwide. VI. SEROLOGIC RELATIONSHIPS Despite the large spectrum of genetic variability, the prototype strain of HIV-1, isolated at the Pasteur Institute, LAV LAI (otherwise named HTLV-IIIB) can still serve as source of antigen for the detection of HIV-1 antibodies in all geographic regions of the world, including antibodies against the envelope. By contrast, little crossreactivity exists between SIV/HIV-2 group and HIV-1 glycoproteins, whereas crossreactivity remains important in Gag and Pol proteins of the two viruses. The lack of crossreactivity in a major epitope of the transmembrane protein (gp41, gp36) has been used to differentiate between HIV-1 and HIV-2 infection, or to detect double infection. So far a few patients have shown, by western blot using HIV-1 and HIV-2 prototypes as antigen, atypical reactivity. Such cases, in which, for instance, no antibodies against the glycoprotein of HIV-1 or HIV-2 could be detected in serum, are probably due to variants distantly related to HIV-1 or HIV-2. With regards to animal retroviruses, SIV-infected primates do react well with all HIV-2 proteins, confirming the close relationship between SIV (particularly SIV from sooty mangabey) and HIV-2. By contrast, other animal lentiviruses (Visna, CAEV, BIV, FIV) are too distantly related to show antibody crossreactivity, with the exception of equine infectious anemia virus which shows crossreactivity with HIV-1, at the level of the Gag p24 protein. 

VII. HOST RANGE AND VIRUS TRANSMISSION HIV-1 has a very limited host range; human and chimpanzee are the only species known so far which can be chronically infected with the virus. However, no disease or deep immune depression have been observed in the hundreds of chimpanzees inoculated with human isolates of HIV-1, with one exception, whereas lack of symptoms is the exception in humans. HIV-2 can chronically infect some macaque species (rhesus, cynomolgus); an AIDS-like syndrome has been observed in some experiments, but not in a reproducible way. The two main routes of transmission of HIVs are blood and blood products and sexual contact. The efficiency of blood transmission (transfusion, needles, i.v. drug abuse) depends on several factors: number of virus particles, volume of blood, immune status of the receiver. Infection is particularly efficient in i.v. drug abusers. Sexual transmission, homosexual and heterosexual, is the major mode of transmission today. All sexual practices are dangerous, but the risk is higher for anogenital intercourse, and is increased by some intercurrent genital infections (herpes, chlamydia, etc.). Transmission from mother to child is also a major mode. In the absence of treatment, 20–30% of seropositive women give birth to an infected child. Infection can occur in the second half of pregnancy, at delivery and also by breast feeding. Severe infection of children results in death in the first year of life. Otherwise the evolution follows that seen in adults. VIII. ROLE OF THE VIRUS IN AIDS PATHOGENESIS Acquired Immune Deficiency Syndrome (AIDS, SIDA in French and Spanish-speaking countries) was defined in the early 1980s, biologically by a profound defect of cellular immunity associated with a deep shortage of CD4_ T lymphocytes and clinically by the occurrence of opportunistic infections and cancers. In Western countries, the most frequent infections are those of the lungs by Pneumocystis carinii and of the brain by toxoplasmas, followed by visceral and retinal infections by cytomegalovirus (CMV). Tuberculosis is frequent in tropical areas (Africa, Asia, South America). Among cancers, the most frequent are disseminated and aggressive Kaposi’s sarcoma caused by human herpesvirus 8 (HHV-8) and B-lymphomas, often caused by Epstein–Barr virus (EBV). The causal relationship between HIV and the immune depression has been assessed as follows: 1. On the basis of epidemiological studies, particularly in blood donors and recipients; 2. From the tropism and cytopathic effect of HIV and viral glycoproteins on CD4_ cells; 3. On the reproduction of the disease in macaques by virus derived from molecular clones of SIV (close to HIV-2). The natural history of HIV-1 infection has been thoroughy studied by histological and molecular techniques. Three phases can be distinguished: primary infection, a clinical latency phase, the clinical phase (full blown AIDS). 

In the primary infection, entry of the virus through mucosa involves its association with dendritic cells which then transport the virus to lymphatic tissues. The virus can rapidly multiply there in activated CD4_ lymphocytes and macrophages. Since the number of activated CD4 lymphocytes is small, it becomes a limiting factor which depends on intercurring microbial infections. Biologically, this phase is characterized by a peak of viral antigens (not always present) and a large number of viral RNA copies in blood, and a peak of interferons. Specific cytotoxic lymphocytes (CTL) then appear followed by the appearance of antibodies against viral glycoproteins and internal proteins. Clinically, when the level of viral replication is high, symptoms can be detected (fever, adenopathy, headaches). However, the infection can be sometimes completely inapparent. This episode is followed by a phase of clinical latency, with no symptoms. However, in a majority of cases, there is a slow and progressive degradation of the immune system, leading finally to clinical AIDS which will last on average 10 years, in the absence of treatment. In a minority of individuals (5–10%) (longterm nonprogressors) this degradation does not take place or is very slow. In some others, on the contrary, immune depression can occur very rapidly within 1–2 years. These variations reflect the complex interaction of HIV with the immune system of the host. In fact, the virus continuously replicates in lymphatic tissues (lymph nodes, spleen), with a rapid clearance from the blood. A relatively small number of infected cells is involved at the beginning and their destruction (either by the virus or by cytotoxic cells) cannot solely account for the large depletion of CD4_ cells. It is likely that indirect mechanisms of cell death are involved, bearing on cells which are not infected, but are in contact with noninfectious viral particles, or gp120 shed from cells or virions. Indeed, interaction of gp120 with the CD4 receptor or coreceptors could induce a wrong signaling leading to apoptotic cell death or anergy, when the T cell receptor is stimulated. 

Evidence for the preapoptotic state of a large fractional circulating CD4 cells has been obtained. Defects in antigen-presenting cells, in bone marrow renewal of precursor cells, of faster thymic involution, of high oxidative stress, have also been involved to explain the specific CD4_ cell depletion. It is also clear that a state of chronic activation bearing not only on CD4_ cells, but also on CD8, NK and B cells exists all through this phase, associated with the production of inflammatory cytokines (interferons, interleukin (IL) 6, tumor necrosis factor). In addition, a small pool of latently infected cells exists and is probably continuously renewed. This pool will escape any kind of treatment. In the clinical phase, with the occurrence of more pathogenic variants (CXC4 tropic) and the decrease in cell-mediated immune response, the infection from local or regional becomes systemic and precipitates the fatal evolution. Neurological signs are also frequent, particularly a subacute encephalopathy which can develop into a dementia syndrome and brain atrophy, and seems to be due to the virus itself. Some foci of brain macrophages infected with HIV can be detected in white matter, but neither neurons nor glial cells seem to be productibly infected in the same situation. Some astrocytes express large amounts of Nef protein. Therefore, indirect mechanisms for the action of HIV (cytokines, nitric oxide?) have also to be postulated in order to explain the neuronal effects. IX. IMMUNE RESPONSE An immune response against HIV is mediated by T-helper cells. Antibody response against major HIV structural protein (Gag, Env) and also against Pol proteins and Nef appear within 2–3 weeks after HIV exposure, exceptionally after 6 months or more. Specific CD8_ cytotoxic T lymphocytes (CTLs) also appear, often earlier than the antibody response, directed against the same proteins. Antibodies against the viral envelope are poorly neutralizing, and are directed against variable regions of gp120, especially the V3 loop. There is also induction of antiviral cytokines (interferons) but for unclear reasons, even large amounts of interferon are not effective to control viral expansion. 

There is also in vitro evidence for viral inhibition by other soluble factors, particularly secreted by CD8_ T lymphocytes; the increase of chemokines may inhibit by competition the entry of C5 viruses, but some other inhibitory factors produced by CD8_ cells have not yet been identified. As the infection progresses, the specific CTL response tends to decrease and to lose its efficacy against mutant viruses, although a strong natural killer response (NK cells) can still also play a protective role. The number and function of CD4_ T lymphocytes decreases progressively with some oscillations during the clinically latent phase. T-helper functions to recalled antigens are precociously affected, in line with a drop of IL-2 and an increase of Th2 cytokines (IL-4). The antibody response against viral proteins decreases at the clinical phase, reaching first the internal proteins. Owing to their higher titers, antibodies against gp120 and gp41 remain detectable even at late stages. These antibodies can also be detected in urine in lower titer. X. TREATMENT Until 1994, there was no effective treatment of HIV infection and AIDS. Monotherapy using nucleosidic or nonnucleosidic inhibitors of HIV reverse transcriptase gave disappointing results in clinical trials. It was realized that the inhibitory effect on virus replication was not strong enough to prevent the rapid occurrence of resistant mutants. Dramatic improvements appeared in 1995–1996 due to the design of a new class of potent antiviral drugs, the inhibitors of the viral protease, which prevent the maturation of infectious virions by inhibiting the cleavage of the gag-pol precursor. Another important development was the general acceptance of the concept of combination therapy, in which two, three, or even four inhibitors act synergistically and decrease the emergence of resistants. Finally the application of sensitive molecular techniques (polymerase chain reaction, branched DNA) to quantitate the viral load in plasma was an important tool to monitor the antiviral effect of the treatment. The combined use of two inhibitors (such as azidothymidine (AZT) and 3-thiodeoxycitidine (3-TC)) of reverse transcriptase with a protease inhibitor (such as Ritonavir, Indinavir, Nelfinavir) results generally, in a strong decrease of viral load within weeks (2–4 logs), a slow but consistent increase of CD4_ cells and an improvement of the patient’s condition.  Opportunistic infections are much less frequent and there is biological evidence of at least partial restoration of T helper-dependent immune functions (response to recall antigens, drop of apoptosis and cell activation markers). However, this treatment called HAART (highly active retroviral therapy) has some constraints and limitations. 

1. It has to be taken daily and indefinitely. 2. Active virus multiplication is not completely suppressed, and multiresistant variants may appear, more rapidly if the patient does not adhere to strict compliance to the treatment. 3. There is a continuously renewed reservoir of cells which are latently infected and therefore have unexpressed proviral DNA which escapes the reverse transcriptase and protease inhibitors. 4. Important side effects, such as mobilization of lipids resulting in hypertriglyceridemia, insulinresistant diabetes, can occur after long-term treatment. 5. The high cost precludes any generalization of its application for patients in developing countries. Therefore, there is a need for intensive research to develop new types of retroviral inhibitors, to find practical tests to evaluate the spectrum of resistance and sensitivity to available drugs of the patient’s virus, to find complementary treatments aimed at reducing the dosage and duration of HAART and at achieving better restoration of the immune system. Among possible promising developments are the use of: very low doses of IL-2 (a daily subcutaneous inoculation for 6 months); hydroxyurea; and inhibitors of cell activation and a specific immunotherapy against viral proteins. XI. PREVENTION AND VACCINE No vaccine is yet available and the only effective prevention at present is education about the ways of transmission, systematic testing of blood donors and the use of condoms. However, an important reduction of transmission from mother to child has been achieved by treatment with AZT of the mother at the end of pregnancy and at the time of delivery and of the newborn in the first weeks of life. A lighter regimen also seems to be effective and applicable to populations of developing countries. A complete control of the AIDS epidemic cannot be achieved without the availability of a protective vaccine. Although the use of whole virus or whole surface glycoproteins has been disappointing, there are new promising approaches based on the use of DNA, mucosal adjuvants and live vectors, internal and regulatory proteins, and conserved parts of the surface glycoproteins. Studies of naturally resistant HIV-exposed individuals have shown that cell-mediated immunity and the secretion of soluble inhibitory factors are more important for protection than antibodies, together with a limited role of genetic factors. Individuals homozygous for mutations impairing the expression of CCR5 coreceptors are totally resistant to infections by ‘R5 tropic viruses’. However, these are rare (_2%) in the caucasian population. Individuals heterozygous for these mutations can still be infected, but seem to evolve more slowly to the clinical phase of AIDS. In fact, a large majority of exposed noninfected individuals have acquired a specific immune resistance to infection, based on CTL and specific IgAs, although the role of other genetic factors (HLA) cannot be excluded. The efficacy trials of candidate vaccine in large populations will raise important logistic and ethical issues difficult to solve, unless an important international mobilization greater than that achieved for vaccinal eradication of poliomyelitis, is achieved.