key: cord-0039689-1nom5hyk
authors: Kanki, Phyllis J.; Essex, Myron E.
title: Virology
date: 2007-05-09
journal: The AIDS Pandemic
DOI: 10.1016/b978-012465271-2/50004-2
sha: 25280c168a34151b9aa24c448b0d71161ab8f0b5
doc_id: 39689
cord_uid: 1nom5hyk

nan

AIDS was first recognized as a new and distinct clinical entity in 1981 (Gottlieb et al., 1981; Masur et al., 1981; Siegal et al., 1981) . The first cases were recognized because of an unusual clustering of diseases such as Kaposi sarcoma and Pneumocystis carinii pneumonia (PCP) in young homosexual men. Although such unusual diseases had previously been observed in distinct subgroups of the population -such as older men of Mediterranean origin in the case of Kaposi sarcoma or severely immunosuppressed cancer patients in the case of PCP -the occurrence of these diseases in previously healthy young people was unprecedented. Since most of the first cases of this newly defined clinical syndrome involved homosexual men, lifestyle practices were first implicated and intensively investigated. These included the exposure to amyl or butyl nitrate 'poppers' or the frequent contact with sperm through rectal sex, which might have acted as immunostimulatory doses of foreign proteins or antigens.

However, AIDS cases were soon reported in other populations as well, including injection drug users (IDUs) (Anonymous, 1982) and hemophiliacs (Davis et al., 1983; Poon et al., 1983; Elliott et al., 1983) . Similar to investigations of male homosexual populations, these new risk groups were exposed to doses of foreign proteins and antigens but through the blood. In the case of IDUs, this would occur through intravenous injection of drugs and needle sharing, and in the case of hemophiliacs through the therapeutic infusion of Factor VIII. Asymptomatic hemophiliacs and IDUs were often found to have unusual immunological tests, such as the inverted T lymphocyte helper:suppressor ratios (i.e. less than the normal 1:1 ratio of helper to suppressor cells). These abnormal tests suggested a problem with the cellular immune system, with low numbers of helper T lymphocytes, also referred to as CD4ϩ lymphocytes, a finding similar to that observed in many AIDS patients.

Three new categories of AIDS patients were soon observed, including blood transfusion recipients (Curran et al., 1984; Jaffe et al., 1984) , adults from Central Africa (Clumeck et al., 1983; Piot et al., 1984; Van de Perre et al., 1984) , and infants born to mothers who had AIDS or were IDUs (Rubinstein et al., 1983; Oleske et al., 1983; Scott et al., 1984) . The transfusion-associated cases had received blood donated from an AIDS patient at least three years before they began showing symptoms (Curran et al., 1984; Jaffe et al., 1984) . Based on the disparate populations afflicted with this new malady and the emerging epidemiology of the disease, the possible infectious etiology for AIDS was considered (Francis et al., 1983) .

Multiple studies were initiated to determine the possible role of various microorganisms, especially viruses, in causing AIDS. These studies measured and compared seroprevalence rates for suspect viruses in AIDS patients and controls. The shortlist of candidate viruses included: cytomegalovirus, which was already associated with immunosuppression in kidney transplant patients; Epstein-Barr virus, which was a lymphotropic virus; and hepatitis B virus, which was known to occur at elevated rates in both homosexual men and recipients of blood or blood products. However, based on the unique clinical syndrome and unusual epidemiology of AIDS, if one of these viruses was to be etiologically involved, it would presumably have been a newly mutated or recombinant genetic variant.

At the same time, our group , Gallo and his colleagues (Gelmann et al., 1983) , and Montagnier and his colleagues (Barre-Sinoussi et al., 1983) postulated that a variant T lymphotropic retrovirus (HTLV) might be the etiologic agent of AIDS. Indeed, HTLV, discovered by Gallo and his colleagues in 1980, was the only human virus known to infect T helper (CD4ϩ ) lymphocytes at that time (Poiesz et al., 1980) . This seemed reasonable since it was already clear that T helper lymphocytes were selectively depleted by the causative agent in clinical AIDS (Rubinstein et al., 1983; Ammann et al., 1983; Fahey et al., 1984; Lane et al., 1985) . In addition, HTLV was known to transmit through the same routes as the etiologic agent of AIDS: sexual contact (with transmission apparently more efficient from males), by blood, and from mother to baby (Essex, 1982) . Finally, HTLV-I was also known to induce immunosuppression (Essex et al., 1984) , as we had previously recognized in animal retrovirus systems (Essex et al., 1985) .

AIDS patient blood samples were repeatedly cultured in an attempt to find a virus related to HTLV-I or HTLV-II (Kalyanaraman et al., 1982) . Although antibodies cross-reactive with HTLV-I and HTLV-related genomic sequences were found in a minority of AIDS patients (Gelmann et al., 1981; Barre-Sinoussi et al., 1983; Essex et al., 1983) , the reactivity was weak, suggesting either AIDS patients were also infected with an HTLV, or that a distant, weakly reactive virus was the causative agent. Soon after, Gallo and his colleagues obtained proof that AIDS was linked to an HTLV Gallo et al., 1984; Schupbach et al., 1984) . Further characterization of the agent -now termed HIV-1 -revealed that it was the same as the isolate detected earlier by Montagnier and his colleagues (Barre-Sinoussi et al., 1983) . Despite controversy over the names and identity of certain isolates, this new and unique human pathogen was clearly not only a distant genetic relative of the known HTLV virus but may have been recently introduced to humans from a primate reservoir (Essex and Kanki, 1988) .

After HIV-1 was recognized as the cause of AIDS, it was also recognized that this virus was new, at least to inhabitants of the Western Hemisphere. This raised the question of whether HIV-1 was also new to Old World human populations, such as Africa, or whether it recently entered humans from another species. Reports of diseases such as AIDS had not been reported in these populations. It is conceivable that if HIV-1 had been present in Africa for some time, the interplay between pathogen and host would have led to a less virulent virus and people with genetic resistance to the lethal properties of the virus. However, early clinical studies did not indicate differences in the pathogenicity of the virus-host interactions in Africa compared with the United States or Europe.

HIV-1 or a related virus was likely present in human populations in Central Africa at the same time or even before AIDS was diagnosed in the United States.

In the early 1980s, Africans residing in Europe were presenting with similar clinical signs and symptoms of AIDS (Clumeck et al., 1983) . Serum samples collected from Africans at earlier periods were also examined for the presence of antibodies reactive with HIV-1. In some cases, the examination of stored samples suggested elevated rates of infection in Africa during the mid-1960s to 1970s (Saxinger et al., 1985) . Subsequently, it was revealed that most of those surveys were conducted with first-stage tests that were imperfect. In addition, the reactors were mostly false positives, due either to contamination of the HIV antigen or to 'sticky sera' containing antibodies that reacted non-specifically because the sera had been repeatedly frozen and thawed and maintained under poor conditions. While examining sera taken from Africa in the period 1955 to 1965, we found one antibody-positive sample that was clearly positive in a specific manner (Nahmias et al., 1986) . When tested with the highly specific test radioimmunoprecipitation, this sample contained high titers of antibodies that were reactive with virtually all the major antigens of HIV-1. This sample represented only a rare positive reactor in a high-risk group of individuals suffering from tuberculosis and AIDS-like illness in a region that subsequently had high rates of infection with HIV-1. However, only 1 per cent or fewer of the individuals who tested positive were from what is now classified as a region of moderate to high prevalence (Kinshasa, Zaire), which suggests that the virus was then only rarely present in places that would now be classified as within the 'AIDS Belt of Africa.'

The recognition that AIDS is caused by HIV was not easy for some to accept. A small group of scientists have persisted in denial of the overwhelming evidence for causation. Additionally, various conspiracy theories have emerged suggesting that HIV-1 was deliberately created by germ warfare scientists. While such proclamations seem silly or irrational to informed medical scientists, they interfere with constructive attempts to educate appropriate population groups.

One reason why some have been reluctant to accept that AIDS is an infectious disease caused by HIV-1 is the very prolonged induction period combined with a very high mortality rate. Most infectious diseases occur after a short induction period. Even for the small number of infectious diseases with a very long induction period caused by viruses, such as tropical spastic paraparesis, adult T cell leukemia or shingles, only a small fraction of infected people experience that clinical outcome. The definition of AIDS as an amalgamation of clinical outcomes ranging from tuberculosis to chronic diarrhea to cancer (e.g. lymphoma and Kaposi sarcoma) also may cause confusion for those trying to accept HIV-1 as the single etiologic cause. This only becomes logical when it is recognized that AIDS disease is fundamentally an irreversible destruction of the immune system. All of the other outcomes are secondary to the immune destruction.

Another problem in understanding AIDS etiology may be a lack of appreciation of the discipline of epidemiology and dissension about the proper definition of cause. For epidemiologists, a very high-risk association, such as for tobacco and lung cancer, is sufficient to ascribe cause. Using analogous logic, the causal association between HIV-1 infection and subsequent destruction of the immune system is overwhelming. In prospective cohort studies, almost all HIV-1-infected people eventually develop immune depletion. This association is much higher than for such viral infections as polio or flu. Concerning time and spatial geographic associations, clinical AIDS rates have exactly paralleled HIV infection rates, whether in Bombay, San Francisco or Nairobi, allowing for the 5-to 15year induction period.

Until recently, a lack of understanding about how HIV-1 caused immune depletion helped those who reject epidemiology to deny causation. This situation has changed dramatically, however, with the recognition that HIVs-1 can be highly lytic for T4 lymphocytes (Yu et al., 1994) , and that very large numbers of T4 cells are killed by the virus in vivo (Wei et al., 1995; Ho et al., 1995) . Further, while the very low rate of infected circulating lymphocytes was interpreted by some as incompatible with the destruction of large numbers of cells, recent studies reveal that most HIV-1 is in lymph nodes rather than blood, and up to 25-50 per cent of lymph node T cells may be infected (Pantaleo et al., 1993) . HIV-1 is also highly unusual as a virus that targets T4 lymphocytes and macrophages, both essential components of the immune system. Finally, HIV-1 is transmitted in exactly the same way as clinical AIDS: by blood, by sex, and from mother to infant. When taken together, these various correlations provide inescapable evidence that HIV-1 must be the cause of AIDS.

Soon after the recognition of clinical AIDS in people, several clinical reports described outbreaks of severe infections, wasting disease, and death in several colonies of Asian macaque monkeys housed at primate centers in the United States (Letvin et al., 1983; Henrickson et al., 1983) . Due to their similarity to the human syndrome, these diseases were designated simian AIDS or SAIDS. As in the case of human AIDS, many possible causes were considered. Following the recognition that SAIDS appeared to be of infectious origin, cytomegalovirus of monkeys was also considered as a possible etiologic agent.

However, seroepidemiological screening revealed that a proportion of the SAIDS monkeys had antibodies that cross-reacted with HIV (Kanki et al., 1985a,b) while healthy monkeys had no such antibodies. Although the antibodies cross-reacted with core antigens of HIV-1, they showed only very weak crossreactivity with the envelope antigens. Further characterization of the cultures revealed the presence of HIV-like particles and antigens detectable with antibodies Phyllis J Kanki and Myron E Essex from either SAIDS monkeys or people with AIDS. The sizes of the protein antigens detected by radioimmunoprecipitation analysis were similar to those of HIV-1. When these antigens were tested with sera from people with AIDS or healthy carriers, virtually all sera had antibodies cross-reactive to core antigens. This primate virus was named STLV-III due to its relationship to HIV-1 (which was then called HTLV-III and/or LAV), and later termed simian immunodeficiency virus or SIV (Biberfeld et al., 1987 ).

An animal model for AIDS was an important advance at the time, particularly given the close similarities of both the disease and the animal host. However, the origin of the virus remained a puzzle. Although cases of AIDS had been reported in Africa, at the time AIDS cases in Asia were considered quite rare. We tested the hypothesis that the natural primate host for SIV would be able to sustain infection without significant disease, and that this primate would most likely reside in Africa. At least half of the healthy wild-caught African green monkeys showed evidence of exposure to SIV on the basis of antibodies (Kanki et al., 1985a (Kanki et al., , 1986a . Although it was possible that SIV caused some type of disease that had not been recognized, it was clear that the disease did not resemble the lethal immunosuppressive syndrome found in Asian macaque monkeys. Similarly, captive African mangabey monkeys infected with SIV revealed no disease symptoms. The pathogenic effects of SIV appeared to be species-specific. Further studies of wild-caught Asian monkeys demonstrated that SIV was not a natural infection of these primates. It is generally believed that SIV was introduced to Asian macaques in captivity, explaining in part the unusual and high mortality induced by SIV.

More recent studies reveal that several African monkey species are infected with different SIVs. These include several species commonly described as African greens, such as vervet, grivet, sabaeus, and tantalus, as well as mona, diana, Sykes', mandrill, and sooty mangabey species (Tsujimoto et al., 1988; Johnson et al., 1989; Allan et al., 1991) . Thus far, SIVs have not been described in Asian species or in baboons, though these species can be infected in captivity with some primate lentiviruses. As a group, the SIVs are more closely related to HIV-2s than to HIV-1s, although some, such as the mandrill SIV, are evolutionarily distant (Tsujimoto et al., 1988) . The sooty mangabey monkey virus and the Senegalese human HIV-2, on the other hand, are essentially the same at the genetic level (Peeters et al., 1989; Essex, 1994; Kirchhoff et al., 1997) . It is this virus that also apparently accidentally infected the Asian macaques in captivity (Essex and Kanki, 1988) .

The possibility that SIV might cause disease rarely in African monkeys in advanced age could not be ruled out. The high prevalence of infection in wildcaught monkeys supported the notion that these African species of primates had evolved to a more benign coexistence with the virus in which infection did not affect survival. This was clearly different from the case of SIV in Asian primates or HIV-1 in people. Because wild macaques do not appear to be infected with SIV, and because the virus is limited to African primates, it appears likely that the virus accidentally infected captive rhesus monkeys in recent times. New SIVs are still being described and it is possible that species specificity may play a role in the level of host-virus adaptation and therefore levels of population infection (Georges-Courbot et al., 1998) . Currently it appears that natural SIV infection of African primates does not result in significant disease, supported by the high seroprevalence rates in most species evaluated. It thus appears that virus-host interaction has resulted in a non-pathogenic infection and the genetic, immunological and virus determinants of this relationship are worthy of further study.

We could speculate that the virus had either moved from subhuman primates to people prior to the mid-1950s or had been introduced to the cities through the migration of a few resistant carriers from a previously isolated group of people. However if HIV-1 had been present in rural areas, we might expect to find that Africans demonstrate greater resistance to infection and disease development, owing to genetic selection and evolution of the human species. In prospective studies conducted to date, Africans infected with HIV-1 appear to develop clinical AIDS and other signs and symptoms of HIV disease as rapidly as individuals in the United States or Europe (Mann et al., 1986; Marlink et al., 1994) . Furthermore, the degree of genomic variation seen in African isolates of HIV-1 was greater than that seen for viruses from Europe or the United States.

A virus that could be a progenitor of HIV-1 was isolated from a chimpanzee in Central Africa (Huet et al., 1990) . This finding, combined with the knowledge that all HIV-1 viruses tested appear to be avirulent when inoculated into chimpanzees, is also compatible with a subhuman primate origin for HIV-1. Some African isolates of HIV-1 appear to be as close to the chimpanzee isolate as to other prototype strains of HIV-1 (De Leys et al., 1990; Zekeng et al., 1994) . These findings appear to have been confirmed by more detailed genetic analysis of a virus from a chimpanzee housed in the United States, although the exact history and pedigree of this animal is unclear (Gao et al., 1999) .

Because a relative of HIV-1 -SIV -had been found in wild African monkeys and was only about 50 per cent related to HIV-1 at the genomic level, it seemed logical that viruses more highly related to SIV might also be present in human populations. Serum samples from West African prostitutes were examined to determine if they had antibodies that were more highly cross-reactive with SIV than with HIV-1 ( Barin et al., 1985) . Through Western blot and radioimmunoprecipitation methods, it became clear that a significant proportion of Senegalese prostitutes had antibodies that were highly reactive with all the major antigens of SIV detected by this technique. When the same SIV antigens were reacted with sera

from HIV-1-infected individuals of either European or Central African origin with classic disease manifestations of AIDS, little or no reaction was seen with the envelope antigens. The class of reactivity seen with serum samples from West African prostitutes was in fact virtually indistinguishable from that seen with serum samples from African monkeys or captive rhesus macaques (Kanki et al., 1986b; Hirsch et al., 1989; Essex, 1994) .

The HIV virus lifecycle requires transcription of the viral RNA into a DNA copy that becomes randomly integrated into the host cellular DNA. The transcription process is performed by the virus enzyme reverse transcriptase, which is error-prone. As a result, each round of replication results in at least one mutation per genome copy. The genetic variation of HIV is hierarchical, as depicted in the simplified schematic of subtype variation, interpatient variation, and intrapatient variation (Figure 1.1) . HIV-1 virus isolates from North Americans and Europeans showed distinct genetic variability compared with the variability in viruses from African patients Burke and McCutchan, 1997) . This variability was more dramatic than the recognized genetic variability between viral isolates from a single geographic region (i.e. inter-isolate variability) (Myers and Pavlakis, 1992; Korber et al., 1995) . In turn, this was also distinguishable from the genetic variation that was seen at the level of an individual patient (i.e. intrapatient variability). At the level of the individual patient, a swarm or quasi-species of highly related but distinguishable viral variants has been demonstrated throughout the course of HIV infection (Goodenow et al., 1989; Balfe et al., 1990; Delassus et al., 1991) . Genetic diversity is therefore a major characteristic of the HIV viruses and provides a major obstacle to drug and vaccine development.

Remarkably, all the HIVs-1 isolated from the United States and Western Europe through 1994 have been of a single subtype, B. Most of the diverse subtypes of HIV-1 have been found in sub-Saharan Africa. Subtypes A, C, and D in particular have been found more frequently than other subtypes in Africa. A high rate of spread of HIV-1 for Africa appeared during the 1980s, at about the same time the epidemic spread in the United States and Europe. The movement of populations and extensive international travel makes the likelihood of mixing subtypes inevitable, and non-B subtypes have already been identified, and are increasing in the United States and Europe.

In Asia, the introduction and spread of HIV-1 appeared about a decade later than in the West (see Chapter 15 for details). In Thailand, HIV-1 subtype B was detected in intravenous drug users during the mid-1980s. During the late 1980s, subtype E was first detected. By the early to mid-1990s, HIV-1 subtype E had spread very rapidly throughout heterosexuals in Thailand, with the highest rates in the northern regions of the country (Weniger et al., 1994) . Although apparently present earlier in the region, HIV-1 subtype B never spread to cause a major heterosexual epidemic as did HIV-1 subtype E. In China, the epidemic in IDUs is a unique recombinant virus of C and B. In heterosexual populations, subtype E appears to be the predominant subtype, similar to that of much of South-East Asia. This complex mixture of HIV-1 subtypes in Asia has challenged vaccine development efforts, since distinct subtype epidemics are being observed in different high-risk populations.

A similar situation occurred in India with HIV-1 subtypes B and C. While subtype B appeared to be introduced earlier among IDUs, this subtype did not spread as rapidly among heterosexuals as did subtype C. Previously associated with the massive heterosexual epidemic in southeastern Africa, subtype C also caused a rapid heterosexual epidemic in western India, initially spreading from the Bombay region (Jain et al., 1994; Weniger et al., 1994) . In the past 7-8 years, it is clear that subtype C is the sole viral subtype responsible for the newest and perhaps most frightening of HIV epidemics, worldwide. In most countries of southern Africa, rapid dissemination of HIV infection has been described with rates ranging from 10 to 40 per cent in pregnant women in most countries of the region. As a result of these significant increases in subtype C infection in Africa, and its predominance in places such as India and perhaps China, this viral subtype is responsible for over half of all HIV infections worldwide (UNAIDS, 2002 ).

An even more distant subtype, designated HIV-O, has been detected in Cameroon (Nkengasong et al., 1994) . The viruses isolated from this subtype are even less related to HIV-1 subtypes A through H than either of the other subtypes are related to each other, yet HIV-O is more related to HIV-1 than to HIV-2 . To emphasize this distance, HIV-1 subtypes A through H are designated the major group (M), and HIV-O is designated the outgroup (O) (Charneau et al., 1994) . Despite extensive serosurveys, the distribution of HIV-O appears to be quite restricted (Peeters et al., 1997) . While HIV-1 subtypes A through H probably had a common human progenitor ancestor, HIV-O no doubt entered independently from a chimpanzee host. HIV-2 almost certainly entered independently from monkey species native to West Africa (Essex and Kanki, 1988; Hirsch et al., 1989) .

The movement and distribution of HIV-1 subtypes throughout the world is often perplexing, particularly when subtypes such as E to H appear to be isolated more frequently in such places as Asia, South America, or eastern Europe than in Africa, where they presumably originated (Louwagie et al., 1993; Bobkov et al., 1994) . However, the viruses that have been isolated and characterized were acquired for analysis from convenience samples and therefore may suffer from extensive regional selection bias and inadvertent clustering. In the future, it will be important to develop more consistent surveillance methods and full-length sequence analysis to generate a true global map of HIV subtypes.

Our understanding of the epidemiology and biology of different HIV-1 subtypes is critical to future intervention efforts, and further studies are clearly needed (Anderson et al., 1996) . Studies have demonstrated differences in the ability of non-B and B subtype viruses to infect Langerhans' cells, a critical cell in heterosexual transmission of HIV. This suggests that the viral properties of non-B subtype viruses would facilitate heterosexual transmission and may have contributed to the dramatic epidemic spread in Asia and Africa (Soto-Ramirez et al., 1996) . A cross-sectional study of heterosexual couples in Thailand suggests a higher risk of heterosexual transmission of subtype E compared with subtype B (Kunanusont et al., 1995) . Studies in many African countries have described multiple HIV-1 subtypes, but it is not known if subtypes enter populations at different time points, or if the distribution of subtypes reflects the dynamics of different subtype-specific transmission potentials. Studies in South Africa demonstrate the association of certain subtypes with different modes of HIV transmission: subtype B viruses are associated with homosexual transmission, and non-B subtypes are associated with heterosexual transmission (Vanharmelen et al., 1997) . Future detailed studies of prospectively followed cohorts will be necessary to determine differences in pathogenicity and transmissibility of different subtypes.

Host selection of HIVs-1 for efficiency of heterosexual transmission may partially explain the high rates of heterosexual transmission seen with subtypes C and E (Soto-Ramirez et al., 1996) . HIV-1 subtype B, the major subtype in the developed world, appears to have undergone counterselection (i.e. less likely to be selected for compared with more virulent or transmissible viruses) to lose the phenotypic property of efficient heterosexual transmission. If efficient heterosexual transmission requires a particular genotype for vaginal infection, such sequences may have been partially lost by these HIV-1 B 'strains' that have been repeatedly passaged by blood exposure or rectal intercourse. Many of the non-B subtypes, on the other hand, could theoretically maintain vaginal phenotypic properties through regular heterosexual transmission in Africa and Asia (Kunanusont et al., 1995; Soto-Ramirez et al., 1996) .

Based on prospective studies of female sex workers in Dakar, Senegal, we have recently reported on disease progression in non-B subtype infections with known time of infection. In evaluating AIDS-free survival curves of women with incident subtype A, C, D, and G infection, we have shown distinct differences in AIDS-free survival (Kanki et al., 1999) . The comparison of non-A subtypes with subtype A demonstrated a significantly longer AIDS-free survival for women infected with subtype A. Due to the small sample size per subtype, our estimate of AIDS incidence should be considered imprecise, and further study is clearly warranted. Cross-sectional studies indicate a significant proportion of AIDS cases with subtype A infection in West and East Africa (PJ Kanki, unpublished data, 1998) . Further study of HIV-1 subtype natural history and progression from different geographic regions is clearly needed to better evaluate the role of viral subtype differences and AIDS pathogenesis.

Human immunodeficiency virus infection results in progressive loss of immune function marked by depletion of the CD4ϩ T lymphocytes, leading to opportunistic infections and malignancies characteristic of the syndrome termed AIDS. A number of host and viral factors influence the rate of disease progression. Studies in the West prior to the implementation of antiretroviral therapy suggested a median time to AIDS ranging from 8 to 10 years. Historically, CD4ϩ T lymphocyte counts provided the most reliable prognostic marker of HIV progression (Fahey et al., 1984) . In addition, other prognostic markers of progression to AIDS included immunologic markers of immune dysfunction which included cutaneous anergy (Redfield et al., 1986; Blatt et al., 1993) , serum ␤2-microglobulin, and neopterin levels (Fahey et al., 1990) . In the past, quantitation of viral infection was performed with imperfect and/or laborious methods. The serologic quantitation of the viral core antigen, p24 was frequently considered as a surrogate marker of high viral burdens (Allain et al., 1987; Dewolf et al., 1997) .

Quantitative viral culture was difficult to perform and not cost effective for clinical monitoring on a regular basis (Coombs et al., 1989; Ho et al., 1989) . HIV viral expression as measured by mRNA was also considered an important marker that preceded immunologic compromise (Gupta et al., 1993; Saksela et al., 1994) . More qualitative characteristics of the HIV virus infection included the syncytium-inducing properties of the virus, which were often coupled with viral burden; the syncytium-inducing viruses were associated with high viral loads and rapid progression and non-syncytium-inducing viruses associated with lower virus loads and slow progression (Tersmette et al., 1988; Fenyo et al., 1989) .

The development of the sensitive PCR-based technology to reliably quantitate RNA levels of virus in the plasma revolutionized our abilities to track viral infection. PCR technology is based on the use of a temperature-sensitive DNA polymerase. A target sequence is identified in the DNA sequence and appropriate primers are designed to span the sequence of interest. The primers need to be as similar to the original target sequence as possible. Once the primers anneal to the target sequence the temperature-sensitive polymerase begins to transcribe sequences, matching the sequence that it has found in the sample. When the temperature increases, the strands of DNA separate and the primer binding and polymerase reaction occurs again. In this way, a large number of exact DNA copies can be made given the cycles of temperature and the availability of primer, nucleotides, and polymerase.

Minute copies of DNA can be detected in large volumes of DNA with this very sensitive technology.

Mellors and colleagues provided important new evidence that such quantitation of plasma levels of RNA virus, or viral load, were important predictors of disease progression (Mellors et al., 1997) . Based on the large US cohort studies of homosexual men, the risk of developing AIDS and death was significantly associated with baseline plasma viral loads, independent of CD4ϩ lymphocyte counts (Mellors et al., 1995 (Mellors et al., , 1996 . It is worth noting that these studies and many of the ensuing viral load studies were conducted in the US and Europe, where subtype B HIV-1 infection is predominant. In viral load studies, as we will discuss further, it is critical to consider the populations studied, HIV viral subtypes, and methodologies employed.

Many studies that have followed the natural history of HIV infection over time have now reported on the utility of viral load determinations to track and predict disease progression. Since a variety of other markers, most notably lymphocyte subset data, were frequently available the relationship of viral load with CD4ϩ lymphocytes counts has been regularly evaluated. It has been important to evaluate the viral load's predictive value over the full course of HIV's natural history, and this has required analysis from long-standing cohort studies, where time of infection has been known and where observation times predated the use of antiretroviral therapy.

The large US Multi-center AIDS Cohort Study (MACS) has demonstrated a strong correlation in initial HIV RNA loads with declines in CD4 lymphocyte counts. Both initial HIV RNA levels and slopes were associated with AIDS-free times. HIV RNA load at the first seropositive visit, similar to three months after seroconversion, was highly predictive of AIDS, and subsequent HIV RNA measurements showed even better prognostic discrimination. However, HIV RNA slopes in the three years preceding AIDS and HIV RNA levels at AIDS diagnosis showed little variation according to total AIDS-free time (Lyles et al., 2000) . The MACS study population is largely a white homosexual male cohort, and it may not be possible to generalize all of these findings to other population groups with different modes of transmission and different viral subtypes.

Sabin and colleagues have reported a similar predictive value of plasma viral loads in the large European hemophilia cohorts where their results confirm the importance of the HIV RNA level in assessing the long-term prognosis in individuals infected with HIV (Sabin et al., 1998) . The risk of developing AIDS and death remained low when the HIV-1 RNA level was below 4 log 10 copies/ml, but increased rapidly thereafter, supporting current guidelines for the initiation of antiretroviral therapy after the viral load has exceeded this level (Sabin et al., 2000) . In the French SEROCO study of HIV seroconverters (n ϭ 330), patients who remained AIDS-free had lower early viral loads and, on average, a longer period of viral load decline after infection (36 versus 18 months), followed by a slower viral load increase compared with those who progressed to AIDS (Hubert et al., 2000) . A true plateau-phase after the seroconversion period was observed, lasting approximately four years, identified only in patients who remained AIDS-free for at least 90 months. In multivariate analysis, both early viral load and later changes were significant predictors of progression to AIDS (Hubert et al., 2000) .

In recent years, through the study of primary HIV infection, we have learned that some degree of virus containment occurs in the very early phases of HIV infection in vivo. During this critical period, a complex dynamic of infecting virus and responding host and immune factors leads to the establishment of a level of viremia, or viral setpoint, that appears predictive of subsequent HIV progression rates and survival (Mellors et al., 1995 (Mellors et al., , 1996 . The incubation period from initial infection to onset of symptoms is an average of 21.4 days (SD ϭ 9.6 days, range: 10-55 days) and the self-limited illness resolves within 1-3 weeks. Current data suggest that HIV viral load in the blood reaches a peak in the first 15-30 days, concurrent with a precipitous drop in CD4ϩ T cell count and increase in absolute number of CD8ϩ lymphocytes (Clark et al., 1991; Clark and Shaw, 1993; Clark and Wolthers, 2000) . Subsequent to these early events and tied to the resolution of the acute clinical syndrome is the lowering of HIV load and rebound of CD4ϩ T lymphocyte levels.

Corey and colleagues have reported considerable variability in the viral burden during these early phases of HIV infection, after 120 days after acquisition, followed by a rapid decrease in plasma HIV RNA levels to an inflection point, after which they gradually increase (Schacker et al., 1998) . The early infection phase continues over the next 6-12 months, with seeding and establishment of HIV viral load in blood and lymphoreticular tissues (Fauci, 1993 (Fauci, , 1996 Haynes et al., 1996) . The establishment of 'steady-state viremia' or viral setpoint is attained during this phase of infection and remains relatively invariant throughout much of the long incubation period.

It is now well established that the level of HIV-1 plasma viremia early in infection is highly predictive of future clinical course (Mellors et al., 1995 (Mellors et al., , 1996 Stein et al., 1997) . Cross-sectional studies of long-term non-progressors (LTNPs) as a group have demonstrated a significantly lower cell and plasma viral burden when compared with rapid progressors (Cao et al., 1995; Rinaldo et al., 1995; Pantaleo et al., 1995) . Increases in plasma viremia are correlated with increase in proviral burden, quantitative virus isolation, and quantity of virus in lymphoreticular tissue (Haynes et al., 1996) . The stability of virion-associated HIV-1 RNA levels suggests that an equilibrium between HIV-1 replication rate and efficacy of immunologic response is established shortly after infection and persists throughout the asymptomatic period of the disease. Thus, defective immunologic control of HIV-1 infection may be as important as the viral replication rate for determining AIDSfree survival.

In HIV-1 infection, treatment with highly active antiretroviral therapy (HAART) has been shown to drastically lower plasma viremia, and this is currently used as an indicator of the effectiveness of treatment (Obrien et al., 1998; Panther et al., 2000) . After discontinuing HAART, individuals had rebounds in their viral burdens approximating pre-HAART levels, even after a significant lapse of time approaching five years (Hatano et al., 2000) . In addition, the viral load at baseline is predictive of the rate of HIV-1 decline following antiretroviral therapy (Notermans et al., 1998) . Multiple studies have shown that the current repertoire of antiretroviral drugs is insufficient to completely eradicate HIV-1 from infected individuals (Wong et al., 1997; Finzi et al., 1997; Dornadula et al., 1999) .

In the era of HIV treatment, the use of sequential HIV RNA measurements may be more meaningful than any single measurement. The pre-treatment slopes of HIV-1 RNA decline in the acutely infected individuals increased significantly ( p ϭ 0.0001) after initiation of antiretroviral therapy. However, these posttreatment slopes were lower than those found in the chronically infected individuals ( p ϭ 0.012). Slopes were inversely correlated ( p ϭ 0.012) with baseline HIV-1 RNA (Putter et al., 2000) .

The combination of multiple adverse side-effects associated with HAART and the stringent demands for regimen adherence have prompted the design of new treatment strategies. Several studies have examined the longitudinal effects of multiple scheduled treatment interruptions where the long-term safety of this experimental protocol is largely unknown, particularly with respect to the generation of resistant viruses. However, such a treatment strategy would improve the likelihood of long-term adherence; the use of individual viral load plateaus will serve as an important guide in the aggressiveness and timing of treatment interruptions (Hatano et al., 2000) .

In just over two decades, the world has come to recognize a uniquely pathogenic new virus. The early epidemiologic pattern of the HIV/AIDS epidemic presented unique challenges to virologists, epidemiologists, clinicians, and public health officials. Throughout its 20-year history the epidemic continues to challenge us within our respective disciplines and in many ways has promoted an integration of these fields in an effort to curb its spread. Although the discovery of this new pathogen occurred relatively quickly after the first cases, there are still many questions as to how such a chronic virus infection can be so uniformly virulent 8-10 years post infection.

Many mysteries still surround the origin of the HIV virus and its relatives. The prevalence of the related simian viruses in African primates is so high that it suggests an ancient and stable virus-host relationship, and one that has developed into a virtually innocuous virus infection. As an intermediate, the HIV-2 virus is significantly less pathogenic than HIV-1, taking 20-30 years to develop AIDS, although the end-stage disease appears similar. We are still trying to discover the viral and host mechanisms by which this virus is able to persist for such a long period of time in an asymptomatic phase and yet somehow be triggered to cause fatal immunosuppression decades later.

Through molecular techniques such as PCR diagnostics and high-throughput sequencing we can now further appreciate the genetic diversity of HIV and SIVs. A decade ago it would take months to generate and analyze the sequence of a single HIV virus, a feat that this now readily achieved in a few days. We have also seen these technologies brought to bear on other epidemics of infectious diseases such as the recent severe acute respiratory syndrome (SARS) epidemic in Asia. There is no doubt that such new techniques will expedite our characterization of the ever-changing HIV epidemic and assist researchers in predicting the viral genotype that is relevant for targeting.

The genetic diversity of the virus at the level of the infected individual is now appreciated as a quasi-species rather than a single genotype that continues to challenge the immune system and clearly plays a role in the pathogenicity of the virus. It also presents a significant barrier to the current armamentarium of antiretroviral drugs that we can provide to our patients with the emergence of viral resistant variants that may continue to require better and different therapies to control virus infection over the long term. At present, we do not have such therapies to affect a cure and the various viral reservoirs and significant viral replication rate suggest that this will be nearly impossible in the near future. Genetic diversity in HIV evidenced by different strains or subtypes has yielded a complex global map of variant viruses, affecting our ability to diagnose, treat, and ultimately vaccinate against an ever-changing set of viruses. We already recognize that these subtypes infecting a single population can generate recombinant viruses at an incredible rate which might predict that vaccine development may never keep pace to provide sterilizing immunity.

New technologies have made significant advances in our research to better understand and control this virus. The ability to measure HIV-1 viral load has revolutionized our ability to track virus infection and replication in the patient. The advent of viral load assays has been both a technological as well as pragmatic feat, supplying a sturdy clinical measure for the management of AIDS patients. Formerly, clinicians and scientists relied on difficult and cumbersome plasma and peripheral blood mononuclear cell cultures for virus isolation or p24 antigen quantitation to monitor infection in the individual patient. The development of PCR-based methods to measure viral RNA has been particularly useful in following patients and their therapeutic management. Importantly the measurement of viral burden has provided new insights into the mechanisms of HIV transmission and pathogenesis.

The twenty-first century will no doubt bring new technologies and research advances to HIV/AIDS research. In the past few years we have seen the global community at least acknowledge the inequalities of the HIV/AIDS global epidemic, and renewed interest and support for prevention, treatment and vaccines are being directed to the developing world. This is where the HIV/AIDS epidemic is most complex from a virological point of view, with multiple subtypes and types of HIV viruses and the ever-growing proportion of complex recombinants. The need to test HIV interventions and vaccines in these parts of the world is therefore critical, since their effects on the HIV/AIDS epidemic in these populations will have the largest impact. As scientists we need to continue our efforts to insure that good science, public health, and effective health policy will end the HIV/AIDS epidemic in the next century.

Long-term evaluation of HIV antigen and antibodies to p24 and gp41 in patients with hemophilia. Potential clinical importance

Species-specific diversity among simian immunodeficiency viruses from African green monkeys

Acquired immune dysfunction in homosexual men: immunologic profiles

Implications of genetic variability in HIV for epidemiology and public health

Epidemiologic aspects of the current outbreak of Kaposi's sarcoma and opportunistic infections

Concurrent evolution of human immunodeficiency virus type 1 in patients infected from the same source: rate of sequence change and low frequency of inactivating mutations

Serological evidence for virus related to simian T-lymphotropic retrovirus III in residents of West Africa

Isolation of T-lymphotropic retrovirus from a patient at risk for acquired immune deficiency syndrome (AIDS)

Meeting report: WHO working group on characterization of HIV-related retroviruses: Criteria for characterization and proposal for a nomenclature system

Delayed-type hypersensitivity skin testing predicts progression to AIDS in HIV-infected patients

Identification of an env G subtype and heterogeneity of HIV-1 strains in the Russian Federation and Belarus

Global Distribution of Human Immunodeficiency Virus-1 Clades in: Rosenberg SA

Virologic and immunologic characterization of long-term survivors of human immunodeficiency virus type 1 infection

Isolation and envelope sequence of a highly divergent HIV-1 isolate: definition of a new HIV-1 group

T-cell dynamics and renewal in HIV-1 infection

The acute retroviral syndrome and the pathogenesis of HIV-1 infection

High titers of cytopathic virus in plasma of patients with symptomatic primary HIV-1 infection

Acquired immune deficiency syndrome in Black Africans

Plasma viremia in human immunodeficiency virus infection

Acquired immunodeficiency syndrome (AIDS) associated with transfusions

Acquired immunodeficiency syndrome in a patient with hemophilia

Isolation and partial characterization of an unusual human immunodeficiency retrovirus from two persons of west-central African origin

Evolution of human immunodeficiency virus type 1 nef and long terminal repeat sequences over 4 years in vivo and in vitro

Aids prognosis based on HIV-1 RNA, CD4ϩ T-cell count and function -markers with reciprocal predictive value over time after seroconversion

Residual HIV-1 RNA in blood plasma of patients taking suppressive highly active antiretroviral therapy

The acquired immunodeficiency syndrome and Mycobacterium avium-intracellulare bacteremia in a patient with hemophilia

Adult T-cell leukemia/lymphoma: Role of a human retrovirus

Simian immunodeficiency virus in people

The origins of the AIDS virus

Antibodies to cell membrane antigens associated with human T-cell leukemia virus in patients with AIDS

Distribution of antibodies to HTLV-MA in patients with AIDS and related control groups in: Gross L, ed. Human T-cell Leukemia Viruses

Retroviruses associated with leukemia and ablative syndromes in animals and in human beings

Quantitative changes in T helper or T suppressor/ cytotoxic lymphocyte subsets that distinguish acquired immune deficiency syndrome from other immune subset disorders

The prognostic value of cellular and serologic markers in infection with human immunodeficiency virus type 1

Immunopathogenesis of HIV infection

Host factors and the pathogenesis of HIV-induced disease

Replicative capacity, cytopathic effect and cell tropism of HIV

Identification of a reservoir for HIV-1 in patients on highly active antiretroviral therapy

Epidemic acquired immune deficiency syndrome (AIDS): Epidemiologic evidence for a transmitted agent

Frequent detection and isolation of cytopathic retroviruses (HTLV-III) from patients with AIDS and at risk for AIDS

Origin of HIV-1 in the chimpanzee Pan troglodytes troglodytes

Molecular cloning and comparative analyses of the genomes of simian sarcoma virus and its associated helper virus

Proviral DNA of a retrovirus, human T-cell leukemia virus, in two patients with

Natural infection of a household pet red-capped mangabey (Cercocebus torquatus torquatus) with a new simian immunodeficiency virus

HIV-1 isolates are rapidly evolving quasispecies: evidence for viral mixtures and preferred nucleotide substitutions

Pneumocystis carinii pneumonia and mucosal candidiasis in previously healthy homosexual men: evidence of a new acquired cellular immunodeficiency

Enhanced expression of human immunodeficiency type 1 correlates with development of AIDS

A new subtype of human immunodeficiency virus type 1 (MVP-5180) from Cameroon

Pre-HAART HIV burden approximates post-HAART viral levels following interruption of therapy in patients with sustained viral suppression

Toward an understanding of the correlates of protective immunity to HIV infection

Epidemic of acquired immunodeficiency in rhesus monkeys

An African primate lentivirus (SIVsm) closely related to HIV-2

Quantitation of human immunodeficiency virus in the blood of infected persons

Rapid turnover of plasma virions and CD4 lymphocytes in HIV-1 infection

Natural history of serum HIV-1 RNA levels in 330 patients with a known date of infection

Genetic organization of a chimpanzee lentivirus related to HIV-1

Transfusion-associated AIDS: serologic evidence of human T-cell leukemia virus infection of donors

Epidemiology of HIV and AIDS in India

Genetic diversity among simian immunodeficiency virus isolates from African green monkeys

A new subtype of human T-cell leukemia virus (HTLV-II) associated with a T-cell variant of hairy cell leukemia

Antibodies to simian T-lymphotropic retrovirus type III in African Green monkeys and recognition of STLV-III viral proteins by AIDS and related sera

Serologic identification and characterization of a macaque T-lymphotropic retrovirus (HTLV) type III

The Pathobiology of Macaque Retroviruses closely related to Human T-cell Lymphotropic Viruses

New human T-lymphotropic retrovirus related to simian T-lymphotropic virus type IIIAGM (STLV-IIIAGM)

Human immunodeficiency virus type 1 subtypes differ in disease progression

Simian immunodeficiency virus variants with differential T-cell and macrophage tropism use CCR5 and an unidentified cofactor expressed in CEMX174 cells for efficient entry

Heterogeneity of HIV-1 and HIV-2

HIV-1 subtypes and male-to-female transmission in Thailand

Correlation between immunologic function and clinical subpopulations of patients with the acquired immune deficiency syndrome

Acquired immunodeficiency syndrome in a colony of macaque monkeys

Genetic subtypes of HIV-1

Natural history of human immunodeficiency virus type 1 viremia after seroconversion and proximal to AIDS in a large cohort of homosexual men

Natural history of human immunodeficiency virus infection in Zaire

Reduced rate of disease development after HIV-2 infection as compared to HIV-1

An outbreak of community-acquired Pneumocystis carinii pneumonia: initial manifestation of cellular immune dysfunction

HIV-1 genetic diversity

Quantitation of HIV-1 RNA in plasma predicts outcome after seroconversion

Prognosis in HIV-1 infection predicted by the quantity of virus in plasma

Plasma viral load and CD4(ϩ) lymphocytes as prognostic markers of HIV-1 infection

Evidence for human infection with an HTLV-III/ LAV-like virus in Central Africa

Genotypic subtypes of HIV-1 in Cameroon

Rate of HIV-1 decline following antiretroviral therapy is related to viral load at baseline and drug regimen

Longitudinal HIV-1 RNA levels in a cohort of homosexual men

Immune deficiency syndrome in children

HIV infection is active and progressive in lymphoid tissue during the clinically latent stage of disease

Studies in subjects with long-term nonprogressive human immunodeficiency virus infection

Genital tract human immunodeficiency virus type 1 (HIV-1) shedding and inflammation and HIV-1 env diversity in perinatal HIV-1 transmission

Isolation and partial characterization of an HIV-related virus occurring naturally in chimpanzees in Gabon

Acquired immunodeficiency syndrome in a heterosexual population in Zaire

Detection and isolation of type C retrovirus particles from fresh and cultured lymphocytes of a patient with cutaneous T-cell lymphoma

Acquired immunodeficiency syndrome with Pneumocystis carinii pneumonia and Mycobacterium avium-intracellulare infection in a previously healthy patient with classic hemophilia. Clinical, immunologic, and virologic findings

Detection, isolation, and continuous production of cytopathic retroviruses (HTLV-III) from patients with AIDS and pre-AIDS

Slower decline of plasma HIV-1 RNA following highly suppressive antiretroviral therapy in primary compared with chronic infection

The Walter Reed staging classification for HTLV-IIILAV infection

High levels of anti-human immunodeficiency virus type 1 (HIV-1) memory cytotoxic T-lymphocyte activity and low viral load are associated with lack of disease in HIV-1-infected long-term nonprogressors

Acquired immunodeficiency with reversed T4/T8 ratios in infants born to promiscuous and drug-addicted mothers

Immune markers and viral load after HIV-1 seroconversion as predictors of disease progression in a cohort of haemophilic men

Course of viral load throughout HIV-1 infection

Human immunodeficiency virus type 1 mRNA expression in peripheral blood cells predicts disease progression independently of the numbers of CD4ϩ lymphocytes

Evidence for exposure to HTLV-III in Uganda before 1973

Biological and virologic characteristics of primary HIV infection

Serological analysis of a subgroup of human T-lymphotropic retroviruses (HTLV-III) associated with AIDS

Acquired immunodeficiency syndrome in infants

Severe acquired immunodeficiency in male homosexuals, manifested by chronic perianal ulcerative herpes simplex lesions

HIV-1 Langerhan's cell tropism associated with heterosexual transmission of HIV

Predicting clinical progression or death in subjects with early-stage human immunodeficiency virus (HIV) infection -a comparative analysis of quantification of HIV RNA, soluble tumor necrosis factor type i receptors, neopterin, and beta(2)-microglobulin

Differential syncytium-inducing capacity of human immunodeficiency virus isolates: frequent detection of syncytium-inducing isolates in patients with acquired immunodeficiency syndrome (AIDS) and AIDS-related complex

Isolation and characterization of simian immunodeficiency virus from mandrills in Africa and its relationship to other human and simian immunodeficiency viruses

Joint United Nations Programme on HIV/AIDS (UNAIDS)

Acquired immunodeficiency syndrome in Rwanda

An association between HIV-1 subtypes and mode of transmission in Cape Town

Viral dynamics in human immunodeficiency virus type 1 infection

The molecular epidemiology of HIV in Asia

In vivo compartmentalization of human immunodeficiency virus: evidence from the examination of pol sequences from autopsy tissues

Killing of primary CD4ϩ T cells by non-syncytium-inducing macrophage-tropic human immunodeficiency virus type 1

Prevalence of HIV-1 subtype O infection in Cameroon: preliminary results