Poster Abstracts

  1. Gert Van der Auwera, et al.: Reanalysis of full-length HIV-1 group M subtype K and sub-subtype F2 with an MS-DOS based bootscanning program
  2. Georg A. Funk, et al.: Viral Fitness in Different HIV-1 Infected Cell Compartments
  3. Vladimir V. Lukashov and Jaap Goudsmit: Molecular Clock In The Recent Evolutionary History Of HIV-1
  4. Monty A. Montano: Determinants of HIV-1C Regulatory Evolution
  5. Miguel Angel Martinez, Angela Ibañez and Bonaventura Clotet: Absence of genetic diversity reduction in the HIV-1 integrated proviral LTR sequence population during successful combination therapy
  6. Rani Meeta, Anand Ramaswami and Françoise Seillier-Moiseiwitsch: Linking env proteins from the HIV1 virus to co-receptor usage
  7. David H. O'Connor, et al.: CTL Responses and Viral Evolution During the Acute Phase of SIV Infection
  8. G. V. Quinnan, et al.: HIV-1 Primary Virus Neutralization Resistance is Determined by Multiple LZ Interactions within the Envelope, Causing Decreased sCD4 Sensitivity and Increased Infectivity
  9. M.E. Quiñones-Mateu, et al.: A Rapid Competition Assay Shows A Direct Correlation Between HIV-1 Fitness And Disease Progression
  10. Roland R Regoes and S. Bonhoeffer: HIV tropism and drug treatment
  11. Marco Salemi, et al.: Uncovering the origin of viral epidemics with a new method to calibrate clock-like phylogenetic trees
  12. Daniel Shriner, Allen Rodrigo and James I. Mullins: Inferring Phylogenies for Large, Serially Sampled Data Sets
  13. Daniel Shriner, Allen Rodrigo and James I. Mullins: Phylogenetic Evidence for Recombination within a HIV-1 Singly Infected Individual
  14. Shambavi Subbarao, et al.: Evidence for transmission networks based on phylogenetic analysis of demographic and temporal clustering among incident HIV-1 infections, within a prospective injecting drug user cohort in Bangkok, Thailand
  15. Florin Vaida and Tony Fitzgerald: Adjusting for Assay Detection Limit in Model-Based Estimation of HIV RNA Decay During HAART
  16. Yang Wang, et al.: HIV-1 Populations in Blood and Semen
  17. C.A. Womack, et al.: Reevaluation Of Amino Acid Variability Of The HIV-1 Gp120 And Prediction Of New Discontinuous Epitopes
  18. Yumi Yamaguchi-Kabata and Takashi Gojobori: Reevaluation Of Amino Acid Variability Of The HIV-1 Gp120 And Prediction Of New Discontinuous Epitopes


Reanalysis of full-length HIV-1 group M subtype K and sub-subtype F2 with an MS-DOS based bootscanning program

Gert Van der Auwera, Wouter Janssens, Leo Heyndrickx, and Guido van der Groen

Recently, five new complete HIV-1 group M genome sequences were published (Triques et al., Aids Res. Hum. Retroviruses 2000;16:139-151). One of these clustered consistently with subtype F sequences, while two others were identified as representatives of a subcluster within the subtype F clade called F2, and the two remaining sequences were described as a new subtype K. We reanalyzed these sequences by means of bootscanning and phylogeny using a newly developed MS-DOS based bootscanning program. Although our analysis confirms the existence of the new subtype K, it does not support the F2 cluster in that it indicates that the F2 sequences are in fact recombinants between subtype F and a still unidentified subtype. If the F2 sequences continue to be treated as a subcluster of subtype F, future analysis of other subtype F related isolates may be seriously hampered, as subtype F would not be monophyletic.

Back to the Top


Viral Fitness in Different HIV-1 Infected Cell Compartments

Georg A. Funk, Marek Fischer, Beda Joos, Richard W. Cone, Milos Opravil, Bruno Ledergerber and Sebastian Bonhoeffer

Mathematical models have lead to fundamental new insights into the replication dynamics of HIV-1 in vivo and its pathogenesis. Continuing this approach, we present a population dynamical model describing CD4+ cells, actively, latently, persistently and defectively HIV-1 infected cells, respectively, and free virus in 21 patients under anti-retroviral treatment. The model was numerically solved and simultaneously fitted to clinical data of HIV-1 infected cell compartments. We found that <1% of the activated CD4+ cells in the blood were HIV-1 infected before treatment. Our estimated half-lives of free virus and productively infected cells further support previously published data. Further, we estimated the basic reproductive ratio of the virus in different HIV-1 infected cell compartments. Although the burst size was higher in persistently infected cells than in actively virus producing cells, our model indicated considerably higher basic reproductive ratios of the virus in actively virus producing cells compared to the other HIV-1 infected cell compartments together, because most of the newly infected cells become actively virus producing cells.

Back to the Top


Molecular Clock In The Recent Evolutionary History Of HIV-1

Vladimir V. Lukashov and Jaap Goudsmit

Department of Human Retrovirology, Academic Medical Center, University of Amsterdam, Meibergdreef 15, 1105 AZ Amsterdam, The Netherlands

The molecular clock hypothesis is widely used to estimate dates of species divergence and to reconstruct phylogenetic relationships among organisms. In the present study, we attempted to test whether the molecular clock is operational in the recent history of HIV-1 subtype B and, if so, to reconstruct the evolutionary history of the global HIV-1 subtype B epidemic. We analyzed evolutionary distances to the common ancestor of HIV-1 env (V3) and pol (prot and partial RT) sequences obtained from several outbreaks of the global HIV-1 epidemic — those in the US as well as among homosexual men and injecting drug users (IDUs) in The Netherlands. The onset of all of these epidemics is known based on intensive retrospective epidemiological data, so we had an opportunity to study which phylogenetic model and which genomic region of HIV would provide the most reliable estimate for the time of recent HIV-1 introduction into a human population. Our analysis of evolutionary distances of the US V3 sequences to their reconstructed common ancestor revealed that, according to both the ML and NJ methods, there is a significant positive correlation between the sampling years of sequences and their nucleotide distances to the common ancestor. There was a strong correlation between evolutionary distances of individual sequences to the common root, estimated by both methods (correlation — 0.81, p<0.001). The years of HIV-1 introduction into the US, calculated by both methods based on nucleotide distances, were 1953 and 1967 (for the NJ and ML methods, respectively), which clearly has to be considered an overestimation of the age of the US epidemic (known to start in the mid-70's). Subsequent separate analysis of synonymous and nonsynonymous distances revealed the reason for this discrepancy. There was a highly significant positive correlation between the sampling years of the US V3 sequences and their synonymous distances to the common ancestor (p<0.00001), while nonsynonymous distances of the sequences to the common ancestor were not significantly increasing (p>0.1) during the observation period (1983-1994). The extrapolation of the regression line of synonymous distances back to the date when no synonymous heterogeneity was present in the US HIV-1 population allowed to estimate 1976 (95% CI 1974-1977) as a year of HIV-1 introduction into the USA, which is in agreement with epidemiological data on the history of HIV-1 epidemic in the US. Similar to our findings for the US epidemic, both the ML and NJ methods, when based on nucleotide distances, robustly underestimated the date of virus introduction into Dutch homosexual men and IDUs. Subsequent separate analysis of nonsynonymous and synonymous distances revealed that this underestimation is resulting from nonsynonymous distances, which were not increasing significantly in both virus populations over the observation period. At the same time, analysis of synonymous distances provided accurate estimates for the years of virus introductions into both risk groups, that were in complete agreement with epidemiological data (1977, 95% CI 1976-1979, and 1980, 95% CI 1979-1981, for homosexual men and IDUs, respectively). In contrast to the V3 region, analysis by either the ML or NJ method of nucleotide distances of pol sequences from Dutch homosexual men and IDUs provided accurate estimates for the years of virus introductions in these risk groups.

Back to the Top


Determinants of HIV-1C Regulatory Evolution

Monty A. Montano

The past ten years of HIV-1 genome research has seen extraordinary advances in our understanding of the extent of HIV-1 genetic diversity within individuals and populations and has identified the global presence of at least 5 major subtypes (A-E). While it is well recognized that the global epidemic is quickly becoming dominated by HIV-1C, the potential link between epidemic expansion, genetic-subtype variation, and an improved biological fitness has only recently begun to be fully recognized and should now be considered a research priority Many studies now suggest a capacity for HIV-1C to be a comparatively more aggressive virus than other subtypes. These features include a preferential use of the HIV-1 co-receptor CCR5, an increased propensity for activation in response to elevated levels of cytokines (e.g., TNF-a), an increased viral load compared to other co-circulating subtypes, and an apparent increase in perinatal transmission in utero. We present evidence, based on studies in our laboratory and others, that may collectively support a role for regulatory evolution of HIV-1C that may provide a viral genetic component to the apparent differential expansion of HIV-1C globally.

Back to the Top


Absence of genetic diversity reduction in the HIV-1 integrated proviral LTR sequence population during successful combination therapy.

Miguel Angel Martinez, Angela Ibañez and Bonaventura Clotet

Fundacio irsiCaixa, Hospital Universitari Germans Trias i Pujol, 08916 Badalona, Spain

By measuring over time levels of unintegrated and integrated cell-associated viral DNA we have previously investigated the reservoir of HIV-1 present in PBMCs of 10 infected patients on highly active anti-retroviral therapy (HAART) (Ibañez et al., AIDS 13: 1045-1049, 1999). Concordant with the decline of virus in plasma, a significant decrease in total HIV-1 DNA was observed after 48 weeks of therapy. However, no statistically significant reduction was detected when copy numbers of integrated HIV-1 DNA were compared. We present here the HIV-1 integrated provirus sequence variation in the PBMCs from four of these patients (B, C, D and H). Integrated proviral fragments of LTR taken from four time points were PCR amplified from PBMCs. The study period included a naïve phase and 1-2 years of therapy. Endpoint dilution of PBMC DNA was performed before nested PCR to make products derived from a single provirus, which were then directly sequenced. Ten individual clones were sequenced for each time point sample. The mean intrasample genetic distances at the beginning of treatment was 1.7%, 2.0%, 1.1% and 2.6% for patients B, C, D and H, respectively. After 48 weeks of therapy no significant change in the intrasample genetic distances was noted (1.6%, 2.3%, 0.4% and 2.3%, respectively). Moreover, samples collected between 48 and 108 weeks of therapy had very similar genetic diversity (1.4%, 1.9%, 0.4% and 2.0%, respectively) compared with the corresponding baseline or 48 weeks samples. Phylogenetic reconstruction of all LTR sequences showed that clones clustered in a patient-specific manner. Phylograms for each patient revealed an intermingling of sequences from the four time points, suggesting an absence of genetic change in this locus after the introduction of treatment. In conclusion, the data presented here imply that 1-2 years of successfully HAART does not significantly reduce the genetic repertoire of the integrated reservoir of HIV-1 present in latently infected CD4 T cells.

Back to the Top


Linking env proteins from the HIV1 virus to co-receptor usage

Rani Meeta, Anand Ramaswami and Françoise Seillier-Moiseiwitsch

Department of Biostatistics, School of Public Health University of North Carolina at Chapel Hill Chapel Hill, NC 27599-7400

We have carried out an analysis of the env protein sequences available in the Los Alamos HIV Database for which we know the cell co-receptor usage (Dorms et al., 1998). We performed a multiple sequence alignment of these 47 proteins (typically over 800 residues long). The alignment is based on a model described by Lamers et al. (1996). We utilized eight different methods to predict the secondary structure of these proteins and considered the consensus predictions. Next we applied a classification method to correlate the amino acids to the information regarding the co-receptor usage. We also correlate the secondary structures to the latter.

Back to the Top


CTL Responses and Viral Evolution During the Acute Phase of SIV Infection

David H. O'Connor*, Todd M. Allen*, Edward J. Dunphy*, Peicheng Jing*, Bianca R. Mothe*, Kevin J. Kunstman#, Xiaochi Wang, John L. Dzuris§, David T. Evans%, Ronald C. Desrosiers%, John D. Altman, Alessandro Sette§, Steven M. Wolinsky#, and David I. Watkins*^

Massive viral replication and the subsequent generation of robust cytotoxic T-lymphocyte (CTL) responses characterize the acute phases of simian immunodeficiency virus (SIV) and human immunodeficiency virus (HIV) infections. These early events are difficult to study in HIV-infected patients since the exact time of infection and the composition of the infecting virus is often unknown. The SIV-infected rhesus macaque model facilitates analysis of the evolution of the virus in the face of this strong immune response during primary infection. In a previous study, we showed that SIV-specific CTL select for amino acid variation during chronic infection in ten of ten CTL epitopes. Given the magnitude of CTL responses and the extent of viral replication during acute viremia, we reasoned that similar selection for CTL epitope variants should occur during the first month of infection. We infected 10 Mamu-A*01 positive rhesus macaques with pathogenic, molecularly cloned SIVmac239 and monitored CD8 responses using seven MHC-Class I tetramers folded with newly identified Mamu-A*01-restricted epitopes. Antigen-specific CD8 positive lympocyte responses peaked between three and four weeks post-infection coincident with a decline in plasma viremia. We are currently investigating whether the robust Mamu-A*01-restricted responses selected for CTL epitope variants during the acute phase.

Back to the Top


HIV-1 Primary Virus Neutralization Resistance is Determined by Multiple LZ Interactions within the Envelope, Causing Decreased sCD4 Sensitivity and Increased Infectivity

G. V. Quinnan, E. J. Park, P. F. Zhang and D. S. Dimitrov*

We investigated the mechanisms of neutralization resistance of a primary isolate of HIV-1-MN (MN-P), compared to the neutralization sensitive (NS) TCLA-MN virus (MN-NS), and a neutralization resistant (NR) variant (MN-NR). MN-P envelope was 250-fold more, and MN-NR was about 10-fold more, NR and highly infectious (HI) than MN-NS. A larger number of mutations distinguished MN-P than MN-NR from MN-NS. Three distinct interdomain interactions contributed independently to both the NR and HI phenotypes of MN-P. Specifically, the gp41 leucine zipper (LZ) interacts with: (1) the gp120 C1-C2 region; (2) the gp120 C2-C3 and C3-V5 regions in a tripartite interaction; and (3) the gp41 membrane proximal alpha helix (AH). Among 21 MN-NS/MN-P recombinants or mutants compared, involving tests of all three interactions, NR and HI were highly correlated (Corr. Coeff.=0.94). MN-NS and MN-P were distinguished by 19 non-V1/V2, or V3 gp120 amino acid substitutions, one in the coreceptor binding site, two that bind CD4, and most of the rest located around the CD4 binding pocket. Two residues at which mutations were critical for distinction of MN-NS and MN-NR were also mutated in MN-P. MN-P was globally NR, except for retained sensitivity to the conformation-dependent apical V3 epitope. It was highly resistant to sCD4. Since the MN-NR and MN-P sequences are more closely related to consensus sequence than that of MN-NS, the findings are generally relevant to understanding primary virus neutralization resistance. The results suggest that primary virus neutralization resistance is associated with decreased dependence on CD4 interaction, complemented by increased efficiency of coreceptor interaction or fusion competence. The findings are pertinent to definition of target epitopes for vaccine design and HIV evolution.

Back to the Top


A Rapid Competition Assay Shows A Direct Correlation Between HIV-1 Fitness And Disease Progression

Quiñones-Mateu, M.E.; Ball, S.C.; Marozsan, A.J; Torre, V.S.; Albright, J.L.; Vanham, G.; Guido Van Der Groen, G., and Arts, E.J.

Rapid evolution of human immunodeficiency virus type 1 (HIV-1) gives rise to drug resistance, escape from immune responses and failure of most single-strain vaccines. Previous studies have suggested that isolates of different clades or even isolates of the same subtype may show variations in replication efficiency/fitness when infecting primary cell isolates. However, few studies have examined the impact of HIV fitness on disease progression. In this study we have developed a competition assay in PBMC to measure the ex vivo fitness of any primary HIV-1 isolates. For viral fitness studies, we established conditions for dual HIV-1 infections of PBMC and a sensitive heteroduplex tracking assay (HTA) to measure relative virus production. Briefly, HIV-1 env DNA (a ~1.6 Kb fragment corresponding to gp120-coding region) was PCR-amplified from infected cells using a pair of universal HIV-1 primers. Env DNA products from the co-infections were then annealed to reference probes for both clades. Production of both HIV-1 isolates in each competition was analyzed by HTA and compared with initial inoculums to calculate a relative fitness value for each isolate. Using a pair wise fitness comparison of various nonsyncytium-inducing (NSI)/CCR5-tropic (R5) and syncytium-inducing (SI)/CXCR4-tropic (X4) HIV-1 isolates, four control strains with moderate ex vivo fitness were selected and competed against primary HIV-1 isolates from an HIV-infected Belgian cohort. HIV-1 isolates from long-term survivors (LTS) were out-competed by control strains and were significantly less fit than HIV-1 isolates from patients with accelerated progression to AIDS (PRO). In addition, ex vivo HIV-1 fitness in both PRO and LTS showed direct and independent correlations with viral load, suggesting that HIV-1 fitness together with viral load may be a strong predictor for the rate of disease progression. Based on these preliminary findings, HIV-1 fitness appears to influence and predict HIV-1 disease progression.

Back to the Top


HIV tropism and drug treatment

Roland R Regoes and S. Bonhoeffer

The human immunodeficiency virus (HIV) targets macrophages in the beginning of the infection, and in the further course gradually shifts towards a preference for T cells. The preference of HIV variants for certain target cell types is referred to as tropism. Some anti-retroviral agents selectively inhibit HIV phenotypes depending on their tropism. We analyze mathematical models, which describe the in vivo interaction of HIV phenotypes, differing in their tropism, with two target cell types (macrophages and T cells). We determine the circumstances, under which the virus population specializes for a certain target cell type, and compare the dynamics of treated with untreated infections. In particular, we investigate the conditions for a treatment-induced phenotype switch.

Back to the Top


Uncovering the origin of viral epidemics with a new method to calibrate clock-like phylogenetic trees.

Marco Salemi, Korbinian Strimmer, William W. Hall, and Anne-Mieke Vandamme

When all the viral strains in a phylogenetic tree exhibit the same evolutionary rate, i.e. they evolve according to the molecular clock hypothesis, the estimation of the time of origin of the most recent common ancestor for each clade of the tree can be easily computed. In fact, in such a tree, the degree of sequence divergence, as measured by the number of substitutions along a branch, is linearly proportional to the time of divergence. However, several observations suggest that viruses such as HIV or HCV do not follow a molecular clock. To overcome these limitations, we developed a new method, called site stripping for clock detection (SSCD), which allows selection of nucleotide sites in a set of aligned sequences evolving at an equal rate in different lineages. The method was validated employing a dataset of HCV patients all infected by the same donor in 1977 and it was able to exactly date the "known" origin of the HCV infection in that cohort. On the other hand, molecular clock dating on the dataset without using the SSCD procedure gave wrong estimates. Our results show that the new method can be reliable used to construct clock-like phylogenetic trees and to calculate divergence times at ancestral nodes. Identifying the sites that distort the molecular clock, the SSCD procedure also allows us to potentially investigate the selective forces leading to differential evolutionary rates.

Back to the Top


Inferring Phylogenies for Large, Serially Sampled Data Sets

Daniel Shriner, Allen Rodrigo and James I. Mullins

Although computational power is increasing in terms of speed and memory, the size and complexity of phylogenetic analyses are also increasing. This problem is especially evident in the case of serial sampling, e. g., when following HIV-1 infected patients longitudinally. Whereas there are numerous simulation studies that have examined the behavior of various algorithms for inferring phylogenies, few have examined phylogenetic inference for large data sets (where large is greater than20). Hence, we investigated computational procedures for inferring phylogenies of 50 and 100 sequences. We addressed three issues: 1. the utility of backbone constraints to account for the time structure present in serially sampled data sets; 2. trees with long diameters and short branches, which are common with serially sampled data sets; and 3. large data sets. We evaluated 10 combinations of optimality criteria and branch swapping algorithms. The results suggest that building a neighbor-joining tree using maximum likelihood distances and then swapping using the subtree-pruning and regrafting algorithm under maximum likelihood is the best compromise between accuracy and speed.

Back to the Top


Phylogenetic Evidence for Recombination within a HIV-1 Singly Infected Individual

Daniel Shriner, Allen Rodrigo and James I. Mullins

Whereas the env gene has been extensively studied in terms of its molecular evolutionary dynamics, much less is known about how other regions of the HIV-1 genome evolve. These studies were initiated with the aim of examining how the six auxiliary genes contemporaneously evolve. Initial phylogenetic analyses revealed topological incongruence among the auxiliary genes. Subsequent analyses using parametric bootstrapping support the notion of frequent recombination, leading to essentially independent assortment of auxiliary genes, within the quasispecies of a singly infected individual.

Back to the Top


Evidence for transmission networks based on phylogenetic analysis of demographic and temporal clustering among incident HIV-1 infections, within a prospective injecting drug user cohort in Bangkok, Thailand.

Shambavi Subbarao1, Lily Nguyen1, Dale J. Hu1, Suphak Vanichseni2, Kachit Choopanya2, Nancy Young1,3, Frits van Griensven3, Timothy D. Mastro1,3.

1: Centers for Disease Control and Prevention, Atlanta, GA. 2: Bangkok Metropolitan Administration, Bangkok, Thailand. and 3: The HIV/AIDS Collaboration, Nonthaburi, Thailand.

Background and Objective: During 1995-96, 1,209 HIV-1-negative IDUs attending 15 methadone treatment clinics in Bangkok were enrolled into a prospective cohort study. The overall HIV-1 incidence rate through 1998 was 5.8 per 100 person-years (PY) of follow up. During August-December 1996, the incidence rate doubled to about 12 per 100 PY. Of 130 seroconvertors identified, sequence analysis of the C2-V4 region of gp120 showed that 103 (79%) were infected with HIV-1 subtype E and 27 (21%) with HIV-1 subtype B. The objective of the study was to explore potential HIV-1 transmission networks in relation to temporal and demographic factors. Methods: All samples used for this study were collected at approximately 3-4 months and again at 12 months following the estimated date of seroconversion (EDS). The C2-V4 region was amplified by nested PCR. Multiple alignments were performed using GCG (Wisconsin package). Phylogenetic analyses were performed with the neighbor-joining distance method using CLUSTAL, PHYLIP, and MEGA programs. Tree topologies were confirmed with maximum-likelihood trees using fastDNAml and with maximum- parsimony trees using PHYLIP. Amino-acid alignments were generated using DNASIS. Results: 1) 6 phylogenetically significant clusters (bootstrap values >80%) were noted for 20/103 subtype E strains which were demographically (same clinic) and/or temporally (similar EDS) associated. Seroconvertors from clinic 9 (n=5), clinic 6 (n=3), and clinic 7 (n=2) were part of 3 strong clinic-based clusters. Seroconvertors from clinics 9 (n=3), 14 (n=2), and 6(n=1) comprised 3 mixed-clinic clusters. 4/6 significant clusters involved temporally associated HIV-1 strains from IDUs who seroconverted during the incidence spike of late 1996.2) 2 phylogenetically significant clusters (bootstrap values >80%) were noted among subtype B strains. One cluster of 2 strains appears to be both demographically (clinic 14) and temporally (EDS late 1996) linked. 3) The deduced amino acid sequences from each of the major phylogenetic clusters showed distinct amino acid patterns that were unique identifiers for each group. Conclusions: Phylogenetic analysis using several inference methodologies are suggestive of distinct HIV-1 transmission networks both among IDUs attending the same clinic, or between different clinics in Bangkok. We also note that the clusters of transmission mainly involve subtype E strains from clinics 9, 6, and 14 and are estimated to have occurred during late 1996. A better understanding of such networks and associated behavioral factors can help improve and focus intervention strategies to reduce HIV-1 transmission.

Back to the Top


Adjusting for Assay Detection Limit in Model-Based Estimation of HIV RNA Decay During HAART

Florin Vaida and Tony Fitzgerald


Back to the Top


HIV-1 Populations in Blood and Semen

Yang Wang1, Susan Frey1, Xi He1, Hong Zhao1, Gerald H. Learn1, David Nickle1, Allen G. Rodrigo2, Phalguni Gupta3, Bruce D. Walker4 and James I. Mullins1.

1: Departments of Microbiology and Medicine, University of Washington, Seattle, WA; 2: School of Biological Sciences, University of Auckland, Auckland, New Zealand; 3: Department of Infectious Diseases and Microbiology, University of Pittsburgh, Pittsburgh, PA; 4: Massachusetts General Hospital, Boston, MA.

Different representations of groups of viral sequences between blood and semen has been found in some patients (Delwart et al.,1997), including drug resistant vs. sensitive populations (Kroodsma et al.,1994). In order to understand the mechanisms underlying this nonuniform distribution, particularly to determine if virus in the two compartments evolve separately, viral populations from blood and semen were compared using longitudinal samples from five patients.

Four out of five patients showed consistently lower nucleotide diversity in semen, whereas one patient showed similar levels of diversity within the two compartments. Using the Sladkin-Maddison method for assessing compartmentalized structure, (Poss et al.,1998), four patients showed evidence of compartmentalization at one or more time points. Three out of four patients showed very similar rates of viral evolution in the two compartments, with the other having a lower evolutionary rate in semen. No signature amino acid residues were found in semen vs. blood viral populations. X4 viruses were only found in one patient, and only in the blood plasma.

Our results suggest that there maybe limited transfer between the two compartments during infection. This results in somewhat parallel evolutionary tracks, however, the biological significance of these two tracks remain obscure. We found no evidence that specific virus populations targeted the blood versus the semen, however, this conclusion is tempered by the fact that we examined only about 7% of the viral genome.

Back to the Top


Molecular Epidemiology and Phylogenetic Analysis of HIV-1 pol Genotypes from STD Clinics in Mumbai

Womack, C.A., Joshi, S., Deshpande, A., Sahni, S., Maniar, J.K., Saple, D.G., Bond, V.C., and Essex, M.



Back to the Top


Reevaluation Of Amino Acid Variability Of The HIV-1 Gp120 And Prediction Of New Discontinuous Epitopes

Yumi Yamaguchi-Kabata1 and Takashi Gojobori2

1: Kyoto University and 2: National Institute of Genetics

To elucidate the evolutionary mechanisms of the HIV-1 gp120 envelope glycoprotein at the single site level, the degree of amino acid variations and the numbers of synonymous and nonsynonymous substitutions were examined from 186 nucleotide sequences for the gp120 (subtype B). Analyses of amino acid variabilities showed that the level of variability was very different from site to site in both conserved (C1-C5) and variable (V1-V5) regions previously assigned. To examine the relative importance of positive and negative selection for each amino acid position, the numbers of synonymous and nonsynonymous substitutions that occurred at each codon position were estimated by taking phylogenetic relationships into account. Among the 414 codon positions examined, we identified 33 positions where nonsynonymous substitutions were significantly predominant. These positions where positive selection may be operating, which we call 'putative Positive Selection sites (PS sites)', were found not only in the variable loops but also in the conserved regions (C1-C4). In particular, we found seven PS sites at the surface positions of the a-helix (positions 335-347 in the C3 region) in the opposite face for the CD4 binding. Furthermore, two PS sites in the C2 region and four PS sites in the C4 region were detected in the same face of the protein. These PS sites found in the C2, C3 and C4 regions were apart in the amino acid sequence but close together in the three-dimensional structure. This observation suggests the existence of discontinuous epitopes in the protein surface including this a-helix, although the antigenicity of this area has not been reported yet.

Back to the Top

Back to the Program

Back to the Seattle HIV Dynamics & Evolution Meeting Home Page

jl, 2may00