Final Program and Oral Abstracts [International Conference on AIDS (8th: 1992: Amsterdam, Netherlands)]

WEDNESDAY, 22 JULY 1992

WeC 1022-WeC 1027

WeC 1 022 IV SIGNATURE PATTERN ANALYSIS AND "FINGERPRTING".
Korber, B.T. and Myers, Gerald Theoretical Division,
Los Alamos National Laboratory, Los Alamos, N.M., 87505 U.S.A.
Objective: To go beyond simple nucleotide or amino acid distance relationships
(percent similarities) as the basis for tracking HIV relatedness, it is desirable
to objectively establish amino acid signature patterns, or "HIV fingerprints."
Methods: HIV amino acid sequences, i.e., sibling sequences, from a given infected
individual are compared to a large reference set of HIV sequences to determine the
positions in the test sequences at which atypical residues occur (amino acids not
found in 50% or more of the reference sequences). Those atypical non-contiguous
amino acids common to all clone sequences from the patient constitute the signature
or "fingerprint" for that individual's viruses. Sequences from epidemiologically
linked individuals, or even sequences from different tissues in the same individual,
are then analyzed in relation to this differentiating pattern. With reasonable
assumptions, probabilities can be assigned to the signature unique to a group of
sibling sequences.
Results: In a study of viral sequences from a dentist and five of his patients
whom he appears to have infected during the course of invasive oral procedures,
signature pattern analysis strongly supports the conclusion of transmission in a
health care setting. The dentist's viral fingerprint was found in all five patients'
viruses but not in viral sequences of 35 local controls. All patient signatures are
found in one or more of the dentist's viral sequences but not in the controls.
Conclusions. Molecular epidemiologic investigations of HIV transmission, especially
linked transmissions and questions of dual infection, will be significantly enhanced
through the employment of signature pattern analysis.
Gerald MYERS, Theoretical Division, Los Alamos National Lab,
Los Alamos, N.M. 87545 U.S.A. 505-665-0480, 505-665-3493
WeC 1024 ANTIGENIC DIVERSIFICATION OF THE
HIV-1 V3 NEUTRALIZATION DOMAIN DURING PROGRESSION OF THE
EPIDEMIC. Kuiken. Carla L*; Zwart, G.*; Baan, E.*; Coutinho, R.A.**;
Goudsmit, J.*.*HRL, UvA, Amsterdam, **Municipal Health Service, Amsterdam
Objective: To investigate the evolution of the gpl20 V3 region in the course of the HIV-1 epidemic in the
Dutch homosexual population.
Materials and methods: Direct sequences, 278 nucleotides long, were made of viral genomic RNA amplified
from early seroconverter sera of homosexual men. Ten sequences were generated from around 1980, and 10
each from 1985 and 1990. Reactivity of sera to synthetic V3-mimicking peptides was measured using ELISAs.
Results: The V3 loop (Cys-Cys) became more diverse (mean difference between sequences 3.9%, 5.7% and
8.6% respectively in 1980, 1985 and 1990). Outside V3, the diversity was already at 8% in 1980 and the
increase was much less marked. Comparing only silent changes over the whole region, the diversity increased
from 8.87% in 1980 to 15.75% in 1990, whereas non-silent changes remained stable. Despite of the increased
variability, no continuous change was found in the consensus sequences, indicating that there is no movement
of the swarm of viruses in a particular direction. Amino acid 308, of particular importance to antibody
reactivity, showed more variation in 1990 than in earlier years. Reactivity to 3 peptides designed to
discriminate at this position also showed this diversification, consistent with earlier findings of our group
[Zwart et al., in preparation].
Conclusions: The V3 region becomes increasingly diverse as the epidemic progresses. The data suggest that
the V3 loop will become a more elusive target region for vaccine-induced antibodies over time.
Carla L Kuiken, Human Retrovirus Laboratory, UvA, Meibergdreef 15, 1105 AZ
Amsterdam, the Netherlands, phone [phone], fax [phone]
WeC 1026 MOLECULAR EPIDEMIOLOGY OF HIV-1 IN
WeC 1 2 EDINBURGH. Holmesr Edward C, Zhang, L.O., Mackenzie, P., Harvey,
E., Baker, C., Simmonds, P. & Leigh Brown, A.J. Division of
Biological Sciences, Univ. of Edinburgh, Edinburgh EH9 3JN, Scotland.
Objectives: To establish, using reliable nucleotide sequence (phylogenetic) markers,
contact networks of the epidemic of HIV-1 in Edinburgh. Such data will provide
information on the number and origin of strains entering the Edinburgh population and
whether different social/behavioural groups are characterised by different strains. The
relationship between strain diversity and the number of transmission events can also
be investigated.
Methods: Approximately 350bp of the p17 region from the gag gene of HIV-1 were obtained
from a large number of Edinburgh HIV patients, including haemophiliacs, intravenous drug
users and heterosexuals. Phylogenetic analysis was then used to infer the relationships
of the different patients and to search for specific sequence patterns.
Results: Analyses showed that HIV entered the Edinburgh population from a number of
different sources. The haemophiliac cohort, originally thought to have been infected
by a single batch of factor VIII, are now seen to have at least 3 different viral types.
Surprisingly, given the large number of possible transmission events involved, the
intravenous drug user epidemic seems to have been founded by a single and different
strain to those found in the haemophiliac population. All heterosexuals analysed are
seen to cluster with the drug users, thus suggesting that intravenous drug use has
created a bridge by which HIV-1 can enter the heterosexual population.
Conclusions: We have found that an analysis of gag HIV-1 sequences is an informative
way to trace epidemiological contact networks and could be useful in cases where the
precise nature of the infection is uncertain. Analysis of sequence data obtained so far
suggests that the Edinburgh haemophiliac epidemic involves a variety of strains whereas
intravenous drug users and heterosexuals are characterised by a single ancestral strain.
Dr. Edward C. Holmes, Division of Biological Sciences, University
of Edinburgh, ICAPB Crew Building, West Mains Road, Edinburgh EH9
3JN, Scotland, UK. Tel.[phone] Fax. [phone]

WeC 1023 IDENTIFICATION OF ADDITIONAL GENETIC SUBGROUPS OF HIV-I AND EVIDENCE
FOR GEOGRAPHIC INTERMIXING McCutchan Francine, Louwagie J*; van der Groen G,
Piot P**; Myers G+; and Burke D++. *Henry MJackson Foundation, Rockville MD; **Prince
Leopold Institute of Tropical Medicine, Antwerp, Belgium; + Los Alamos National Laboratory,
Los Alamos, NM; ++Walter Reed Army Institute of Research, Washington DC.
Backgound: Genetic variation of HIV-1 and the geographic distribution of genetic variants are
important considerations for HIV-1 vaccine development. A global surveillance of HIV-1 isolates
is being conducted to supplement and extend the current genetic database for HIV-1.
Methods: More than two hundred geographically diverse HIV-1 isolates were surveyed for genetic
similarity to reference isolates using a Polymerase Chain Reaction (PCR) typing procedure.
Twenty-four genetically diverse isolates were selected and the complete gag gene was PCR
amplified, cloned, and sequenced.
Results: A phylogenetic tree of gag DNA sequence relationships among HIV-1 isolates including
twenty-four sequences generated from this study and fourteen previously published sequences has
been constructed. The geographic origins of the newly added isolates were Belgium (n=2), Brazil
(n=2), Gabon (n=8), Ivory Coast (n=l), Kenya (n=4), Philippines (n=l), Somalia (n=l),
Thailand (n=2), Zaire (n=l), and Zambia (n=2). Six genetic subgroups with at least two members
and a seventh subgoup containing a single isolate were found. Most of the subgroups contained
isolates from more than one continent, and some geographic locales contained members of several
subgroups.
Conclusions: Based on genetic relationships in the gag gene, the genetic diversity of HIV-1
appears more extensive that previously appreciated. Several subgroups equidistant from previously
sequenced isolates have emerged. Most subgroups have apparently migrated to more than one
continent. Geographic dispersal may be playing a significant role in broadening the genetic
diversity of HIV-1 in specific locales.
McCutchan, Francine, Henry M. Jackson Foundation, 1500 E. Gude Dr.
Rockville, MD 20850 U.S.A. Phone [phone]; FAX [phone].
WeC 1025 IDENTIFICATION OF TWO DISTINCT HIV-1 SUBTYPES IN TAILAND WITH
APPARENT SEGREGATION BY MODE OF TRANSMISSION OU Chin-Yih1
Auwanit, W4.; Pau, C.-P.1; Luo, C.-C.1; Kalish, M.1; Takebe, Y.2;
Bandea, C.1; Lee-Thomas, S.1; George, J. R.1; Schochetman, G.1;
Gayle, H1.; Young, N.1,3; and Weniger, B.1,3 1Division of HIV/AIDS,
Centers for Disease Control, US Public Health Service, Atlanta, Georgia, USA;
2National Institute of Health, Tokyo, Japan. 3The HIV/AIDS Collaboration, Bangkok, and
4National Institute of Health, Bangkok, Thailand.
objectives: To determine the molecular epidemiology of the HIV explosion in Thailand
since 1988 and to develop peptide-based serologic assays to facilitate HIV-1
subtyping.
Methods: Blood samples were collected in 1991 from infected persons with various risk
behaviors from 7 provinces. HIV proviral DNA was amplified through the polymerase
chain reaction and sequenced. Enzyme-linked immunoassays based on peptides derived
from the V3 loops of HIV-1 subtypes were developed for serosurveys.
Results: Only 2 genetically distinct BIV-1 subtypes (A and B) were identified among
52 infected persons. Subtype A was genetically distant from American HIV-1 consensus
sequences while subtype B was genetically more related. Average sequence difference
between subtypes A and B was 22%, too high to have diverged from a common HIV-1
precursor since 1988. Both genetic and serologic subtyping revealed that 91% (20/22)
of persons presumably infected through sexual transmission had HIV-1 subtype A, while
75% (21/28) of injecting drug users had subtype B (p<0.001). Subtype A also appeared
to be predominant in the northern region of Thailand.
Conclusion: Two HIV-1 subtypes independently entered Thailand in the late 1980s and
infected most of the estimated 300,000 HIV-positive Thais by 1991. Subtype A was
primarily sexually transmitted whereas subtype B was parenterally transmitted. These
findings will assist in understanding the epidemiologic patterns of HIV transmission,
and developing a vaccine in Thailand.
Dr. Chin-Yih Ou/Centers for Disease Control
1600 Clifton Road, MS D12
Atlanta, GA 30333 USA Tel [phone] FAX [phone]
WeC 1027 HIV CAUSES AIDS: A CONTROLLED STUDY.
Craib. Kevin JP; Schechter, Martin T; Le TN; O'Shaughnessy MV; Montaner JSG.
The Vancouver Lymphadenopathy-AIDS Study (VLAS) Group, UBC & BC Centre for Excellence
in HIV and Related Viral Diseases, Vancouver, CANADA.
Issue/Problem: A considerable number of people continue to question the role of HIV in the pathogenesis of AIDS. One
of the main proponents of this school of thought. Prof. P. Duesberg, wrote in 1988 (Science 1988; 242: 997-8) and repeated
in a public address in 1991 that the necessary comparisons in controlled cohorts were not available to address this question.
The purpose of this analysis was to conduct just such a controlled analysis within a cohort of homosexual men.
Descrintion of Project: The VLAS is a cohort study which has followed HIV- and HIV+ homosexual men since 1982.
This analysis involves 350 persistently seronegative men (SN), 237 seroprevalent men (SP), and 128 seroincident men (SI)
followed for a median of 103 months. A total of 134 cases of AIDS-defining illnesses have occurred in these men. We studied
the incidence of AIDS-illnesses by HIV status, by exposure to psychoactive
120 SN Group drugs, and by other risk behaviours. The latter are popular counternou "hypotheses to HIV. We also studied changes in CD4 count in HIV+ and
O100 HIV- men through time.
800 - o Results: The SN group continues to be exposed to a number of risk
o- "- -- -- factors. As many as 49% of the SN group reported using psychoactive
SI Grou, p I drugs during follow-up. About 25% of the SN group reported receptive
anal intercourse with casual partners. Of 134 AIDS illnesses in the cohort,
200 - every single one occurred in men with pre-existing evidence of HIV
infection (52 PCP, 33 KS, 24 other OI, 10 candidiasis, 10 lymphoma, 4
Swasting syndrome, 1 neurologic disease). No evidence of any immune
0 2 3 4 5S6rr7 8 1011 dysfunction has occurred in the SN group. The figure illustrates that CD4
counts have remained stable in the SN group while they have fallen steadily
in the SI group after seroconversion.
Lessons Learned: Claims that AIDS is caused by other exposures and not by HIV are clearly not borne out by these data.
Part of the problem may be semantics. No one has ever disputed that cofactors play a very important role in determining the
rate and type of clinical progression just as cofactors must determine why some smokers develop lung cancer, why some
develop bronchitis and why still others remain well. The search for cofactors must continue but it is a disservice to the many
persons infected with HIV and a hindrance to public health initiatives for responsible scientists to claim that HIV is harmless.
Mr. Kevin JP Craib, Vancouver Lymphadenopathy-AIDS Study, 205-1033 Davie Street;
Vancouver, BC, CANADA V6E 1M7. Telephone [phone]; Facsimile (604)-631-5210;

We49