Abstract Book Vol. 2 [International Conference on AIDS (14th: 2002: Barcelona, Spain)]

XIV International AIDS Conference

Abstracts WePeA5713-WePeA5716 7

GHOST-CD4 and U87-CD4 cells. Replication capacity was determined in primary coculture. Positively charged amino acid substitutions were evaluated in
V3, and viral phenotype was based on total net charge (NSI: +2 to +4 or SI:
>+4).
Results: The genetic forms(pol/V3) were: A(2); C(3); F(1); G(4); H/A(1); BG/B(3);
BG/G(1); DF/F(1); FBF/F(1); G/B(10); GH/K(1); AGA/A(1); GU/A(5); KU/A(1);
AJU/G(3); dual infection G+B(1)(U=unknown fragment). The phenotype based on
V3 resulted NSI/CCR5 in 29 of 36(80.5%) and SI/CXCR4 in the other 7(19.5%),
corresponding in the 7 cases to G/B recombinants. A correlation between phenotype based on V3 and biological phenotype in primary isolates was observed.
G/B recombinants resulted SI/CXCR4,CCR5,CCR3 and with several mutations in
positions 13, 14, 20, 25 and 32 as characteristics of these G/B recombinants. G
clone from dual G+B HIV infection resulted NSI/CCR5 and B clone SI/CXCR4.
Conclusions: The most frequent viral phenotype detected in these recombinant
strains circulating in Galicia was NSI/CCR5. However, in the majority of the G/B
recombinants the phenotype was SI/CXCR4/MULTITROPIC, underlining the importance of performing a follow-up of these cases, in order to establish whether
these findings are associated with a faster progression of the disease.
Presenting author: Lucia Perez-Alvarez, Area de Patogenia Viral. CNBF. Instituto de Salud Carlos III.B, 28220 Majadahonda. Madrid, Spain, E-mail: Iperezal
@ isciii.es
WePeA5713 Relationship between Tonsillar Tissue Viral
Burden (TTVL) and Plasma Viral Load (PVL),
peripheral blood T lymphocyte subsets and
Lymphoproliferative Responses (LPR)
F. Garcia, M. Arnedo, M. Plana, C. Gil, M. Caballero, L. Alos, E. Martinez,
J.M. Miro, J.L. Blanch, A. Cruceta, G. Mestre, T. Pumarola, T. Gallart,
J.M. Gatell. Hospital Clinic, Infectious Diseases Unit, Villarroel, 170, Spain
Background: To analyze the association of the tonsillar tissue viral burden with
plasma viral load, peripheral blood T lymphocyte subsets and LPR before any
ART and after 1 year of ART.
Methods: Measurements of plasma, tonsillar tissue viral load (TTVL), LS, and
LPR to mitogens and specific antigens were performed at day 0 before any ART
and 1 year thereafter.
Results: We performed tosillar biopsies in 81 patients with a CD4+ >500/mm3
and were repeated in 70 patients after 1 year of different ART The median weight
of tonsillar biopsy was 34 mg (IQR: 20-46). The weight was not correlated with
TTVL. At baseline the mean ~ SE of log of TTVL was 5.4 ~ 1.3 copies/mg of tissue. TTVL was undetectable in 38/40 patients (95%) who received a PI containing
regimen (either with 2 NRTI or 1 NRTI+ 1NNRTI) vs in 10/18 patients (44%) who
received a nevirapine containing regimen (plus 2 NRTI) and vs in 2/12 patients
(17%) who received only two NRTI (p= 0.004). At baseline, the mean ~ SE of log
of PVL was 3.9 ~ 0.9 copies/mi. PVL was undetectable in 67 out of 70 patients
(95%) after 1 year of ART At baseline, there was a significative positive correlation between log10 of TTVL and log10 of PVL, and CD8+CD38+ T cells. There
was a significative negative correlation between log1 0 of TTVL and CD4+ T cells,
ratio of CD4+/CD8+ T cells, ratio of naive/memory CD8+ T cells and CD8+
CD28+ T cells. After 1 year of ART, we divided the cohort in two groups: patients
with undetectable TTVL and with detectable TTVL. We did not detect significant
differences between these two groups in PVL, and LPR SI after 1 year of ART
Conclusions: In antiretroviral naive HIV-1 chronic infected patients seems to be
a correlation of the TTVL with PVL and peripheral blood T lymphocyte subsets.
However, after 1 year of antiretroviral therapy we could not detect useful peripheral blood surrogate markers that could indicate which patients had undetectable
tonsillar tissue viral load.
Presenting author: Felipe Garcia, Infectious Diseases Unit, Villarroel, 170,
Spain, Tel.: [phone], Fax: [phone], E-mail: [email protected].
es
WePeA5714 Cell-derived membrane protein profile and
quasispecies evolution in HIV from plasma and
spinal fluid of AIDS patients
I. Abbate1, G. Cappiello2, R. Longo1, A. Ursitti1, S. Calcaterra2, A. Spano1,
M.R. Capobianchi2. S.Pertini Hospital, Rome, Italy; 2Nationa/ Institute for
Infectious Diseases L. Spallanzani, Rome, Italy
Background: Cell-derived membrane proteins (CMP) acquired by HIV provide
information of its cellular source. CMP profile of HIV present in paired plasma
and spinal fluid samples from AIDS patients, as well quasispecies evolution in
env and gag regions.
Methods: CMP profile was established by immunocapture of purified virions from
plasma and spinal fluid with MAbs to cell membrane markers including: CD26,
CD45RO, CD36, CD58, N-CAM, VCAM-1, ELAM-1, CD44 and glutamate-R. Viral
quasispecies was analyzed by cloning and sequencing of V3 and gag PCR prod
ucts. Phylogenetic trees were constructed by the neighbour joining programme,
based on distance matrix values.
Results: Seven patients were included in the study. Direct comparison between
plasma and spinal fluid was possible on 5 paired samples. Incorporation of CMP
in the envelope of HIV from spinal fluid was lower than plasma virions. Although
no distinctive CMP pattern was observed in virions from the two body sites,
CD44 was the most represented CMP in HIV from spinal fluid, while the mono
cyte marker CD36 was the most prominent CMP on plasma virions. In some
paired plasma/spinal fluid samples, profoundly different CMP patterns could be
observed. V3 loop sequences in HIV from both compartments were compatible
with a predominant CCR-5 co-receptor usage. Heterogeneity in both V3 loop and
gag was generally lower in spinal fluid. Phylogenetic analysis indicated a general
trend to separate quasispecies evolution in the two body compartments.
Conclusions: Our data indicate that in AIDS patients HIV present in spinal fluid
and plasma may evolve separately. Accordingly, CMP profile of virions derived
from the two body compartments may be quite different, suggesting distinct cellular source of the HIV, although a similar pattern of co-receptor usage is suggested
by the V3 loop aminooacid pattern.
Presenting author: Maria R. Capobianchi, Laboratory of Virology, National
Institute for Infectious Diseases, Via Portuense 292, 00149 Rome, Italy, Tel.:
[phone], Fax: [phone], E-mail: [email protected]
WePeA57155 Shedding of human herpesvirus-8 in oral and
genital secretions from HIV-1 positive and HIV-1
negative women
M.M. Taylor1, B. Chohan1, L. Lavreys', M.L. Huang2, L. Corey1,
B. Richardson', R. Ashley3, K. Mandaliya4, J. Bwayo5, J. Kreiss1. 1 University of
Washington, Seattle, WA, United States; 2Fred Hutchinson Cancer Research
Center, Seattle, United States; 3Childrens Hospital and Medical Center, Seattle,
United States; 4University of Nairobi, Mombasa, Kenya; 5University of Nairobi,
Nairobi, Kenya
Background: Human herpesvirus-8, the causative agent of Kaposi's sarcoma,
was evaluated to determine risk factors and correlates of shedding among commercial sex workers in Mombasa, Kenya.
Methods: Women found to be seropositive for HHV8 by ELISA (OD value >0.2)
were examined for the shedding of virus from mouth, saliva, cervical, vaginal,
and PBMC samples using PCR. None of these patients had evidence of Kaposi's
sarcoma.
Results: Shedding of HHV8 in the oral cavity was detected in 42% (24/57) of HIV1 seropositive women and 18% (12/61) of HIV-1 seronegative women (p= 0.01).
Detectable virus was found in 11% (6/57) of cervical or vaginal secretions from
HIV-1 seropositive women as compared with 1.6% (1/61) from HIV-1 seronegative women (p = 0.055). The presence of virus in PBMCs was detected in 16%
of HIV-1 seropositive women and 16% of HIV-1 seronegative women (p = 1.0).
Women with CD4 counts < 200 cells/ were not more likely to have detectable
virus by PCR from any of the evaluated shedding sites (p = 0.8). The sole risk
factor for viral shedding from any site in HIV-1 seronegative women was age (p
= 0.05). Risk factors for viral shedding in HIV-1 seropositive women from the oral
cavity included age (p = 0.02) and years of prostitution (p = 0.03). Use of hormonal contraceptives was not associated with an increased risk of viral shedding
from any site in either group.
Conclusion: Genital and oral shedding of HHV-8 occurs in women with and without HIV-1 infection. Women with HIV-1 were more likely to shed virus from oral
and genital sites and this was not related to degree of immunosuppression as
measured by CD4 count.
Presenting author: melanie taylor, iartp, box 359909, 325 9th Ave, seattle, wa
98105, United States, Tel.: [phone], Fax: [phone], E-mail: [email protected]
WePeA57166 HIV-1 in genital tract and plasma of women:
compartmentalization of both viral sequences and
coreceptor usage
K. Sherefa-Kemal', B. Foley2, H. Burger1, K. Anastos3, H. Minkoff4, C. Ramirez
Kitchen5, W. Gao3, E. Robison3, S. Holman4, C. Dehner1, W.A. Meyer 1116,
A. Landay7, A. Kovacs8, J. Bremer7, K. Davenny9, B. Weiser'. IWadsworth
Center, New York State Dept. of Health, David Axe/rod Institute, Albany, NY
United States; 2Los Alamos National Laboratory, Los Alamos, NM, United
States; 3Montefiore Medical Center-NYC WIHS, Bronx, NY United States;
4State University of New York at Brooklyn, Brooklyn, NY United States;
5University of California at Los Angeles, Los Angeles, CA, United States; 6Quest
Diagnostics, Inc, Baltimore, MD, United States; 7Rush-Presbyterian-St. Lukes
Medical Center, Chicago, IL, United States; 8University of Southern California,
Los Angeles, CA, United States; 9Nationa/ Institute on Drug Abuse, Rockville,
MD, United States
Background: Worldwide, a large fraction of HIV-1 infections have been transmitted through exposure to virus in the female genital tract, yet the vast majority of
virologic analyses have focussed on HIV-1 in blood of men. To understand HIV-1
pathogenesis and halt its spread, it is critical to study virus in the genital tract of
women.
Methods: We compared the HIV-1 RNA sequences of the gpl20 env gene obtained contemporaneously from the genital tract and plasma of 13 women in the

Women's Interagency HIV Study. They displayed a variety of HIV-1 risks, patterns
of drug use, clinical states, and viral loads in genital tract (cervicovaginal lavage,
CVL) and plasma. We sequenced 8-11 clones from each specimen, a total of 265
clones.
Results: Computational analyses of HIV-1 genomes, which focussed in particular on variation in CVL vs. plasma, showed striking compartmentalization of sequences in 10 of 13 women. In these 10 women, quasispecies from CVL and