Bridging the Gap: Conference Record [Abstract book, International Conference on AIDS (12th: 1998: Geneva, Switzerland)]

526 Abstracts 31156-31159

12th World AIDS Conference

Design: Data were used from 36 individuals prevalent with HIV infection
followed prospectively with median follow-up of 3.0 years.
Methods: HIV-1 CTL activity was measured in a 4 hr chrom release assay using
autologous target cells expressing HIV-1 gene products. CD4+ lymphocyte count
was measured at enrolment and HIV RNA in plasma was measured retrospectively. The specific CTL activity was calculated by subtracting total cytotoxicity
against HIV-1 gene products (gp-120, gag, pol, nef) from cytotoxicity generated
against control. Risk of reaching death was analysed in Cox proportional hazards
model.
Results: The survival analysis showed that HIV specific CTL activity was of
no prognostic importance in the prediction of death. However, total CTL activity
showed a significant association with the risk of death in the univariete analysis
(RH = 1.96, P = 0.02), individuals with lower CTL activity were at higher risk
of death. The association between killing target of gp-120 and survival was still
significant after adjusting for CD4 lymphocyte count (data not shown) but were
no longer significant after adjusting for HIV RNA baseline levels.
Conclusion: The present study shows that, high levels of HIV-1 specific CTL
do not protect against disease progression. However, our data also indicate that
a retain in general cytotoxic capacity may protect against disease progression.
S31156 1 Bulk measurements of HIV-1-specific cytolytic
T lymphocyte activity using frozen PBMC and correlation
with disease progression
Ndongala Michel Lubaki, M.S. Shepherd, M. Kashamuka, T. Quinn,
R.S. Brookmeyer, R.C. Bollinger. 720 Rutland Avenue Ross 1150, Johns
Hopkins University, Baltimore, MD, USA
Background: The objectives of this study were to determine the biological variability of a bulk culture antigen-specific stimulation method for measurement of
HIV-1-specific cytolytic T lymphocyte (CTL) activity using frozen PBMC and to
correlate CTL activity with HIV-1 plasma viral load and CD4 count.
Methods: Blood samples were collected and stored at enrollment and at 2
weeks from 5 HIV-infected individuals from each of three baseline CD4 counts
(<200, 200-500, >500). Frozen lymphocytes were assayed in a blinded fashion
for HIV-1-specific CTL activity using an antigen-specific stimulation method in the
absence (BCM) or presence (TCM) of rlL2 and heterologous feeder cells. CTL
activity was correlated with CD4 count and viral load using 2 separate statistical
methods: 1) Spearman's Correlation of area under the net HIV-specific lysis curve
(AUC) and by 2) Log-linear regression (LLR) on net HIV-specific lysis.
Results: Lack of PBMC cell growth following in vitro stimulation correlated with
a lower CD4 count and higher viral load. For both BCM and TCM assays, the
AUC estimate of CTL activity correlated best with CD4 count and viral load. For
BCM assays, a higher CD4 count correlated with a higher CTL activity against
nef (R =.70; p <.001), pol (R =.69; p <.001), gag (R =.76; p <.001) and env
(R =.46; p =.01). For the TCM assays, CD4 count correlated with CTL against
nef (R =.63; p <.001), pol (R =.65; p <.001), gag (R =.71; p <.001), and env
(R =.45; p =.01). For the BCM assays, a lower viral load correlated with a higher
CTL activity against nef (R = -.50; p =.01), pol (R = -.49; p =.01) and gag (R =
-.49; p =.01). For the TCM assays, viral load correlated with CTL against nef (R
= -.85; p <.001), pol (R = -.78; p <.001), gag (R = -.84; p <.001), and env
(R = -.45; p =.01). CTL activity also had a statistically significant correlation with
viral load and CD4 count using the LLR analysis.
Conclusions: CTL assays are labor intensive and studies of their reliability
are limited. In this blinded analysis, specific CTL activity directed against multiple
HIV-1 antigens, measured by bulk culture antigen-specific stimulation of frozen
PBMC, demonstrated statistically significant correlation with viral load and CD4
count. The best method for correlating HIV-specific CTL with clinical stage
utilized an area under the specific lysis curve analysis. This method may also be
useful for measuring CTL activity in clinical trials.
31157 Strong cytotoxic T cell responses detected in the majority
of uninfected infants born to HIV-positive mothers
Rosemary Ffrench1, S.A. Maplestone1, M. Goode1, C. Hughes1, G.J. Stewart2,
E.M. Benson2, J.B. Ziegler1. 1'Paediatric Research Laboratories, Sydney
Children's Hospital, High St. Randwick, NSW, 2031; 2 Westmead Hospital,
Westmead, NSW, Australia
Objectives: To examine the cytotoxic T cell (CTL) responses in uninfected infants
born to HIV positive mothers, and compare this to CTL responses in HIV positive
and unexposed control infants
Design: Three year prospective and retrospective analysis of CTL responses,
at approximately three monthly intervals
Methods: Micro CTL assay on thawed PBMC assayed without in vitro expansion, and assessed using autologous B cell lines infected with recombinant
Vaccinia viruses (env, gag, pol, nef, tat, lac) in a chromium release assay
Results: We have been able to detect HIV-specific CTL responses in approximately 90% of infants who have been exposed to HIV but remain uninfected.
These responses can be detected as early as one day after birth and in some

infants are still present at 18-20 months of age. Interestingly, the majority of the
CTL responses are directed to the HIV env IIIB gene product, while only a third
of these infants have responses to other gene products (gag, pol, nef, gp120).
This is substantially lower than the proportion of HIV positive infants that can
respond to one or more of these HIV genes (-70%). When these responses were
compared to a group of age-matched unexposed infants we found that 30% of
the unexposed group also had responses to env IIIB but not to other HIV gene

products. Thus it appears that the env IIIB cytotoxic activity we are detecting may
be cross reactive with another, as yet unidentifed, antigen found in individuals
never exposed to HIV.
Conclusions: In most HIV exposed, uninfected infants, we have detected
cytotoxic activity to an epitope cross-reactive between the env IIIB and an
unidentified antigen. As this activity is higher in exposed, uninfected infants
than controls (p = 0.01) it seems that this response is primed or boosted by
HIV. More importantly, since there are more env 111B responders in the HIV
exposed/uninfected group than in the HIV +ve infants it is possible that this
cross-reactive response confers some protection against vertical transmission.
31158 ]HIV-specific cytotoxic T lymphocytes in asymptomatic
HIV-infected Thai adults with CD+ T lymphocyte >500
cells/pL
Supranee Buranapraditrun1, S. Ubolyam2, N. Sae-Sol3, S. Sirivichayakul1,
S. Nookhai4, P. Phanuphak1. 1Dept. of Medicine, Chulalongkorn Uni,. Thailand
Rama IV Road Pratumwan; 2Dept. of Microbiology, Chulalongkorn Hos.,
Bangkok; 3Dept. of Nuclear Medicine, Chulalongkorn Hos., Bamgkok;
4Anonymous Clinic, Thairedcross Society, Bangkok, Thailand
Background: Cytotoxic T lymphocyte (CTL) immune response is a major host
defence against viral infection, including HIV. There is evidence to suggest
that HIV-specific CTLs plays important roles in down-regulation of HIV-replication during acute seroconversion, protection from infection in babies born from
HIV-infected mothers and prolonged disease progression in long-term survivors.
Several studies had reported the CTL responses, CTL epitope mapping, and
MHC-restriction in Caucasian patients infected with HIV-1 subtype B. Nevertheless, little has been investigated in Thai patients who predominantly carrying
subtype E. There is also a need to establish a CTL assay in Thailand for future
involvement in studies of CTL-inducing HIV candidate vaccines.
Objective: To investigate HIV-specific CTL responses in asymptomatic HIV-1
infected Thai adults with CD4+ T lymphocyte >500 cells/p L
Methods: PBMCs from 12 eligible patients were used for the preparation of
both autologous EBV-transformed B cell lines (as target cells) and CTL-effector
cells. The CTL assays were performed according to Walker et al. Fresh PBMCs
and 3 days PHA-stimulated PBMCs were used for the assays as effector cells.
Wild-type vaccinia, HIV-1 (IIIB) gag, pol and env vaccinia-constructed were used
to infect the target cells. The effectors: targets (E:T) rotios were 50:1 and 100:1.
The results were expressed as % specific cytotoxicity and the assay using results
which showed 10% greater than that of wild type vaccinia was considered as
positive of HIV-specific CTL activity.
Results: Four out of the 12 patients showed detection of HIV-specific CTL response. All of them carried only HIV-gag-specific CTL with % specific cytotoxicity
of 37.92%, 28.49%, 20.17% and 21.66% respectively.
Conclusions: Our preliminary study found that four out of twelve (33.33%)
HIV-infected asymptomatic patients with CD4+ > 500 cells/piL showed HIV gagspecific CTL activity. No env, pol-CTL responses were observed. HIV-subtyping
and HLA class-I typing will be performed to elaborate HIV-1 subtype and HLA restriction in each individuals. CTL-cloning, CTL-epitope mapping and cross-clade
CTL-activity will be further investigated.
31159 ]Role of 3-chemokines and chemokine receptor on HIV
rgpl20-induced B lymphocyte function
Christina Patke1, C. Green2, W. Shearer2. Texas Children's Hospital Dept
Immuno 6621 Fannin Houston TX 77030 USA; 2Baylor College of Medicine
Houston TX, USA
Objectives: To examine the interaction of ý/-chemokines (MIP-1P, RANTES) and
their chemokine receptor (CCR5) on HIVgp 120-induced B cell functions of cAMP
generation, proliferation, IgM production and the cell surface expression CD23
and CD79b.
Methods: Highly purified normal human B cells from seronegative donors
(>97% CD20+) were isolated by adherence to CD19-coated beads. IL-4 plus
antiCD40-activated B cells were treated with recombinant gp120 (10 ng/ml) prior
to exposure to If chemokines or receptor. Cyclic nucleotide generation was assessed at six hr, DNA synthesis at day three and IgM production on day seven.
Cell surface marker expression was determined by flow cytometric analysis using
the Coulter EPICS-XL.
Results: Previously, we have shown enhanced IgM production, proliferation,
and cAMP generation by gp 120 on B cells. MIP1fl or RANTES (1-100 mg/ml)
further enhanced the gpl20-induced B cell IgM secretion, proliferation and cAMP
generation, (65%, 95% and 73%, respectively) (p < 0.01). Inhibitory effects on
gpl20-induced cell proliferation and cAMP generation occurred in the presence of
CCR5 1-100 ng/ml), (50% and 62% respectively) (p < 0.01) along with enhancement of IgM secretion (147%). Flow cytometric analysis revealed that CCR5,
MIP-1lp or RANTES down-regulated the cell surface marker expression of the B
cell anti associated with the Ig superfamily-BCR complex, CD79b (44%, 62% and
42%, respectively) whereas only CCR5 altered the expression of CD23 (51%).

Conclusions: The evidence presented here indicates a role for MIP-1/ and
RANTES on the early B cell events of proliferation, cyclic nucleotide generation
and cell surface marker receptor modulation in opposition to the /-chemokine
receptor, CCR5. The ability to regulate early B cell events could be a targeted
area in the development of novel designer molecule therapeutic approaches to
AIDS.