Prevalence and Characterization of HIV-Specific Adaptive Immune Responses in Genital Mucosal Secretions from a Cohort of High Heterosexual Risk, Seronegative Women in Chicago
Richard Novak*1, L Baum2, P Graham1, H Chen1, W Yang1, and B Parekh3
1Univ of Illinois at Chicago, US; 2Rush Univ, Chicago, IL, US; and 3CDC, Atlanta, GA, US
Several studies have detected mucosal immune responses in
African and Asian cohorts of HIV seronegative sex workers, implying that these
responses may be protective. We studied a similar cohort in Chicago to
determine the prevalence of HIV-specific mucosal responses in a Clade B region.
We collected behavioral risk data and cervicovaginal lavages
(CVL) from a cohort of 64 high risk seronegative (HSN) women, and a low risk
(LR) cohort of 33 women at 6 month intervals. Samples were subjected to the
following assays: binding IgG and IgA to gp120 (ELISA, HIV-1MN);
ADCC assay (51Cr release, HIV-1 MN gp120-coated target
cells), neutralizing assay to 3 pseudoviruses with B clade gp160 envelopes
(env, TZM-bl assay).
Analysis of relationships between the presence of binding
IgG, IgA to gp120, ADCC activity, neutralizing activity (NA) to the 3 clade B
gp160 clones as well as risk behaviors, coexisting vaginoses were done after
adjusting for repeated sampling, (GEE with working correlation being independent),
excluding samples containing semen.
17% of the HSN cohort had IgG, 4% had IgA binding antibodies
(Abs) to gp120 (HIV-1MN) vs 0% of the LR cohort. NA to env clones RHPA4259.7,
AC10.0.29 and QH0692.42 was present in 11, 49 and 32% of the HSN cohort
respectively, vs 0, 12 and 0% of the LR cohort; 5.5% of baseline samples had NA
to all 3 envs.
In the HSN group, risky sex (>10 partners in the past 6
months, and seldom/never using condoms) correlated with NA of all 3 envs
(RHPA4259.7, p=0.04; AC10.0.29 P=0.03; QH0692.42 P=0.01). The presence of gp120
IgG Abs correlated with NA of RHPA4259.7 (P=0.04), but not the others.
There was a significant correlation between ADCC and NA of
AC10.0.29 (P=0.03), but not with the other 2 envs, nor with IgG or IgA gp120
Abs. The presence of ADCC, gp120 IgG and IgA antibodies and NA of RHPA4259.7,
correlated with the presence of CVL leukocytes but not AC10.0.29 or QH0692.42.
There was a negative association between the presence of BV and gp120 binding
IgG titer (P<0.0001).
HIV-specific humoral immune responses can be found in the
genital tracts of HSN women in the US. Some samples neutralize all env clones
tested thus far. The presence of BV may interfere with the naturally acquired
adaptive host immune response to HIV.
These immune responses may play a role in natural protection
from HIV infection, and thus further characterization of these Abs may provide
guidance in vaccine development.