CCR5 Ligands in HIV Vaccines

Therefore, recall antigen stimulation can be easily measured by chemokine release, so
that it is likely that this parameter can be readily measured in HIV immunization
protocols. We then evaluated the kinetic of release of CCR5 ligands, we found that MIP1 a and MIP-1 are released very early upon antigen activation; a first set of experiments
(figure 2a) revealed that these chemokines were released as early as after one day. When
we evaluated the production of CCR5 ligand within one day of stimulation, we found
that MIP-1c and -13 production can be detected as early as 2 hours after stimulation,
becomes very prominent at 4-6 hours, and by 16-20 hours is at its peak (Fig 2b and 2c).
This suggests that chemokine release is a very early event in the triggering of the
immune response. Such a quick response may be very important in effectively
preventing infection of cells when immune system cells encounter the antigen. Finally,
we wanted to establish whether the responses that we observed diminish in magnitude
with time from vaccination. We analyzed chemokine production and proliferation in four
subjects who did not receive a tetanus vaccination booster for more than ten years. All of
these subjects responded to activation by producing chemokines (Fig. 3). However, we
observed that we could not detect cell proliferation by thymidine uptake, at any of the
timepoints that we tested. Therefore, it appears that chemokine responses to antigen
activation constitute a very long-lasting response (one of the subject tested declared that
his last tetanus vaccination occurred in 1972).
It is important to stress that not all chemokines are induced by antigen stimulation: We
tested these same supernatants by ELISA for the presence of two other CC chemokines
which have been shown to inhibit HIV, MDC and 1309. Our results indicate that MDC
production is much delayed as compared to the production of CCR5 ligands, as MDC is
barely detectable at day 3, while its concentration rises at day 6 and, more robustly, at
day 9 (Fig. 4). In contrast, we could not observe increase of 1309 production by any
stimulus, including PHA (Fig. 4).
In summary, our preliminary data shows that we can organize the study of a large cohort
of human subject (29), where we can measure chemokine release in response to antigen,
and analyze in relationship to clinical parameters. In addition, we have determined that
chemokines are relevant to vaccination, as they are released upon stimulation with recall
antigen. In our experience, chemokine production from cells is a reproducible parameter,
so that it can be reliably used in cohort study to evaluate the impact of chemokines in the
immune response to antigen stimulation. Finally, we have shown that we can develop
assays for chemokine isoforms that are potentially relevant to HIV infection.
In conclusion, we think that these data fully support the hypothesis that chemokines
release constitute a reliable, reproducible, and informative marker of immune response
that can be measured in vaccine trials and that is equally or more sensitive than
proliferative assays. Previous reports have shown that all of the subjects immunized with
canarypox-based HIV immunogen were responders when tested with a proliferation assay
similar to the one that we used (51). Since our preliminary data support the concept that
chemokine assay is equally or more sensitive than the proliferation assay, we fully expect
that cell samples from 30 vaccinated subjects and a similar number of placebo recipient
will provide a sample with adequate statistical power to test our hypotheses. For example,
assuming a low 50% rate in vaccinees for MIIP-loc and -3 and a 5% false positive
response rate in placebo recipients, we have greater than 90% power to detect these
differences with a sample of 30 vaccinees.