Secondary Logo

Journal Logo

Basic Science

Detection of HIV Vaccine-Induced Cell-Mediated Immunity in HIV-Seronegative Clinical Trial Participants Using an Optimized and Validated Enzyme-Linked Immunospot Assay

Dubey, Sheri MS*; Clair, James PhD; Fu, Tong-Ming PhD*; Guan, Liming MS*; Long, Romnie MS*; Mogg, Robin MS; Anderson, Kiersten BS*; Collins, Kelly B MS*; Gaunt, Christine MS*; Fernandez, V Rose MS*; Zhu, Lan MS*; Kierstead, Lisa PhD*; Thaler, Scott MD; Gupta, Swati B DrPH§; Straus, Walter MD§; Mehrotra, Devan PhD; Tobery, Timothy W PhD*; Casimiro, Danilo R PhD*; Shiver, John W PhD*

JAIDS Journal of Acquired Immune Deficiency Syndromes: May 1, 2007 - Volume 45 - Issue 1 - p 20-27
doi: 10.1097/QAI.0b013e3180377b5b
  • Free

Abstract

Twenty-five years after the identification of HIV as the pathogen that causes AIDS,1,2 an effective vaccine remains elusive. Given the accumulation of evidence that cell-mediated immune responses play a substantial role in containment of primary HIV infection and suppression of viral replication,3-7 it is likely that a component of an effective vaccine is the ability to elicit HIV-specific T cells. Several candidate vaccines aimed at inducing this type of response are currently in early clinical testing, most using naked DNA and/or viral vectors to deliver HIV antigens.8,9 One of the most promising candidates is the replication-incompetent human adenovirus type 5 (Ad5) vector expressing internal HIV proteins such as gag, pol, and nef. To evaluate the immunogenicity of these vaccine modalities, highly robust, sensitive, and specific assays are needed to detect and quantify cell-mediated immunity against HIV antigens.

The interferon-γ (IFNγ) enzyme-linked immunospot (ELISPOT) assay has been widely used to measure antiviral immunity and vaccine immunogenicity, with secretion of IFNγ by CD4+ and CD8+ T cells serving as a marker of antigen-specific T-cell function.10-13 The assay is compatible with the use of cryopreserved peripheral blood mononuclear cells (PBMCs), which can be stimulated with synthetic peptides directly in the assay well without prior expansion of precursor cells in vitro.14-16 Thus, by using cryopreserved PBMCs and batched assays, the ELISPOT assay can provide a relatively high-throughput mechanism for evaluating cell-mediated vaccine immunogenicity in large-scale clinical trials.

To evaluate and compare ELISPOT responses effectively across vaccine recipients and vaccine regimens, rigorous ELISPOT assay optimization and validation are needed.17,18 To this end, there have been several published reports describing HIV ELISPOT assay validation and determination of a positive response.15,17,19,20 These studies typically included relatively small sample sizes, however, and data analyses relied primarily on inclusion of simulated data to extend the data set in statistical models. Additionally, conclusions were often drawn using responses from only HIV-infected donors tested with 15mer or 20mer peptide pools and did not include analysis of background responses in HIV-negative donors. In this report, we describe validation of our optimized HIV ELISPOT assay by assessing responses in a large number of HIV-negative volunteers with pools of 20mer, 15mer, and 9mer peptides using an empirical method to establish positivity criteria that limit false-positive rates to ≤1%. Using this optimized validated assay, we also present a comparison of ELISPOT responses to 20mer, 15mer, and 9mer pools in relatively large cohorts of HIV-infected donors and experimental HIV vaccine recipients. Our data demonstrate sensitivity of the assay to detect HIV-specific cell-mediated immune responses induced by natural infection and vaccination, and therefore confirm its utility to monitor vaccine immunogenicity in support of clinical development of our vaccine candidates.

METHODS

Peripheral Blood Mononuclear Cell Samples

Whole blood was collected in ethylenediaminetetraacetic acid (EDTA)-containing Vacutainer tubes (BD Diagnostics, Franklin Lakes, NJ), and PBMCs were isolated using standard Ficoll-gradient centrifugation or Accuspin System Histopaque-1077 (Sigma, St. Louis, MO). PBMC aliquots were stored frozen in liquid nitrogen until the time of assay. Specific procedures for optimal sample preparation, handling, and transport are described elsewhere.16 Samples were collected from healthy HIV-negative donors, from asymptomatic HIV-infected donors (stable HIV-1 infection with or without antiretroviral treatment, CD4+ T-cell counts >350 cells/μL, and viral loads <100,000 copies/mL), and from HIV-negative experimental vaccine recipients. Collectively, the vaccinated subjects received placebo or one of the following vaccine candidates: (1) DNA expressing clade B CAM-1 gag (DNA gag); (2) replication-defective Ad5 expressing the CAM-1 gag (Ad5 gag); (3) a prime/boost combination of the DNA gag with Ad5 gag; and (4) a trivalent vaccine consisting of 3 Ad5 viruses, with each expressing CAM-1 gag, IIIB pol, or JRFL nef (Ad5 gag + Ad5 pol + Ad5 nef).8 These studies were approved by the institutional review boards of all participating centers, and informed consent was obtained from all volunteers.

Peptides

Synthetic peptides representing the open reading frames of HIV-1 clade B gag (CAM-1), nef (JRFL), and pol (IIIB) proteins were prepared as 20mers overlapping by 10 amino acids (aa), 15mers overlapping by 11 aa, and 9mers overlapping by 8 aa. Lyophilized peptides were solubilized in dimethyl sulfoxide (DMSO) at 20 to 50 mg/mL, were pooled within each protein (1 gag pool, 1 nef pool, and 2 pol pools [pol-1, which represents the NH2-terminal half, and pol-2, which represents the COOH-terminal half]), and were stored in aliquots at −70°C.

Interferon-γ Enzyme-Linked Immunospot Assay

PBMC samples were tested in a modified ELISPOT procedure21 that has been described previously.22 Briefly, sterile 96-well polyvinylidene difluoride (PVDF) membrane-backed microtiter plates (Millipore, Billerica, MA) were coated overnight at 4°C with 100 μL per well of mouse antihuman IFNγ monoclonal antibody clone 1-D1K (MabTech, Stockholm, Sweden) at 10 μg/mL. Frozen-thawed cells were cultured overnight in “R10,” complete RPMI 1640 medium plus 10% fetal bovine serum, in a humidified carbon dioxide (CO2) incubator at 37°C before enumeration and assay input. Fresh PBMCs, which were tested only as part of the 20mer validation studies, were obtained from the same sample preparations as the corresponding frozen vials. Viable PBMCs were enumerated by hemacytometer with trypan blue exclusion or using the Guava ViaCount assay (Guava Technologies, Hayward, CA). When using manual hemacytometer counting versus the Guava, the viable cell counts and the resultant ELISPOT data were shown to be concordant in an extensive cell counting validation study before implementation of the Guava to replace manual counting (data not shown). After cell enumeration, cells were added to the blocked plates at 2 × 105 and 1 × 105 viable cells per well. Diluted peptide pools were added in duplicate per cell input at final peptide concentrations of approximately 2.5 μg/mL for 20mers and 15mers and 0.625 μg/mL for 9mers. Optimal peptide concentrations were previously determined in titration experiments (data not shown). Peptide-free DMSO diluent matching the DMSO concentration in the peptide solutions (<1% final) was used as a nonantigen control (“mock”) for each sample. Concanavalin A was used as a positive assay control. Assay plates were incubated overnight in a 37°C CO2 incubator. To detect captured IFNγ, plates were incubated 2 to 4 hours at room temperature with 100 μL per well of biotinylated antihuman IFNγ monoclonal antibody clone 7-B6-1 (MabTech) at 1 μg/mL, followed by a 2-hour room temperature incubation with 100 μL per well of alkaline phosphatase-conjugated antibiotin monoclonal antibody (Vector Laboratories, Burlingame, CA) at a dilution of 1:750. To develop spots, plates were incubated 5 to 10 minutes at room temperature with 100 μL per well of 1-Step NBT/BCIP substrate (Pierce, Rockford, IL). Spots were imaged and counted on an automated ELISPOT imaging system (Autoimmun Diagnostika, Strassberg, Germany). The number of spots per well at each cell input was normalized per 1 × 106 cells and averaged for each sample and antigen; these normalized values are hereafter referred to as ELISPOT responses.

RESULTS

Enzyme-Linked Immunospot Assay Validation and Positivity Criteria

Responses by HIV-negative donors to HIV peptides in the ELISPOT assay were assessed to establish a positivity cutoff that would limit identification of false-positive responders to ≤1%. Fresh and frozen-thawed PBMC samples from 20 seronegative donors were each tested 2 times for responses to 20mer peptide pools in the ELISPOT assay, as previously described for 20mer gag peptides.23 Responses by fresh and frozen-thawed PBMC samples (after overnight culture) were statistically similar in this study (data not shown; see similar analysis in the study by Kierstead et al16), and were thus analyzed together. In all samples, responses by the seronegative donors to the 20mer HIV peptides were relatively low (mean = 17 spots per 106 PBMCs, range: 0-115 spots per 106 PBMCs), and peptide responses were similar to corresponding mock responses, (mean = 14 spots per 106 PBMCs, range: 0-118 spots per 106 PBMCs). The highest observed value in response to 20mer peptides (115 spots) was from the same sample that gave the highest observed mock value (118 spots), illustrating why ELISPOT positivity criteria must consider the inherent background of each sample (response in the absence of in vitro antigen stimulation) in addition to the magnitude of the antigen-stimulated response. Because each sample tested in the ELISPOT assay has 2 measured responses (non-antigen-stimulated “M” and peptide antigen-stimulated “C”), we evaluated their joint distribution in a 2-dimensional region of positivity using the natural log (ln) of the measured peptide responses. Figure 1A plots the points (c,m) where m = ln(M) and c = ln(C) for the 20 seronegative samples tested with gag, nef, and pol 20mer peptide pools. The region of positivity is defined by antigen spot counts ≥C0 and an antigen/mock ratio ≥R0. Here, we defined C0 = 55 and R0 = 4. In the 20mer validation data set, no data points were in the positive region as defined; thus, the observed false-positive rate to each of the 20mer peptide pools was 0% (5.50% upper 90% confidence interval [CI]; Table 1). By comparison, samples from 6 HIV-infected donors had mock responses similar to the seronegative donors (mean = 27 spots per 106 PBMCs, range: 0-189 spots per 106 PBMCs) and substantially higher responses to 20mer HIV peptides pools (mean = 1090 spots per 106 PBMCs, range: 158-2700 spots per 106 PBMCs), all of which fell in the positive region as defined in Figure 1A (data not shown).

F1-4
FIGURE 1:
ELISPOT responses in HIV-seronegative donors to HIV gag, nef, pol-1, and pol-2 peptide pools are summarized for 20mer peptides (n = 40) (A), 15mer peptides (n = 290) (B), and 9mer peptides (n = 242) (C). Responses are shown in a 2-dimensional region of positivity model in which the ln of peptide responses, “C,” is plotted against the ln of “mock” (non-antigen-stimulated) responses, “M.” The region of positivity is defined by peptide spot counts ≥C0 and a peptide/mock ratio ≥R0, where we defined C0 = 55 (solid vertical line) and R0 = 4 (dashed diagonal line). For 9mer peptide responses, an alternative criterion, C0 = 470, is also shown (bold vertical line to right of first vertical line). Symbols in the “positive region” represent false-positive responses in these samples.
T1-4
TABLE 1:
Observed False-Positive Rate (Upper 90% Confidence Interval) When Positivity Criteria are ≥55 Spots per 106 Cells and ≥4-Fold Mock

In addition to validating the ELISPOT assay with 20mer peptides, we assessed responses induced by smaller peptides such as 15mers and 9mers. It has been previously reported that ELISPOT and intracellular cytokine staining (ICS) response magnitudes tend to be higher with peptides shorter than 20 aa, particularly CD8-mediated responses.15,24-26 These studies used individual peptides or small pools of peptides targeting individual epitopes in small cohorts. To assess the effect of 15mer peptides more fully, we validated the ELISPOT assay using gag, nef, and pol 15mer peptide pools with frozen-thawed PBMCs from 290 unvaccinated HIV-negative donors. Figure 1B shows the ELISPOT responses of these samples in the 2-dimensional region of positivity model. Responses by these seronegative donors to the 15mer peptide pools were similar to those seen in the 20mer validation study, where responses by the seronegative donors to the HIV peptides were relatively low and peptide responses were similar to corresponding mock responses. Therefore, the same positivity criteria can be applied to the 15mer peptide data (≥55 spots per 106 PBMCs and ≥4-fold over mock) to result in an observed false-positive rate of ≤1% to each of the 4 15mer peptide pools (see Table 1). We conducted additional validation studies to assess assay specificity and precision using a panel of samples with a range of 15mer responses. Samples were tested over multiple days by multiple operators (data not shown). The assay relative standard deviation (RSD) was estimated to be approximately 48% (95% CI: 41% to 58%).

Because major histocompatibility (MHC) class I-restricted CD8+ T-cell responses are considered to be an important component of an effective HIV vaccine, we sought to measure these responses in an unfractionated ELISPOT assay using pools of overlapping 9mer peptides. Analysis of responses to 9mer peptide pools has not been previously reported. We analyzed the ELISPOT responses using HIV gag, nef, and pol 9mer peptide pools with frozen-thawed PBMCs from 242 HIV-negative donors (see Fig. 1C). Responses by these seronegative donors to the 9mer peptide pools were generally of higher magnitude than those seen to 15mer and 20mer peptide pools in previous validation studies. Application of the positivity cutoff established for the 15mers and 20mers (≥55 spots per 106 PBMCs and ≥4-fold over mock) resulted in observed false-positive rates of 2% to 9% for the 9mer peptide pools (4% to 12% upper 90% CI; see Table 1). Alternate combinations of minimum spot count and minimum fold over mock were evaluated to target a ≤1% false-positive rate, and the lowest potential cutoff was determined to be ≥470 spots per 106 PBMCs (shown in Fig. 1C by the bold vertical line) and ≥4 times mock.

The use of this higher and more stringent cutoff diminished the sensitivity of the 9mer ELISPOT assay to detect responders, thereby conferring no advantage to the use of 9mer peptide pools over 20mer or 15mer peptide pools in the HIV ELISPOT assay. These false-positive 9mer responses seem to be donor dependent and were still evident in these donor samples at 9mer peptide pool concentrations as low as 62 ng/mL (data not shown). Epitope mapping of a subset of the 9mer responders leads us to suspect that at least some of these responses may be cross-reactive, where T cells are recognizing a similar epitope from a pathogen unrelated to HIV (data not shown). Based on our observations and reports of cross-reactivity observed in cytotoxic T lymphocytes (CTLs), as summarized in the article by Vieira and Chies,27 it seems plausible that detection of this cross-reactivity is more frequent when using 9mers compared with larger peptides, because the higher affinity of 9mer peptides for binding to MHC class I molecules leads to greater avidity for T-cell receptor (TCR), and therefore increased likelihood of stimulating T-cell responses.

Effect of Peptide Length in Detection of Enzyme-Linked Immunospot Responses in HIV-Infected Samples

ELISPOT responses to 20mer peptides were compared with 15mer responses for gag (n = 186; Fig. 2A) and nef (n = 139; see Fig. 2B) in HIV-infected donors from the United States, Brazil, South Africa, Malawi, and Cameroon. Responses were plotted relative to the 45° concordance line, and a sign test was used to determine if the fraction of data points above (or below) the concordance line was significantly different from 0.50. These data show that 15mer responses were significantly higher in magnitude than 20mer responses in 82% of subjects tested for gag and in 76% of those tested for nef (sign test P < 0.001). Using the positivity cutoff of ≥55 spots per 106 cells and ≥4-fold mock, the 15mer response rates (98% for gag and 91% for nef) were somewhat higher than the 20mer response rates (95% for gag and 88% for nef).

F2-4
FIGURE 2:
ELISPOT responses to 20mer peptides were compared with 15mer responses for gag (A) and nef (B) in HIV-infected donors from the United States, Brazil, South Africa, Malawi, and Cameroon (n = 186). ELISPOT data (spots per 1 million PBMCs) were plotted relative to the 45° concordance line, and the sign test was used to determine if the fraction of data points above (or below) the concordance line was significantly different from 0.50. P values <0.05 are considered statistically significant.

ELISPOT responses to 9mer peptides were also compared with those for 15mer peptides for gag (n = 37; Fig. 3A), nef (n = 37; see Fig. 3B), pol-1 (n = 18; see Fig. 3C), and pol-2 (n = 18; see Fig. 3D) in HIV-infected donors from Botswana. Responses to 9mers were higher in magnitude than responses to 15mers in 76% of subjects tested for gag (P < 0.001), 78% tested for nef (P < 0.001), 78% tested for pol-1 (P < 0.005), and 61% tested for pol-2 (P = 0.012). An increased magnitude with 9mers was also observed in the assay validation study using seronegative samples; therefore, the apparent increased sensitivity obtained with 9mers may not confer an advantage to detecting HIV-specific responses compared with 15mers or 20mers, because a higher positivity cutoff is required as a result.

F3-4
FIGURE 3:
ELISPOT responses to 9mer peptides were compared with 15mer responses for gag (A), nef (B), pol-1 (C), and pol-2 (D) in HIV-infected donors from Botswana (n = 37 for gag and nef, n = 18 for pol). ELISPOT data (spots per 1 million PBMCs) were plotted relative to the 45° concordance line, and the sign test was used to determine if the fraction of data points above (or below) the concordance line was significantly different from 0.50. P values <0.05 are considered statistically significant.

Effect of Peptide Length in Detection of Enzyme-Linked Immunospot Responses in Seronegative HIV Vaccinees

The validated ELISPOT assay was used to evaluate responses in vaccine study volunteers who received 4 priming doses of DNA gag (1 or 5 mg in saline or placebo), followed by a single boost vaccination with Ad5 gag at 1 × 109 or 1 × 1011 viral particles. Figure 4 shows ELISPOT responses to 20mer gag and 15mer gag before vaccination, 4 weeks after priming doses with DNA gag, and 4 weeks after Ad5 gag boost vaccination. Data for all 3 time points with both peptide pools were available for a cohort of 56 volunteers. Of these 56, 29 were positive to gag 15mer or gag 20mer at 4 weeks after boost vaccination, and their ELISPOT responses are shown in Figure 4 relative to the 45° concordance line. Prevaccination responses were comparable in both peptide pools, with a geometric mean response of 19 spots per 106 cells to 15mer and 20mer gag and no positive responses observed with either pool. After DNA priming doses, the overall magnitude of ELISPOT responses increased to a geometric mean of 63 to gag 15mer and 65 to gag 20mer and the response rate was 24% to both. There was no statistically significant difference in responses to either peptide length after DNA priming doses. Four weeks after Ad5 gag boost vaccination, the overall magnitude of ELISPOT responses increased further to a geometric mean of 299 for gag 15mer and 222 for gag 20mer. The post-Ad5 boost vaccination responses were statistically significantly higher to 15mer gag than to 20mer gag in 24 of 29 subjects (83%; P < 0.001) Using the positivity cutoff of ≥55 spots per 106 cells and ≥4-fold mock, the 15mer response rate was higher than the 20mer response rate (97% vs. 86%). These differences confirm our previously described observations with 15mer and 20mer peptide pools in HIV-infected volunteers.

F4-4
FIGURE 4:
ELISPOT responses to 20mer gag and 15mer gag before vaccination, 4 weeks after priming doses with DNA gag, and 4 weeks after Ad5 gag boost vaccination were compared in a cohort of 29 study volunteers who were positive to gag 15mer or gag 20mer at 4 weeks after boost vaccination. ELISPOT data (spots per 1 million PBMCs) were plotted relative to the 45° concordance line to compare 20mer and 15mer responses at each time point. The sign test was used to determine if the fraction of data points above (or below) the concordance line was significantly different from 0.50. P values <0.05 are considered statistically significant.

In another clinical vaccine study, ELISPOT responses to gag, nef, and pol peptides were evaluated for 72 subjects at 4 weeks after a single priming vaccination with Ad5 gag + Ad5 pol + Ad5 nef trivalent vaccine at 1 × 109, 1 × 1010, or 1 × 1011 viral particles. For this cohort, none of the prevaccination ELISPOT responses was ≥55 spots per 106 cells and ≥4-fold mock, and each subject had at least 1 positive response to any of the 15mer peptide pools at week 4. ELISPOT data were plotted as 15mer versus 9mer responses to each peptide pool relative to the 45° concordance line (Fig. 5). Overall, 15mer responses were similar to 9mer responses in these vaccinees, although responses to 9mers trended higher in magnitude in 58% of subjects for gag (P > 0.06), in 53% for nef (P > 0.27), and in 69% for pol-1 (P < 0.001). Conversely, responses to 15mer pol-2 were higher than responses to 9mer pol-2 in 61% of subjects (P < 0.03). Positive response rates based on the cutoff of ≥55 spots per 106 cells and ≥4-fold mock were similar for 15mers and 9mers; however, differences based on peptide length varied from one antigen to the next (79% 15mer vs. 83% 9mer gag, 67% vs. 61% nef, 46% vs. 58% pol-1, and 47% vs. 42% pol-2). However, the 9mer response rates using this cutoff may include false-positive responses at the rate of 2% to 9% (4% to 12% upper 90% CI) as determined in assay validation, even with the selection of subjects with no false-positive responses before vaccination. Therefore, the increased sensitivity provided by the 9mers in this vaccine cohort study may be offset by the decreased specificity of 9mers compared with 15mers. These observations and the earlier data showing higher responses with 15mers compared with 20mers lend further support to use of 15mer peptide pools over 20mer or 9mer peptide pools in the ELISPOT assay.

F5-4
FIGURE 5:
ELISPOT responses to 9mer and 15mer gag (A), nef (B), pol-1 (C), and pol-2 (D) peptides at 4 weeks after a single priming vaccination with Ad5gag-Ad5nef-Ad5pol at 1 × 109, 1 × 1010, or 1 × 1011 viral particles were compared in a cohort of 72 study volunteers. None of the prevaccination ELISPOT responses was ≥55 spots per 106 cells and ≥4-fold mock, and each subject had at least 1 positive response to any of the 15mer peptide pools at week 4. ELISPOT data (spots per 1 million PBMCs) were plotted relative to the 45° concordance line, and the sign test was used to determine if the fraction of data points above (or below) the concordance line was significantly different from 0.50. P values <0.05 are considered statistically significant.

DISCUSSION

Although an IFNγ ELISPOT assay is a relatively easy assay to conduct compared with other T-cell-based immunoassays, it requires appropriate sample handling16,18,28 and rigorous optimization and validation before it can be used in large-scale clinical testing. The ELISPOT assay described here represents the culmination of our optimization and validation efforts, resulting in a robust and sensitive ELISPOT assay that has been an effective tool in evaluating the immunogenicity of our HIV vaccine candidates in phase I/II studies.

We conducted validation studies to establish empirical positivity criteria (≥55 spots per 1 × 106 cells and ≥4-fold corresponding mock) that provide an observed false-positive rate of ≤1% against each of the 20mer or 15mer peptide pools. This criterion has also been cross-validated in more than 500 HIV-negative healthy volunteer samples using 3 statistical models (empiric, parametric, and nonparametric), and all 3 approaches confirmed an estimated false-positive rate of approximately 0.5%.29

The approach we described in this report for determining the responder/nonresponder status of an analyte in a clinical ELISPOT assay is an empiric one that has been adapted in a similar fashion by several other testing laboratories.30-32 In contrast, other laboratories have used various statistical approaches to define ELISPOT positivity.15,33 Cox and coworkers34 compared 5 different statistical and empirical methods for determining assay positivity, including our empirical cutoff, using data generated by 11 laboratories (excluding our laboratory). Our positivity criteria were successful in predicting both positive and negative controls in this study; however, it should be noted that the positivity criteria defined in this report strictly apply only to the ELISPOT procedures and reagents that were used to validate them. Similar methodology can be applied to other ELISPOT assays which use different procedures or reagents in order to establish empirical cutoffs appropriate for those assays. Our method for defining responder/nonresponder status is particularly attractive because of the ease with which it can generally be used by operators to determine the status on the basis of the assay results.

We included an analysis of the effect of peptide length in our assay validation and additional studies presented here and demonstrated that ELISPOT responses induced by pools of overlapping peptides of 9, 15, or 20 aa in length correlated well; however, 15mers offer the advantage of increased magnitude over 20mers (ie, greater sensitivity), ability to stimulate both CD4 and CD8 T cells, and fewer false-positive or cross-reactive responses than 9mers (ie, greater selectivity). Therefore, we have selected 15mer peptide pools for use in clinical trial monitoring.

The ultimate test for a validated assay is in a large-scale vaccine study in which the assay needs to be sensitive enough to detect vaccine-induced responses yet stringent enough to limit false-positive responses. We implemented our validated ELISPOT procedure and positivity criteria to measure vaccine immunogenicity in global epidemiology studies22,35 and in multiple phase I vaccine trials evaluating our DNA and Ad5 vaccine candidates. In 2 vaccine studies presented here, our observed false-positive rates were on target at ≤1% and we were able to detect HIV-specific responses after vaccination with DNA and/or adenovirus vector-based HIV vaccines, with positive ELISPOT responses in the analyzed subsets ranging from 61 to 256 spots per 106 cells after DNA vaccination and 55 to 4456 spots per 106 cells after Ad5 vaccination.

These studies and analyses support the conclusion that our validated ELISPOT assay using 15mer peptide pools and associated empirical positivity criteria provides a sensitive and robust assay for evaluation of HIV vaccine-induced cell-mediated immunity. An association between ELISPOT responses and vaccine efficacy remains to be shown. We are currently testing the trivalent Ad5 HIV vaccine in a phase II proof-of-concept clinical trial in which we plan to use this ELISPOT assay to explore potential correlates of protection.

ACKNOWLEDGMENTS

The authors thank Michael Robertson, Erin Quirk, Randi Leavitt, James Kublin, Daniel C. Freed, Paul Coplan, Alex Nikas, and Christopher Mast for their contributions in initiating and conducting the vaccine and epidemiologic studies described in this report. They also extend sincere appreciation to the volunteers who participated in these studies and to the principal investigators and staff who conducted these studies at their respective sites.

REFERENCES

1. Barresinoussi F, Chermann JC, Rey F, et al. Isolation of a T-lymphotropic retrovirus from a patient at risk for acquired immune-deficiency syndrome (AIDS). Science. 1983;220:868-871.
2. Gallo RC, Sarin PS, Gelmann EP, et al. Isolation of human T-cell leukemia-virus in acquired immune-deficiency syndrome (AIDS). Science. 1983;220:865-867.
3. Jin X, Bauer DE, Tuttleton SE, et al. Dramatic rise in plasma viremia after CD8(+) T cell depletion in simian immunodeficiency virus-infected macaques. J Exp Med. 1999;189:991-998.
4. Borrow P, Lewicki H, Hahn BH, et al. Virus-specific CD8+ cytotoxic T-lymphocyte activity associated with control of viremia in primary human-immunodeficiency-virus type-1 infection. J Virol. 1994;68:6103-6110.
5. Koup RA, Safrit JT, Cao YZ, et al. Temporal association of cellular immune-responses with the initial control of viremia in primary human-immunodeficiency-virus type-1 syndrome. J Virol. 1994;68:4650-4655.
6. Schmitz JE, Kuroda MJ, Santra S, et al. Control of viremia in simian immunodeficiency virus infection by CD8(+) lymphocytes. Science. 1999;283:857-860.
7. Shiver JW, Fu TM, Chen L, et al. Replication-incompetent adenoviral vaccine vector elicits effective anti-immunodeficiency-virus immunity. Nature. 2002;415:331-335.
8. Shiver JW, Emini EA. Recent advances in the development of HIV-1 vaccines using replication-incompetent adenovirus vectors. Annu Rev Med. 2004;55:355-372.
9. Letvin NL. Progress toward an HIV vaccine. Annu Rev Med. 2005;56:213-223.
10. Lalvani A, Brookes R, Hambleton S, et al. Rapid effector function in CD8(+) memory T cells. J Exp Med. 1997;186:859-865.
11. Schmittel A, Keilholz U, Scheibenbogen C. Evaluation of the interferon-gamma ELISPOT-assay for quantification of peptide specific T lymphocytes from peripheral blood. J Immunol Methods. 1997;210:167-174.
12. Sun Y, Iglesias E, Samri A, et al. A systematic comparison of methods to measure HIV-1 specific CD8 T cells. J Immunol Methods. 2003;272:23-34.
13. Mashishi T, Gray CM. The ELISPOT assay: an easily transferable method for measuring cellular responses and identifying T cell epitopes. Clin Chem Lab Med. 2002;40:903-910.
14. Kreher CR, Dittrich MT, Guerkov R, et al. CD4(+) and CD8(+) cells in cryopreserved human PBMC maintain full functionality in cytokine ELISPOT assays. J Immunol Methods. 2003;278:79-93.
15. Russell ND, Hudgens MG, Ha R, et al. Moving to human immunodeficiency virus type 1 vaccine efficacy trials: defining T cell responses as potential correlates of immunity. J Infect Dis. 2003;187:226-242.
16. Kierstead LS, Dubey SA, Meyer BE, et al. Enhanced rates and magnitude of immune responses detected against an HIV vaccine: effect of using an optimized process for isolating PBMC. AIDS Res Hum Retroviruses. 2007;23:86-92.
17. Lathey JL. Preliminary steps toward validating a clinical bioassay-a case study of the ELISpot assay. Biopharm International-The Applied Technologies of Biopharmaceutical Development. 2003;16:42-50.
18. Janetzki S, Cox JH, Oden N, et al. Standardization and validation issues of the ELISPOT assay. In: Kalyuzhny A, ed. Handbook of ELISPOT: Methods and Protocols. Humana Press, Totowa, NJ; 2005:51-86.
19. Mwau M, McMichael AJ, Hanke T. Design and validation of an enzyme-linked immunospot assay for use in clinical trials of candidate HIV vaccines. AIDS Res Hum Retroviruses. 2002;18:611-618.
20. Hudgens MG, Self SG, Chiu YL, et al. Statistical considerations for the design and analysis of the ELISpot assay in HIV-1 vaccine trials. J Immunol Methods. 2004;288:19-34.
21. Czerkinsky C, Andersson G, Ekre HP, et al. Reverse ELISpot assay for clonal analysis of cytokine production. 1. Enumeration of gamma-interferon-secreting cells. J Immunol Methods. 1988;110:29-36.
22. Coplan PM, Gupta SB, Dubey SA, et al. Cross-reactivity of anti-HIV-1 T cell immune responses among the major HIV-1 clades in HIV-1-positive individuals from 4 continents. J Infect Dis. 2005;191:1427-1434.
23. Fu TM, Dubey SA, Mehrotra D, et al. Evaluation of cellular immune responses in subjects chronically infected with HIV-1. AIDS Res Hum Retroviruses. 2007;23:67-76.
24. Maecker HT, Dunn HS, Suni MA, et al. Use of overlapping peptide mixtures as antigens for cytokine flow cytometry. J Immunol Methods. 2001;255:27-40.
25. Draenert R, Altfeld M, Brander C, et al. Comparison of overlapping peptide sets for detection of antiviral CD8 and CD4 T cell responses. J Immunol Methods. 2003;275:19-29.
26. Kiecker F, Streitz M, Ay B, et al. Analysis of antigen-specific T-cell responses with synthetic peptides-what kind of peptide for which purpose? Hum Immunol. 2004;65:523-536.
27. Vieira GF, Chies JAB. Immunodominant viral peptides as determinants of cross-reactivity in the immune system-can we develop wide spectrum viral vaccines? Med Hypotheses. 2005;65:873-879.
28. Smith JG, Liu X, Kaufhold RM, et al. Development and validation of a gamma interferon ELISPOT assay for quantitation of cellular immune responses to varicella-zoster virus. Clin Diagn Lab Immunol. 2001;8:871-879.
29. Mogg R, Fan F, Li X, et al. Statistical cross-validation of Merck's IFN-gamma ELIspot assay positivity criterion [abstract 495]. Presented at AIDS Vaccine 2003, September 18-21, 2003; New York, NY.
30. Alter G, Merchant A, Tsoukas CM, et al. Human immunodeficiency virus (HIV)-specific effector CD8 T cell activity in patients with primary HIV infection. J Infect Dis. 2002;185:755-765.
31. Slyker JA, Lohman BL, Mbori-Ngacha DA, et al. Modified vaccinia Ankara expressing HIVA antigen stimulates HIV-1-specific CD8 T cells in ELISpot assays of HIV-1 exposed infants. Vaccine. 2005;23:4711-4719.
32. Goonetilleke N, Moore S, Dally L, et al. Induction of multifunctional human immunodeficiency virus type 1 (HIV-1)-specific T cells capable of proliferation in healthy subjects by using a prime-boost regimen of DNA- and modified vaccinia virus Ankara-vectored vaccines expressing HIV-1 gag coupled to CD8(+) T-cell epitopes. J Virol. 2006;80:4717-4728.
33. Mulligan MJ, Russell ND, Celum C, et al. Excellent safety and tolerability of the human immunodeficiency virus type 1 pGA2/JS2 plasmid DNA priming vector vaccine in HIV type 1 uninfected adults. AIDS Res Hum Retroviruses. 2006;22:678-683.
34. Cox JH, Ferrari G, Kalams SA, et al. Results of an ELISPOT proficiency panel conducted in 11 laboratories participating in international human immunodeficiency virus type 1 vaccine trials. AIDS Res Hum Retroviruses. 2005;21:68-81.
35. Gupta SB, Mast CT, Wolfe ND, et al. Cross-clade reactivity of HIV-1-specific T-cell responses in HIV-1-infected individuals from Botswana and Cameroon. J Acquir Immune Defic Syndr. 2006;42:135-139.
Keywords:

ELISPOT; HIV-1; interferon-γ; interferon-γ enzyme-linked immunospot assay; peptides; vaccine

© 2007 Lippincott Williams & Wilkins, Inc.