Additional testing in A2.1 Tg mice reveals a more extensive repertoire overlap

We previously detected 14 different epitopes utilizing PBMC from A2.1-positive human volunteers vaccinated with Dryvax (Figure 1A) [9]. Of these epitopes, only two were also independently identified in VACV-infected A2.1 Tg mice [5, 13, 14, 15, 16]. Based on these results, Terajima and Ennis hypothesized that the large number of potentially immunogenic peptides encoded by a complex pathogen such as VACV might explain the incomplete overlap observed [17]. Herein, we addressed this issue in more detail.

First, A2.1 Tg mice were inoculated with VACV and the reactivity against each of the 28 individual A2.1-restricted epitopes was evaluated. Responses were detected for all epitopes previously reported in HLA Tg mice (Figure 1A; white and grey bars). In addition, responses were detected for 6 peptides that had been originally described in A2.1-positive humans [9, 14, 18], but not reported in studies utilizing Tg mice [5, 15] (Figure 1A; black bars). Altogether, responses directed against 8 of 14 (57%) epitopes reported in humans were now also detected in A2.1 Tg mice.

The present epitope screen differs in two experimental settings from the original study by Pasquetto et al. [5]. First, individual peptides were used here instead of peptide pools. Second, the initial epitope identification used splenocytes as effector cells [5], whereas the present study utilized purified CD8⁺ T cells. Herein we examined the impact of these two experimental differences. When three representative epitopes (A14L_51–59, B14R_327–335, and M1L_374–383) were tested either as individual peptides or in pools of 10 peptides, similar response magnitudes were observed (Figure 1B). We also tested the response to the same 3 epitopes using either purified CD8⁺ T cells or splenocytes, and found that the higher non-specific interferon (IFN)-γ production obtained when using splenocytes led to a reduced stimulation index, and thereby decreased sensitivity (Figure 1C).

We observed that the average responses against these epitopes not initially identified in Tg mice were ~3 times lower than those against the epitopes identified in the original screen (P = 0.0098; Figure 1D). Although the 3 dominant epitopes (A14L_51–59, A17L_81–90, and M1L_347–383) contribute to a significant portion of the higher response identified in mice, exclusion of these 3 epitopes still results in a significant difference in the magnitude of the response for the epitopes identified in humans versus mouse (P = 0.0012). Taken together, these results suggest that while 8 of 14 (57%) peptides reported in humans can also be detected in A2.1 Tg mice, differences in the patterns of immunodominance might render them less prominent and consequently make their detection more difficult.

The mice utilized by Pasquetto et al. were A2.1 bxd mice [5]. To examine whether expression of endogenous mouse MHC molecules might also contribute to the incomplete overlap, we tested the reactivity of the 14 epitopes identified in humans using CD8⁺ T cells from VACV-infected HHD A2.1 mice, which do not express endogenous mouse MHC [19]. However, no additional epitopes were recognized in the HHD mice, and the response magnitude and immunodominance pattern was similar to that in A2.1 bxd mice (data not shown).

Testing of a broader panel of human vaccinees reveals a more extensive overlap

As described above, a total of 16 epitopes were initially identified in HLA Tg mice [5, 13, 15]. Of these, only two (13%) were also independently identified in human vaccinees (Figure 1A). Most of the candidate peptides were tested in only 6 donors, with each individual donor yielding a rather distinct pattern of epitope recognition. While some epitopes were identified in multiple donors, most of the epitopes were only detected in 1 out of the 6 donors investigated [9].

To test the hypothesis that the epitopes not initially detected in humans were associated with lower and/or less frequent responses, we examined the responses to the set of 16 A2.1-restricted epitopes identified in Tg mice using PBMC from 8 different A2-positive vaccinees. The samples tested included frozen cell aliquots from 5 of the donors tested in the original study of Oseroff et al. [5], as well as PBMC from 3 additional A2-positive vaccinees. PBMC were cultured in the presence of individual peptides, and the numbers of IFN-γ-producing cells were enumerated in an enzyme-linked immunosorbent spot (ELISPOT) assay. The same two peptides previously identified as positive in these donors were also re-identified in the course of the present experiments, and responses to an additional 5 peptides originally identified in A2.1 Tg mice, but not identified in the human donor screen of Oseroff et al. [9], were detected. Similar to what we observed in the mouse experiments, responses to the newly detected epitopes were in general lower than to epitopes identified in the initial screen (see Additional file 1). Overall, responses directed against 7 of 16 (44%) peptides initially identified in HLA Tg mice, were also detected in humans. Considering the large donor-to-donor variation observed, it is likely that testing of additional donors would further increase the overlap between the two systems.

Peptides

Peptides were synthesized by A and A Labs and purified to >95% homogeneity by reverse-phase high-performance liquid chromatography (HPLC). Purity was determined using analytical reverse-phase HPLC and amino acid analysis, sequencing, and/or mass spectrometry. Peptides were radio-labeled with the chloramine T method as described [31].

Characteristics of study population

Healthy males and females between 18 and 59 years of age were used in this study. Exclusion criteria were body weight of < 45.4 kg and established pregnancy. Recruited donors were vaccinated by arm scarification with a VACV (Dryvax) vaccination as a prophylactic measure either because of their potential exposure to VACV in a laboratory or hospital setting or because of their enrollment into military and health worker vaccination program. The blood draw was performed within 1 year of the Dryvax vaccination. All experiments carried out with human subjects were in compliance with the Helsinki Declaration. Institutional Review Board approval (FWA# 00000032) and appropriate informed consent were obtained for this study.

PBMC isolation and HLA typing

PBMC were isolated from heparinized blood by gradient centrifugation with a Histopaque-1077 (Sigma-Aldrich, St. Louis, MO) [32], suspended in fetal bovine serum containing 10% dimethyl sulfoxide, and cryo-preserved in liquid nitrogen. Donor's PBMC were typed for HLA-A by high-resolution polymerase chain reaction (Forensic Analytical Molecular Genetics, San Francisco, CA).

Mice

A2.1 Tg mice used in this study were the F₁ generation derived from crossing homozygous Tg mice (H-2^b haplotype) expressing a chimeric gene (A2.1/K^b) consisting of the α1 and α2 domains of HLA and the α3 domain of H-2K^b with BALB/c mice (The Jackson Laboratory, Bar Harbor, ME) [4, 33, 34]. The HHD A2.1 strain [19] was kindly provided by Dr. Jack Bennink (NIAID, NIH, Bethesda, MD). All mice were bred and maintained following NIH guidelines and Institutional Animal Care and Use Committee-approved animal protocols (AAALAC# 000840 and OLAW# A3779-01).

Viruses and infection

The Western Reserve strain of VACV was obtained from Dr. Bernard Moss (NIAID). Mice were infected i.p. with 2 × 10⁶ PFU of VACV. On day 7 post-infection, the mice were sacrificed and splenic CD8⁺ T cells were used in mouse IFN-γ ELISPOT assays.

Ex vivo IFN-γ ELISPOT

IFN-γ ELISPOT assays were performed as described [5, 9]. In brief, for murine assays, 2 × 10⁵ splenic CD8⁺ T cells were cultured with 10⁵ human Jurkat cells (expressing the same A2.1/K^b construct as in the Tg mice) pulsed with 10 μg/ml of peptide. For human assays, 2 × 10⁵ PBMC were incubated with 5 μg/ml of peptide. After a 20 h incubation at 37°C, plates were developed, and responses calculated as described [5, 9, 19]. Criteria for positivity were net spot-forming cells (SFC)/10⁶ cells ≥ 20, stimulation index ≥ 2.0, and p-value ≤ 0.05 using a Student's t test in at least 2 out of 3 experiments.