Lymphocytic Choriomeningitis Virus Infection Yields Overlapping CD4⁺ and CD8⁺ T-Cell Responses

Peptide synthesis.Fifteen-mer peptides overlapping by 11 amino acids for the entire proteome of LCMV strain Armstrong (GP, NCBI accession no. AAX49341; NP, NCBI accession no. AAX49342; L, NCBI accession no. AAX49344; Z, NCBI accession no. AAX49343) were synthesized (Pepscan, Lelystad, The Netherlands). The resultant 664 LCMV peptides for the entire proteome were divided into 83 pools, with eight peptides per pool for initial screening purposes. Specific peptides were subsequently tested individually at a concentration of 10 μg/ml.

Mice and infections.C57BL/6 (H-2^b) mice were purchased from The Jackson Laboratory (Bar Harbor, ME). All studies were conducted at the California State University—San Marcos and the La Jolla Institute for Allergy and Immunology, in facilities approved by the Association for Assessment and Accreditation of Laboratory Animal Care and according to Institutional Animal Care and Use Committee-approved animal protocols.

C57BL/6 (H-2^b) mice were infected with 2 × 10⁵ PFU of LCMV Armstrong intraperitoneally (i.p.). Mice were sacrificed by CO₂ inhalation, and spleens were harvested 8 days postinfection.

For LCMV clone 13 experiments, C57BL/6 mice were infected with 2 × 10⁶ PFU of LCMV clone 13 intravenously. Mice were sacrificed by CO₂ inhalation, and spleens were harvested 8 days postinfection.

ELISPOT assay.The gamma interferon (IFN-γ) ELISPOT assays were performed as previously described (30, 48). Briefly, mouse CD4⁺ T cells were isolated from the spleens of LCMV-infected mice with anti-CD4⁺ magnetic beads (Miltenyi Biotech, Inc., Auburn, CA). Purified CD4⁺ T cells (1 × 10⁵ to 2 × 10⁵) were cultured with 1 × 10⁵ to 2 × 10⁵ syngeneic splenocytes from uninfected mice and peptides (either peptide pools or individual peptides, tested at 10 μg/ml) in flat-bottom 96-well nitrocellulose plates (Immobilon-P membrane; Millipore, Bedford, MA), which had been precoated with 2 μg/ml anti-mouse IFN-γ monoclonal antibody (Mabtech, Inc., Mairemont, OH). After 20 h, plates were washed with phosphate-buffered saline (PBS)-0.5% Tween 20 and then incubated with 1 μg/ml biotinylated anti-mouse IFN-γ monoclonal antibody (Mabtech) for 3 h at 37°C. After additional washes with PBS-0.5% Tween, spots were developed by incubation with Vectastain ABC peroxidase (Vector Laboratories, Burlingame, CA) and then 3-amino-9-ethylcarbazole solution (Sigma-Aldrich, St. Louis, MO) and counted by computer-assisted image analysis (Zeiss KS ELISPOT reader).

Experimental values were expressed as the mean net spots per million CD4⁺ cells for each peptide pool or individual peptide. For the initial screening of the 83 pools, responses against each pool were considered positive if (i) the number of spot-forming cells (SFCs)/10⁶ CD4⁺ T cells exceeded the absolute value of the mean negative control wells (effectors plus antigen-presenting cells [APCs] without peptide) by twofold, (ii) the value exceeded 50 SFCs/10⁶ CD4⁺ cells, and (iii) these conditions were met in at least two replicate independent experiments. Positive pools were deconvoluted into their eight individual components and tested again, to determine which individual peptides were responsible for the pooled IFN-γ response. Responses against individual peptides were considered positive if they exceeded the threshold of the mean negative control wells (effectors plus APCs without peptide) by twofold and exceeded a threshold of 200 SFCs/10⁶ CD4⁺ cells.

ICCS for IFN-γ.ICCS assays were performed as previously described (30). Splenocytes from LCMV-infected mice were cultured in 96-well U-bottom plates (1 × 10⁶ cells per well) in complete RPMI (RPMI 1640 with 10% fetal bovine serum, 2 mM l-glutamine, 1 U/ml penicillin G, and 100 μg/ml streptomycin), at 37°C with 5% CO₂ for 1 to 2 h, in the presence of the indicated peptides (at a final concentration of 10 μg/ml). Brefeldin A (1 μg/ml) was added 2 h later, and the cells continued in culture for 6 to 10 h. Cells were harvested, washed with PBS containing 5% fetal bovine serum, and stored on ice. Cells were stained according to the BD Cytofix/Cytoperm Plus kit (BD Biosciences, San Jose, CA) with antibodies to CD4 or a combination of CD4 with CD62L and/or CD8 and anti-IFN-γ antibodies (BD Pharmingen, San Jose, CA). The cells were collected using a BD FACSCalibur, and data were analyzed with FlowJo software (Tree Star, Inc., Ashland, OR). A CD4⁺/IFN-γ⁺ response was considered positive if it exceeded the mean background (response against irrelevant peptide) + 3 standard deviations (SDs), corresponding to 0.73%. A CD8⁺/IFN-γ⁺ response was considered positive if it exceeded the mean background (response against irrelevant peptide) + 3 SDs, corresponding to 0.59%.

IA^b-peptide binding assays.H-2^b class II MHC was purified, and peptide binding assays were performed, essentially as previously described (45). Briefly, mouse B-cell lymphoma LB27.4 cells were utilized as sources of murine IA^b. MHC molecules were purified by affinity chromatography using the anti-IA^b monoclonal antibody Y3JP. Quantitative peptide binding assays were based on the inhibition of binding of radiolabeled probe peptide ROIV (peptide 569.01, an artificial ligand with sequence YAHAAHAAHAAHAAHAA) to purified IA^b molecules. Assays were performed at pH 5.5 in PBS containing 1% digitonin and in the presence of a cocktail of protease inhibitors (45). MHC binding of the radiolabeled peptide was determined by capturing MHC-peptide complexes on antibody-coated Lumitrac 600 plates (Greiner Bio-one, Frickenhausen, Germany) and measuring bound cpm using the TopCount (Packard Instrument Co., Meriden, CT) microscintillation counter. The average 50% inhibitory concentration (IC₅₀) of ROIV (peptide 569.01) for IA^b was 28 nM.

Memory and secondary challenge of LCMV Armstrong.Two groups of six C57BL/6 mice were infected with 2 × 10⁵ PFU LCMV Armstrong as previously described. Three mice from each group were sacrificed by CO₂ inhalation 35 days postinfection and tested in ELISPOT assays as previously described. The three remaining mice from each group were infected with 2 × 10⁵ PFU LCMV Armstrong for a secondary challenge at day 35 of the original infection, as previously described. Eight days postinfection (or 43 days post-original infection), each group was sacrificed by CO₂ inhalation and tested in ELISPOT assays as described above. Uninfected C57BL/6 mice were used as controls.

Sequence analysis.Amino acid sequences of the GP and NP proteins were aligned using EMBOSS pairwise alignment algorithms (http://www.ebi.ac.uk/Tools/emboss/align/ ). Each sequence was individually compared to LCMV Armstrong. The following sequences were used: LCMV Armstrong 53b (NP 694851), LCMV clone 13 (DQ361066), LCMV WE (AAA46265), LCMV CHV1 (AAA82176), LCMV CHV2 (AAA82177), LCMV CHV3 (AAA82178), LCMV WE-HPI (CAC01231), LCMV CH-5692 (AAL01687), and LCMV CH-5871 (AAL01687). The sequences were derived from the Arenavirus Sequence Database (http://epitope.liai.org:8080/projects/arena/arena_home ). Each CD4⁺ T-cell epitope was compared to the peptide sequences for sequence homology.

Statistical analyses.For each peptide or pool studied, three mice were used per experiment, and either the splenocytes were pooled (ELISPOT) or each mouse was tested independently (ICCS). All peptides and pools were tested in triplicate and in three or more independent experiments. The average and SD of the triplicate samples were then calculated. Responses among groups of mice are presented as means with SDs. Comparisons of mean immune responses were performed using one-sided t tests, including Fisher's exact test, with P values of <0.05 considered significant.

Identification of LCMV-derived CD4⁺ T-cell epitopes.A set of 664 overlapping peptides spanning the entire LCMV Armstrong proteome was subdivided into 83 pools of eight peptides each, grouped in sequential order. To identify CD4⁺ T-cell epitopes, C57BL/6 mice were infected with 2 × 10⁵ PFU of LCMV Armstrong via i.p. inoculation and 8 days later splenic CD4⁺ T cells were enriched and stimulated with each peptide pool. CD4⁺ T-cell reactivity against each pool was determined in standard IFN-γ ELISPOT assays. A total of 17 positive pools from NP, GP, and L proteins were identified (data not shown).

To define specific epitopes, positive pools were decoded, and each of the eight peptides contained within a positive pool was tested. Responses greater than twice the mean negative control wells and exceeding 200 SFCs/10⁶ CD4⁺ cells were considered positive. Nine pools failed to yield significant responses against individual peptides (data not shown). In the remaining eight pools, 10 antigenic peptides were identified that met the positivity criteria. Figure 1 shows the positive responses observed against individual peptides from the original positive pools. The 10 responses detected against peptides were located in their source proteins at GP 6-20, GP 31-45, GP 61-75, GP 66-80, GP 126-140, GP 171-185, GP 186-200, NP 201-215, NP 236-250, and NP 311-325.

• Open in new tab
• Download powerpoint

FIG. 1.

Identification of LCMV-derived CD4⁺ epitopes by ELISPOT. C57BL/6 (H-2^b) mice were infected i.p. with 2 × 10⁵ PFU of LCMV Armstrong. Eight days postinfection, CD4⁺ T-cells were purified and tested against each of the eight peptides contained in each positive pool at a concentration of 10 mg/ml per peptide. Responses against individual peptides were considered positive if they exceeded the threshold of double the mean negative control wells (effectors plus APCs without peptide) and exceeded a threshold of 200 SFCs/10⁶ CD4⁺ cells. Each positive peptide is labeled by its proteomic location.

To verify that the IFN-γ responses observed in the ELISPOT assays were indeed due to CD4⁺ T-cells, we performed ICCS experiments. Mice were infected with LCMV, and 8 days later, splenocytes were isolated and incubated with each of the peptides and then stained for intracellular IFN-γ and CD4 and CD8 surface staining. The 10 peptides identified as positive in the ELISPOT assays, as described above, were tested (GP 6-20, GP 31-45, GP 61-75, GP 66-80, GP 126-140, GP 171-185, GP 186-200, NP 201-215, NP 236-250, and NP 311-325). In Fig. 2, the mean responses against each of the peptides are shown for both CD4⁺/IFN-γ⁺ (Fig. 2A) and CD8⁺/IFN-γ⁺ (Fig. 2B) cells. The percentage of CD4⁺ cells producing IFN-γ in response to each peptide was measured from LCMV Armstrong-infected animals. Each of the peptides was tested using at least six different mice. The percentage of CD4⁺/IFN-γ⁺ releasing cells from unstimulated mice was 0.16% ± 0.57% (mean + 3 SDs), and accordingly, we defined a value of 0.73% as a threshold value to consider IFN-γ responses against a given peptide as positive.

• Open in new tab
• Download powerpoint

FIG. 2.

CD4⁺ versus CD8⁺ IFN-γ production as measured by intracellular cytokine staining. C57BL/6 (H-2^b) mice were infected i.p. with 2 × 10⁵ PFU of LCMV Armstrong. Eight days postinfection, splenocytes were purified and tested in ICCS assays measuring IFN-γ production and surface staining for CD4 (A) and CD8 (B) against 10 peptides identified as positive in the ELISPOT assays.

Out of the 10 peptides tested, 6 peptides identified in the ELISPOT assay scored positive for IFN-γ production by gated CD4⁺ cells in the ICCS assay (Fig. 2A). The percentage of CD4⁺/IFN-γ⁺ cells ranged from 0.25 to 3.04% against the peptide panel. The strongest response observed was against GP 66-80, followed by GP 126-140, GP 186-200, NP 311-325, GP 31-45, and GP 6-20. Interestingly, peptide GP 61-75, nested within the previously characterized GP 61-80 epitope, did not give a significant response in the ICCS, whereas the overlapping peptide, GP 66-80, elicited the strongest response.

Next, we assessed if the lack of response against some peptides in the ICCS assay could be explained by contaminating CD8⁺ cells. Figure 2B shows the IFN-γ production by CD8⁺ T cells in response to each of the 10 peptides. The response observed in unstimulated mice was 0.20% ± 0.39% (mean + 3 SD), yielding a threshold value of 0.59% to consider a response against a given peptide as positive. For the four peptides which did not yield significant IFN-γ production by CD4⁺ T cells, GP 61-75, GP 171-185, NP 201-215, and NP 236-250, three peptides yielded strong CD8⁺/IFN-γ⁺ staining: GP 171-185, NP 201-215, and NP 236-250. Out of the remaining six peptides, which had significant IFN-γ responses by CD4⁺ T cells, for two peptides, GP 6-20 and GP 186-200, the newly identified epitopes were not associated with CD8⁺/IFN-γ⁺ staining. However, each of the other four epitopes yielded CD8⁺/IFN-γ⁺ staining. GP 31-45 contains the previously identified MHC class I-restricted epitope, GP 33-41 (34, 42, 50, 54). Indeed, the GP 31-45 peptide elicited a CD8⁺ T-cell response, with 11.7% of the CD8⁺ cells producing IFN-γ, thus demonstrating that this region contains overlapping CD4⁺ and CD8⁺ T-cell epitopes. The NP 311-325 epitope contains peptide NP 314-322, predicted to bind to the MHC molecule K^b with an affinity of 774 nM (J. Sidney, personal communication). The NP 311-325 peptide elicited a CD8⁺ T-cell response with 0.66% of the CD8⁺ cells producing IFN-γ. GP 66-80 contains a recently identified epitope, GP 67-77, which is restricted by D^b (20, 27). GP 126-140 also elicited a CD8⁺/IFN-γ response consisting of 0.97%. Therefore, out of the six CD4⁺ epitopes, four of them also yielded CD8⁺/IFN-γ⁺ staining.

We were interested in determining whether the overlap between the set of CD4⁺ and CD8⁺ T-cell epitopes is statistically significant. To pursue this, each of the 664 15-mer peptides overlapping the LCMV proteome was analyzed for the presence or absence of a CD4⁺ T-cell response as well as whether it contained a nested CD8⁺ T-cell epitope in the setting of H-2^b. The Fisher exact test (one tailed) yielded a P value of 0.0033. We further examined if this overlap was due only to a preferred recognition of peptides from the GP and NP proteins. Limiting the same test to the 208 peptides overlapping GP and NP still gives a significant association of CD4 and CD8 response targets (P = 0.0093). This demonstrates that the overlap of the two sets of immune response targets is statistically significant.

MHC peptide binding affinity of identified epitopes.To determine the MHC binding affinity for each of the identified epitopes, we conducted MHC-peptide binding assays. H-2^b class II MHC was purified, and quantitative inhibition binding assays were performed for each peptide using IA^b, as described in Materials and Methods (45). The binding assays revealed that five out of the six peptides bound to IA^b with an affinity of less than 11,000 nM (Table 1). Although found to yield strong immunogenicity responses, as measured in both ELISPOT and ICCS assays, peptide GP 186-200 did not efficiently bind to IA^b. The greatest affinity to IA^b was detected against GP 66-80, which interestingly also elicited the strongest ICCS response. Table 1 summarizes the binding affinities as well as the IFN-γ levels detected both by ELISPOT and ICCS.

Characterization of responses after chronic infection.To determine if responses were present after infection with clone 13, a virus which yields persistent infection, two groups of three mice per group were infected with 2 × 10⁶ PFU LCMV clone 13 intravenously. Eight days postinfection, the mice were sacrificed by CO₂ inhalation and tested in IFN-γ ELISPOT assays with the six peptides which resulted in responses after LCMV Armstrong infection. Responses greater than twice the mean of negative control wells and exceeding 200 SFCs/10⁶ CD4⁺ T cells were considered positive. Of the six characterized CD4⁺ epitopes identified above, responses were observed against two epitopes, GP 66-80 and GP 126-140. Figure 3 shows the responses observed against the individual peptides. We observed weak and not statistically significant responses against GP 31-45, GP 186-200, and NP 311-325. The epitope GP 6-20 was negative in two replicate experiments.

• Open in new tab
• Download powerpoint

FIG. 3.

Identification of CD4⁺ responses after chronic infection. C57BL/6 (H-2^b) mice were infected retroorbitally with 2 × 10⁶ PFU of LCMV clone 13. Eight days postinfection, CD4⁺ T cells were purified and tested against each of the six epitopes at a concentration of 10 mg/ml per peptide. Responses against individual peptides were considered positive if they exceeded the threshold of double the mean negative control wells (effectors plus APCs without peptide) and exceeded a threshold of 200 SFCs/10⁶ CD4⁺ cells. Each peptide is labeled by its proteomic location.

Magnitude of CD4⁺ T-cell epitopes in the memory stage and after secondary challenge with LCMV Armstrong.To determine the magnitude of CD4⁺ T-cell responses during the memory phase of LCMV Armstrong infection, C57BL/6 mice were infected with 2 × 10⁵ PFU LCMV Armstrong and tested in ELISPOT assays as described in Materials and Methods.

The responses observed after 35 days postinfection were from two of the six epitopes tested (GP 66-80 and NP 311-325). Both epitopes met the criteria established for a positive response (twice the mean for negative control wells and exceeding 200 SFCs/10⁶ CD4⁺ T cells). Furthermore, after secondary challenge with 2 × 10⁵ PFU LCMV Armstrong at 35 days of infection, we observed similar responses (Fig. 4). No response was detected with the uninfected control mice. After both infections, weak responses were observed for GP 31-45, GP 126-140, and GP 186-200. The GP 6-20 epitope was negative in two replicate experiments.

• Open in new tab
• Download powerpoint

FIG. 4.

Identification of CD4⁺ responses during memory stage and secondary infection of LCMV Armstrong. C57BL/6 (H-2^b) mice were infected i.p. with 2 × 10⁵ PFU of LCMV Armstrong. Thirty-five days postinfection, three mice from each group were used in ELISPOT assays and three mice from each group were challenged with a second (2°) infection of LCMV Armstrong. Eight days postinfection, CD4⁺ T cells were purified and tested against each of the six epitopes at a concentration of 10 mg/ml per peptide. Responses against individual peptides were considered positive if they exceeded the threshold of double the mean negative control wells (effectors plus APCs without peptide) and exceeded a threshold of 200 SFCs/10⁶ CD4⁺ cells. Each positive peptide is labeled by its proteomic location.

Sequence analysis of newly identified CD4⁺ T-cell epitopes.To determine whether there was sequence homology among the newly characterized CD4⁺ T-cell epitopes, sequences from several strains of LCMV were aligned to reveal any differences among the strains of LCMV using the sequence of the newly identified CD4⁺ T-cell epitopes. Table 2 shows the sequence alignments among various strains of LCMV. There were no amino acid changes from LCMV Armstrong to LCMV clone 13.

There were several changes in the overlapping CD4/CD8 epitope GP 31-45, where there was a G→S change at position 1 in seven of the nine strains compared. There was an I→L change at position 13 in three strains. There were an F→L change at position 14 in two strains and an A→T change at position 15 in two strains. The other five epitopes had few changes among the different strains. This shows us that while the majority of the epitopes were conserved between the several different strains assessed, differences were observed in the immunodominant CD8⁺ epitope GP 31-45.