Antigenic Epitopes Regulate the Phenotype of CD8⁺ CTL Primed by Exogenous Antigens

Mice

Female 8- to 12-wk-old B6 mice were purchased from the National Cancer Institute (Frederick Cancer Research and Development Center, Frederick, MD). IL-4-deficient, IL-4^tm1Nnt (16) (IL-4^−/−) mice were purchased from The Jackson Laboratory (Bar Harbor, ME). All procedures on animals were conducted according to the principles outlined in the guidelines of the Committee on Care and Use of Laboratory Animals, Institute of Laboratory Animal Resources, National Research Council.

Reagents

Purified OVA (grade VI) and crystalline BINS were purchased from Sigma (St. Louis, MO). CFA containing Mycobacterium tuberculosis strain H37Ra and IFA was obtained from Difco (Detroit, MI). Emulsions of OVA, BINS, or their peptides in CFA or IFA were prepared by mixing aqueous Ag solution with oil phase at a 1:1 ratio and achieved a final concentration of 1 mg/ml as described previously (14). Dr. Maurice Gately (Hoffman-La Roche, Nutley, NJ) generously provided recombinant human IL-2. The OVA_257–264 (SIINFEKL) and INS A_12–21 (SLYQLENYCN) peptides used in this paper were synthesized at the Emory University Microchemical Facility or provided by Dr. Brian Evavold (Emory University, Atlanta, GA).

Tumor cells and cell cultures

H-2^b thymoma cells (EL4) were purchased from American Type Culture Collection (Manassas, VA). EG7-OVA, generated by transfection of EL4 cells with the OVA cDNA gene (17), were kindly provided by Dr. M. J. Bevan (University of Washington, Seattle, WA). EL4-INS cells were generated by tranfection of EL4 cells with human proinsulin DNA (18). RMA-S (H-2^b) cells (19) were generously provided by Dr. K. Karre (Stockholm, Sweden). All cell cultures were maintained in RPMI 1640 medium supplemented with 5% FBS, 1 mM l-glutamine, 1 mM sodium pyruvate, 50 μM 2-ME, and antibiotics at 37°C in 6% CO₂ in air.

Immunization and establishment of T cell lines

Unless otherwise noted, B6 mice were immunized in the hind footpad with 100 μg of OVA or BINS in CFA as previously described (14). In some experiments, 100 μg each of OVA and BINS were mixed together in CFA. After 10–14 days, splenocytes were cultured with irradiated (20,000 rad) stimulator cells, E.G7-OVA, or EL4-INS. Thereafter, T cells were restimulated weekly with irradiated syngeneic splenocytes and stimulator cells plus 20 U/ml recombinant human IL-2. For peptide priming, B6 mice were immunized with 50 μg of the indicated peptides in CFA at the hind footpad. After 12 days, these mice were boosted with 50 μg of the same peptide in IFA. One week later, splenocytes from mice immunized with peptides were stimulated with RMA-S cells pulsed with the corresponding peptides rather than E.G7-OVA or EL4-INS in vitro.

Cr ⁵¹ release assay

Cytotoxicity was quantified by a 4-h ⁵¹Cr release assay (14). Syngeneic targets (E.G7-OVA, EL4-INS, or peptide-pulsed RMA-S cells) were labeled with Na₂⁵¹CrO₄ (DuPont, Boston, MA) at 37°C for 1 h. After washing, ⁵¹Cr-labeled target cells were incubated with INS-CTL or OVA-CTL at different E:T ratios in 96-well round-bottom plates. After 4-h incubation at 37°C, supernatants were collected and radioactivity was detected in a scintillation counter (Wallac, Turku, Finland). Percent specific lysis was calculated as 100 × ([release by CTL − spontaneous release]/[maximal release − spontaneous release]). Maximal release was determined by the addition of 1% Triton X-100 (EM Science, Gibbstown, NJ). The spontaneous release, in the absence of CTL, was generally <15% of the maximal release.

Cytokine assay

Effector T cells (10⁶) were incubated for 24 h with 5 × 10⁵ stimulator cells (unless otherwise noted) in a final volume of 1 ml/well in 24-well plates. Supernatants were collected and tested for lymphokines using paired mAbs specific for cytokines IL-4, IL-5, IL-10, IFN-γ, or TNF-α (PharMingen, San Diego, CA). Biotinylated Ab were added and detected with avidin-peroxidase (Vector Laboratories, Burlingame, CA) plus 2.2-azino-di[3-ethyl-benzthiazoline sulfonate] substrate containing H₂O₂ (Kirkegaard & Perry, Gaithersburg, MD). The colorimetric reaction was read at 450 nm using an automatic microplate reader (Molecular Devices, Menlo Park, CA). The concentrations of the cytokines were calculated according to the standard curves of the appropriate recombinant cytokines.

Abs

The Abs used in this study are listed in Table I⇓.

Flow cytometry

Viable T cells were isolated by Ficoll-Hypaque gradient centrifugation (26) and then incubated for 30 min on ice in 50 μl of staining buffer (PBS, pH 7.4, containing 1% BSA and 0.1% sodium azide) containing 1 μg of fluorochrome-conjugated or unconjugated Ab. Fluorochrome-conjugated rat or hamster IgG was used as the isotype control. The cells were fixed in 0.5% paraformaldehyde after washing three times with staining buffer. In the case of CT-1, a secondary FITC-conjugated goat anti-mouse Ig (A,M,G) (Cappel, Durham, NC) was used. For Fas and Fas ligand (FasL) staining, CTL were stimulated with Con A (5 μg/ml) for 4 h and stained with hamster anti-mouse Fas or rabbit anti-mouse FasL Abs followed by either biotinylated anti-hamster IgG mixture or biotinylated anti-rabbit IgG, respectively. Biotinylated Abs were detected with streptavidin-PE.

The stained cells were examined on a FACScan Cytofluorimeter using LYSIS-II software (Becton Dickinson, San Jose, CA). Forward angle, light scatter was used to exclude dead and aggregated cells. The results are presented as fluorescence histograms with the relative number of cells on a linear scale plotted vs the relative fluorescence intensity on a log scale.

Depletion of CD4⁺ and phagocytic cells

B6 mice were injected i.p. with 200 μg of anti-CD4 (GK1.5) mAb or isotype controls on days −3, −2, and −1 (28). For depletion of phagocytic cells, mice were injected i.v. on day −2 and i.p. on day −1 with 0.5 mg silica (29, 30) (kindly provided by Dr. R. L. Hunter, University of Texas, Houston, TX) before immunization.

Serum Ab detection

Fourteen days postimmunization, mice were bled individually and the sera were tested for OVA- and INS-specific Abs by ELISA. Microtiter plates were coated with 100 μl of either OVA or BINS (10 μg/ml) in borate-buffered saline. Sera from individual mice were first diluted 1:100 followed by serial 2-fold dilutions. Ag-specific IgG was detected with alkaline phosphatase-conjugated goat anti-mouse IgG. Subsequent addition of the substrate ρ-nitrophenyl phosphate in buffer diethanolamine and MgCl₂ caused a colorimetric reaction, which was detected by an ELISA reader (Molecular Devices, Menlo Park, CA).

A minimum of three mice per group was used in each experiment in this manuscript. Representative results are shown for experiments that were repeated at least three times with similar results.

Phenotypic and functional differences between OVA-CTL and INS-CTL

OVA-CTL and INS-CTL cell lines were generated by priming B6 mice with 100 μg of either OVA or BINS in CFA. Splenocytes harvested from primed mice were cultured in vitro with irradiated stimulator cells (E.G7-OVA or EL4-INS) for 1 wk. The cultures were then restimulated weekly with the same irradiated target cells plus spleen cells and exogenous IL-2. Such CTL generated are specifically lytic toward their corresponding target cells. However, no CTL were detected without in vivo Ag priming (14, 15, 17), after priming with soluble protein, or protein in IFA (30)

Although OVA-specific and insulin-specific CD8⁺ lines are cytolytic, they secreted different cytokines upon stimulation. OVA-CTL produced significant amounts of IFN-γ and TNF-α but no IL-4, IL-5, or IL-10, which represents a Tc1 cytokine profile (Fig. 1⇓). By contrast, INS-CTL produced IL-5, IL-10, IFN-γ, and some IL-4 but no detectable TNF-α (Fig. 1⇓). It should be noted that independently derived OVA-CTL lines always produced more IFN-γ than did INS-CTL lines. Because the CTL clones generated from OVA-CTL and INS-CTL lines expressed the same phenotypes as the bulk lines (data not shown), we hereafter refer to OVA-CTL and INS-CTL as Tc1 and Tc0 cells, respectively. Neither OVA-CTL nor INS-CTL produced IL-2 upon activation (data not shown), which is not unusual for cells that have been restimulated with supplemental IL-2. OVA-CTL and INS-CTL represent two polarized subsets of CD8⁺ T cells, as their phenotypes have been stable over long-term culture.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

Lymphokine production by OVA-CTL and INS-CTL. OVA-CTL or INS-CTL were incubated for 24 h at 10⁶ cells per well with 5 × 10⁵ EG7-OVA or EL4-INS stimulator cells, respectively. Supernatants were then assayed for IL-4, IL-5, IL-10, IFN-γ, and TNF-α by ELISA using recombinant lymphokines as standards. Results are shown as cytokine concentrations ± SD.

Cell-surface markers were determined for INS-CTL and OVA-CTL (Fig. 2⇓). Both CTL are CD8⁺, CD3⁺, αβTCR⁺, and CD28⁺ but CTLA-4⁻. OVA-CTL but not INS-CTL express the CT-1 (T200), a carbohydrate epitope of CD45 expressed by activated CTL (25). Fas was expressed by both resting (not shown) and activated INS-CTL but not OVA-CTL, while both CTL expressed FasL upon activation. Thus, INS-CTL and OVA-CTL express similar, but not identical, phenotypic features.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

Cell-surface phenotype of OVA-CTL and INS-CTL cell lines. OVA-CTL and INS-CTL were stained with the indicated Abs. Isotype controls are the left most peak in each panel.

Requirements for priming CTL precursors

In some systems, priming of Ag-specific CD8⁺ T cells has been shown to require CD4 T cell help (31, 32, 33). Thus, we asked whether CD4⁺ T cells were required for the priming of Tc1 or Tc2 cells. Priming of OVA-CTL by OVA in CFA was previously shown to be independent of CD4 helper cells but critically dependent on phagocytic cells (14). To determine whether INS-CTL require CD4 cells for priming, B6 mice were depleted with anti-CD4 Ab before priming. Production of insulin-specific Ab was attenuated by anti-CD4 Ab, as expected (Fig. 3⇓A). By contrast, INS-CTL with similar lytic activities were primed in both the mice treated with GK1.5 and the isotype control (Fig. 3⇓B). Inactivation of phagocytic cells by injection of silica did not inhibit the Ab responses (Fig. 3⇓A), but completely abrogated the induction of INS-CTL in the same mice (Fig. 3⇓B). Thus, priming of INS-CTL, like priming of OVA-CTL (14), is relatively independent of CD4⁺ T cells but dependent on phagocytic cells. In addition, the cytokine profiles of INS-CTL generated from mice treated with either rIgG (Fig. 3⇓C) and anti-CD4 Ab (Fig. 3⇓D) were similar, suggesting that Th cells exerted little effect on the phenotype of CTL primed with exogenous Ags in CFA.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

Cellular requirements for Ab production and generation of INS-CTL in vivo. B6 mice were treated with either GK1.5, rat IgG1 as the isotype control, or silica and then immunized with BINS in CFA. Mice were bled after 14 days and tested for insulin-specific IgG by ELISA (A) and priming for CTL using either EL4-INS as target cells at the indicated E:T ratios (B). CTL generated from mice treated with either rIgG (C) or anti-CD4 Ab (D) were also tested for cytokine production with or without stimulation.

IL-4 plays a very important role in skewing both CD4⁺ (34) and CD8⁺ (35) T cells to produce type 2 cytokines. Although CD4⁺ T cells were not required for the activation of CD8⁺ T cells in mice primed with Ag in CFA, other cells might provide sufficient IL-4 to active CD8⁺ T cells in our system. To test whether endogenous IL-4 in B6 mice was responsible for priming Tc0 by BINS in CFA, CD8⁺ T cells from insulin immunized B6 or B6 IL-4^−/− mice (36) were compared. Fig. 4⇓ shows that INS-CTL generated from IL-4^−/− mice were as lytic as those from the normal B6 mice (Fig. 4⇓A). Even without IL-4 in the priming microenvironment, INS-CTL produced the other type 0 cytokines (IFN-γ, IL-5, and IL-10), albeit at levels lower than those from the wild-type B6 mice (Fig. 4⇓B). Thus, IL-4 was not required for the differentiation of naive CD8⁺ T cells into type 2 cytokine-producing cells in vivo.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

Priming of INS-CTL in IL-4^−/− mice. Wild-type B6 or IL-4^−/− mice were immunized with 100 μg of BINS in CFA. Cytolytic activities by B6 mice for EL4-INS (•) or E.G7-OVA (○) targets and IL-4^−/− mice for EL4-INS (▴) or E.G7-OVA (▵) were measured (A). Cytokines produced by B6 (▩) and IL-4^−/− mice (□) after stimulation with EL4-INS cells are shown (B).

The same immunization protocols and subsequent culturing conditions were used for generating INS-CTL and OVA-CTL, suggesting either that Ag itself might regulate the phenotypes of CTL or that the hormonal activity of insulin might be responsible for the induction of type 0 INS-CTL differentiation. To assess whether the Ag itself induced changes in the APC or cytokine microenvironment, B6 mice were immunized with a mixture of OVA and BINS in CFA. Mouse splenocytes were then divided in half and cultured separately with either E.G7-OVA or EL4-INS. We reasoned that if the Ags themselves were the determining factors that regulated CTL differentiation then INS-CTL and OVA-CTL generated from mixed Ag immunization should maintain their respective cytokine profiles. In contrast, if the OVA or BINS induced changes in the cytokine environments or the properties of APC, then both CTL would assume the same (Tc1 or Tc0) phenotype. Mixed Ags induced serum Abs against both OVA (Fig. 5⇓A) and BINS (Fig. 5⇓B) in B6 mice. The CTL that were cultured in the presence of either E.G7-OVA (Fig. 5⇓C) or EL4-INS (Fig. 5⇓D) were lytic toward their corresponding stimulator cells, although some low reactivity toward the other Ag was detected in these early cultures. More importantly, CTL stimulated with E.G7-OVA produced only IFN-γ (Fig. 5⇓E), whereas CTL stimulated with EL4-INS produced IL-5, IL-10, and some IFN-γ (Fig. 5⇓F). The cytokine profiles induced by priming with a mixture of OVA and BINS were similar to those produced by mice primed with a single Ag. Thus, we conclude that the generation of Tc1 or Tc0 is determined by the intrinsic property of the Ag rather than a nonspecific effect.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 5.

Priming of CTL with a mixture of OVA and BINS. B6 mice were primed with 100 μg of OVA and BINS mixed in CFA. Serum Abs against OVA (A) and BINS (B) were determined by ELISA. Cytolytic activities of OVA-CTL (C) or INS-CTL (D) were assayed using OVA- or INS-transfected cells as targets. The cytokines produced by OVA-CTL (E) and INS-CTL (F) are shown after activation by different stimulator cells.

There are several potential explanations for how Ag might regulate the phenotype of CD8⁺ T cells. Because OVA and BINS are two very different protein Ags, differences in their processing and in presentation by APC might influence the cytokines produced by these CTL. To eliminate any potential bias in the processing and presentation of the OVA and BINS Ags, B6 mice were immunized with the dominant antigenic epitopes of OVA or BINS. OVA_257–264 induced CTL that were lytic toward OVA_257–264-pulsed RMA-S cells but not RMA-S cells pulsed with irrelevant INS_12–21 peptide (Fig. 6⇓A). Similarly, BINS A_12–21 induced CTL that were lytic toward insulin peptide-pulsed RMA-S cells but not RMA-S cells pulsed with OVA peptide (Fig. 6⇓B). More importantly, OVA_257–264 induced OVA-CTL to produce only IFN-γ (Fig. 6⇓C), whereas INS-CTL primed by BINSA_12–21 produced IFN-γ, IL-5, and IL-10 upon activation (Fig. 6⇓D). These data demonstrated that the antigenic peptides were sufficient to prime Ag-specific CTL that produce either type 1 or type 0 cytokines. Thus, the phenotypes of the peptide-primed CTL were identical with those of the CTL primed by native protein Ags, which rules out a processing difference between OVA and BINS Ags.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 6.

Phenotype of CTL induced by priming with peptides. B6 mice were primed with 50 μg of OVA_257–264 or BINS_12–21 in CFA on day 0 and boosted with 50 μg of corresponding peptides in IFA on day 12. Splenocytes were harvested on day 19 and cultured with irradiated, peptide-pulsed RMA-S cells. Specific lytic activity of OVA-CTL (A) and INS-CTL (B) were examined after the third week of culture. Cytokine production for the CTL was determined after activation with the corresponding peptide-pulsed APC (C and D).