Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2013 Mar 15.
Published in final edited form as: J Immunol. 2012 Feb 6;188(6):2575–2582. doi: 10.4049/jimmunol.1102393

Selective dependence of H2-M3-restricted CD8 responses on IL-151

Jürgen R Müller 1, Thomas A Waldmann 1, Sigrid Dubois 1,2
PMCID: PMC3294130  NIHMSID: NIHMS351644  PMID: 22312130

Abstract

We studied whether CD8 T cell responses that are mediated by non-conventional MHC class Ib molecules are IL-15-dependent in mice. CD8+ T cell responses to listeria monocytogenes infection that are restricted by the MHC class Ib molecule H2-M3 decreased in the absence of IL-15 while other primary MHC class Ib- and MHC class Ia-restricted responses were IL-15-independent. This result was confirmed in MHC class Ia-deficient mice in which IL-15-deficiency also reduced H2-M3-restricted but not all CD8 T cell responses to listeria monocytogenes. IL-15-deficiency did not affect proliferation or survival of responding H2-M3-restricted CD8+ T cells but IL-15 was necessary to detect H2-M3-restricted CD8+ T cells in naïve mice. This suggests that these CD8+ T cells require IL-15 during development but become IL-15-independent after activation. IL-15 was necessary for the survival of most class Ib-restricted CD8+ T cells starting at the mature thymocyte stage in naïve mice but does not affect a distinct CD44low/CD122low sub-population. Together these data suggest that the nature of the selecting MHC class Ib molecule determines whether CD8+ T cells acquire IL-15-dependence during thymic development.

Introduction

Interleukin-15 belongs to the group of cytokines that require the common γ-chain for signaling (1). The cytokine is necessary for the survival of distinct lymphocyte populations that include all NK cells and subsets of CD8+ T cells (1–3). IL-15 displays an unusual mode of action: It is expressed by activated monocytic cells but stays membrane-anchored via its hetero-dimeric association with IL-15Rα (4). Signaling in trans through the two receptor chains CD122 and CD132 on responding lymphocytes causes their survival and proliferation. In addition, other cell types such as activated mastocytes and dendritic cells are affected by the presence of IL-15 (5, 6).

The effects of IL-15 signaling on CD8 responses have been studied in mice deficient in either IL-15 or IL-15Rα (7, 8). Primary CD8 responses showed only modest changes in the absence of IL-15. In contrast, IL-15 was necessary for the homeostasis of memory cells in that an accelerated decrease in the number of epitope-specific CD8+ T cells was observed during the memory stage (7, 8). Also, numerous publications exist that generally show supporting effects of IL-15 injections on CD8 responses (9). However, IL-15 injections also caused paradoxical increases of viral numbers in SIV-infected primates despite elevated numbers of CD8+ T cells that recognized viral epitopes (10, 11).

Infections of mice with listeria monocytogenes (LM) 3 have been widely used as a model to study CD8 responses (12). Two phases of the primary CD8 response are distinguished. An early expansion and activation stage usually measured by the number of CD8+ T cells restricted by the non-conventional MHC class Ib molecule H2-M3 is followed by class Ia-restricted responses. Responses restricted by class Ia and Ib molecules differ in several ways (13, 14) that include: A. Besides the earlier primary response peak, class Ib responses do not form a memory population and are nearly absent after rechallenges with LM. B. A higher number of class Ib than class Ia molecules is encoded in the murine genome. Class Ib molecules are not polymorphic, and similar class Ib-restricted responses can be measured across various murine inbred strains. C. Class Ib responses are promiscuous in that CD8+ T cells that are elicited with one peptide will respond to a range of alternative peptides if presented by the same class Ib molecule. In case of H2-M3, several peptides of bacterial origin have been identified whose main requirement is the presence of an N-terminal formylated methionine. D. Class Ib molecules can present epitopes with non-peptide moieties. E. Class Ib-restricted CD8+ T cells show an activated phenotype (CD44high/CD122high) prior to activation that resembles the subset of IL-15-dependent CD8+ T cells (15).

Despite the phenotypical characterization of IL-15-dependent CD8+ T cell sub-populations, no link between cytokine dependence and TCR reactivity with specific epitopes has been established. Here we study the hypothesis that subsets of CD8+ T cells recognizing distinct epitopes depend on IL-15 whereas CD8+ T cells recognizing other epitopes do not. We report that IL-15 is necessary for the survival of non-activated CD8+ T cells that recognize epitopes presented by some but not all MHC class Ib molecules.

Materials and Methods

Mouse strains

C57BL/6 wild-type, IL-15Rα−/− and β2M −/− mice were purchased from The Jackson Laboratory, and IL-15−/− and Kb−/−Db−/− mice were from Taconic. IL-15−/− mice were backcrossed with Kb−/−Db−/− mice and genotyped as described (2, 16, 17). All mice used were females between 8 and 12 weeks of age. IL-15 signaling was increased in vivo by a single i.p. injection of 2 µg IL-15 super-agonist (murine IL-15 bound to a chimeric fusion protein of murine IL-15Rα and human IgG1-Fc, R&D Systems) 24 h after infections. IL-15 signaling was inhibited by repeated injections of the CD122-neutralizing antibody Tm-β1 that blocks the action of trans-presented IL-15 (20 µg at days -7 and -1, R&D Systems). All the animals were housed and treated within published guidelines of humane animal care, and all procedures were approved and performed according to National Cancer Institute Animal Care and Use Committee–approved protocols for animal research.

Listeria monocytogenes

LM was handled as described (18). Briefly, LM was grown in brain-heart infusion broth. For infections of cells and mice, log phase cultures of LM expressing full-length ovalbumin or LCMV-gp protein, provided by Drs. E. Pamer, Memorial Sloan-Kettering Cancer Center and R. Ahmed, Emory University were used. To analyze anti-bacterial activities in various mouse strains that could affect antigen availability, wild-type, IL-15−/− and IL-15Rα−/− mice were injected i.v. with various bacterial numbers, and spleen lysates (0.25 % Triton-X 100, Sigma) were analyzed at day three by spreading samples on brain-heart infusion plates and incubating them overnight at 37 °C. We observed that a higher number of LM had to be injected into mice deficient in IL-15 or IL-15Rα to result in similar numbers of live bacteria in vivo when compared with wild-type mice (not shown). In contrast, treatments with Tm-β1 or IL-15 super-agonist (IL-15Rα-IgFc/IL-15) did not significantly alter the number of live bacteria (not shown). Based on these results we induced CD8 responses in wild-type mice with 105 cfu while IL-15−/− and IL-15Rα−/− mice received 5 × 105 cfu of LM. H2-M3 responses were generally induced with LM expressing recombinant LCMV-gp. To further equalize antigen availability, mice were i.p.-injected with 1 mg ampicillin 2 days after infection.

Vaccinia

The vaccinia virus Western Reserve strain was purchased from the American Type Culture Collection, propagated in embryonic fibroblasts from C57BL/6 mice, and titered in BSC-1 cells as described (19). Vaccinia virus was isolated by three freeze-thaw cycles of cell cultures 3 days after infection. The culture supernatant was cleared of debris by centrifugation at 2000 × g for 30 min at 4 °C. Similar to LM infections, we determined the numbers of vaccinia virus in spleens 5 days after infections but found no significant difference between mouse strains. Mice were infected i.p. with 106 pfu of vaccinia virus in PBS. After 7 days, the mice were sacrificed and the splenocytes were used for ex vivo IFN-γ production analyses (20).

Reagents and cytometry

A list of antibodies and tetramers used is provided in Table S1. For cytometry analyses we used cells from spleens and lymph nodes after mechanical disruptions. Lymph node analyses were done on pooled cervical, mesenteric and inguinal lymph node cells. Cells were blocked with a mixture of rat IgG1, IgG2a, IgG2b, mouse IgG1 and hamster IgG1 for 15 min at room temperature that was followed by a 30-min incubation on ice with the specific antibody. For biotinylated antibodies, an additional 15-min incubation on ice was done with streptavidin-PE-CY5.5 (Ebioscience). For tetramer stainings (Table S1), cells were incubated with tetramer for 30 min at 37°C before blocking. BrDU stains were done 9 h after the i.p.-injections of 1 mg BrDU (Sigma) using the BrDU Flow Kit (BD Biosciences). Data analysis was performed using FlowJo software.

Peptides and cytokine production

All peptides had been synthesized by JPT, Berlin, Germany. Peptide sequences to detect H2-M3-restricted and ovalbumin-specific responses are given in Table S1, and KAVYNFATC was used for responses against LCMV-gp. Cytokine production was induced by incubating 106 splenocytes for 6 h in 200 µl RPMI containing 10 % FBS, 55 mM 2-mercaptoethanol, 100 pM human IL-2 (Peprotech) and 500 nM of the respective peptide with brefeldin A present during the last 5 hours. The BD Cytofix/Cytoperm Kit was used to detect intracellular cytokines according to the manufacturer’s instructions (BD Biosciences).

Lysis Assay

To detect CD8+ T cell-mediated cytotoxicity, EL-4 target cells were propagated in RPMI-10 % FBS and were labeled with 0.1 mCi Indium-111 Oxine (GE Healthcare, 30 min, 37°C in 100 % FBS), washed in PBS and incubated in the presence or absence of 500 nM peptide (2 h, 37°C in growth medium). Negatively selected CD8+ T cells (CD8a+ T cell Isolation Kit II, Miltenyi) from spleens were incubated with labeled EL-4 target cells for 20 h at various effector:target ratios, and the radioactivity in the liquid phase was measured. Specific lysis was determined by using the formula: % lysis = 100 * [(mean experimental cpm - mean spontaneous cpm)/(mean maximum cpm - mean spontaneous cpm)]. The maximum release value was determined from target cells treated with 1% (v/v) Triton X-100.

Dendritic cells

Bone marrow-derived dendritic cells (BMDCs) were prepared by growing C57BL/6 bone marrows in RPMI supplemented with 10 % FBS, 55 mM 2-mercaptoethanol and 40 ng/ml recombinant murine GM-CSF (Peprotech) for 5 days, and the DC-like line DC2.4 was propagated in RPMI-10 % FBS. LM-infected DCs were generated by a 6-h co-incubation with live bacteria at MOIs of 0.25 for BMDCs and 0.5 for DC2.4. This was followed by an overnight exposure to 20 ng/ml murine IFN-γ (Peprotech) and a combination of 100 U/ml Penicillin/ 100 µg/ml Streptomycin/ 50 µg/ml Gentamycin in RPMI-10 % FBS after which no live bacteria could be detected. For peptide-labeling, BMDCs were matured overnight in RPMI containing 10 % FBS, 50 ng/ml LPS (E. coli 055:B5, Sigma) and 20 ng/ml IFN-γ that was followed by a 2-h incubation with 1 µM of the respective peptide at 2.5 × 106 cells/ml RPMI at 4°C.

BMDCs and DC2.4 were used to immunize mice and for ex vivo cytokine inductions. For immunizations, peptide-loaded BMDCs were washed and injected i.p. in 200 µl PBS. To analyze the survival of H2-M3-restricted CD8+ T cells in immunized mice, Ly5.2+ recipient mice were immunized with H2-M3 peptide-loaded BMDCs that was followed 4 h later by i.p. injections of negatively sorted splenic CD8+ T cells from Ly5.1+ donor mice that had been immunized with H2-M3 peptide-loaded BMDCs five days prior. The number of surviving H2-M3-restricted donor CD8+ T cells was determined 24 h later. For survival analyses of activated H2-M3-restricted CD8+ T cells in naive mice, we immunized Kb−/−Db−/− mice with H2-M3 peptide-loaded BMDCs. Splenic CD8+ T cells were negatively sorted 4 days later, pooled, CFSE-labeled and i.p. injected into naïve Kb−/−Db−/− and Kb−/−Db−/− IL-15−/− mice. Donor CD8+ T cells were identified and enumerated 24 h later using the CFSE label and H2-M3 tetramer binding.

For ex vivo cytokine inductions, 5 × 105 BMDCs were washed and co-incubated with 106 splenocytes in RPMI containing 10 % FBS and 100 pM human IL-2 for 6 h with Brefeldin A present during the last 5 h.

Bone marrow chimeras

Bone marrow chimeras were prepared by γ-irradiating (1000 rad, 137Cs source) 6–week old wild-type or β2M−/− recipient mice (Thy1.1-) followed by i.v. injections of 107 T cell-depleted bone marrow cells from wild-type or IL-15−/− donors (Thy1.1+). T cell depletion was done with Thy1.1 beads (Miltenyi). Chimeras were analyzed 8–15 weeks later.

Statistical analyses

Data are represented as means ± SD. Significance levels were determined using the Student’s t test.

Results

H2-M3 responses are IL-15-dependent

We studied CD8+ T cell responses to infection with LM in mice. Two peaks of CD8+ T cell responses are known to occur in this model (12): The early peak involves MHC class Ib-restricted CD8+ T cells exemplified by responses that are mediated by H2-M3 while a later peak comprises mostly MHC lass Ia-restricted responses. We analyzed the amplitudes of the non-conventional, H2-M3-restricted CD8 responses by tetramer staining, IFN-γ/TNF-α production induced by an H2-M3-specific peptide or by H2-M3 peptide-directed cytotoxicity (Fig. 1A–C). In all cases, the absence of IL-15 signaling in IL-15- or IL-15Rα-deficient mice was associated with decreased percentages and total numbers of H2-M3-specific CD8+ T cells (Fig. 1A–C). Similarly, reducing the number of IL-15-dependent CD8+ T cells by pre-treating mice with the CD122-neutralizing antibody Tm-β1 that blocks the activity of trans-presented IL-15, decreased the number of responding H2-M3-restricted CD8+ T cells. In contrast, a single injection of an IL-15 super-agonist [comprising murine IL-15 and a chimeric fusion protein of murine IL-15Rα and human IgG1-Fc (21, 22)] 24 hours post infection had the opposite effect (Fig. 1A, B). IL-15 super-agonist injections was also associated with the presence of H2-M3-reactive CD8+ T cells 8 days post infection that was not observed in untreated mice (not shown). In addition to lower amplitudes of H2-M3 responses, the absence of IL-15 also appeared to affect the function of H2-M3-restricted CD8+ T cells since little H2-M3 peptide-directed cytotoxic activity was detected with CD8+ T cells from immunized IL-15−/− mice (Fig. 1C). In contrast to the H2-M3-restricted responses, IL-15 did not have clear positive effects on late MHC class Ia-mediated CD8 responses to LM infection (Fig. S1), and MHC class Ia-mediated CD8 responses to vaccinia virus infection were similar irrespective of IL-15-proficiency of infected mice (Fig. S2). Together these data show that the amplitudes of H2-M3 responses depend on IL-15.

Figure 1. IL-15-dependence of H2-M3-restricted CD8 responses to infection with LM.

Figure 1

Open in a new tab

Infected mice were analyzed at day 5 post infection for splenic H2-M3-restricted CD8 cells. (A) Representative panels depict the percentage of CD3+/CD8+ T cells that bind H2-M3 tetramer (upper) or express IFN-γ/TNF-α after exposure to H2-M3 peptide (lower) in LM-infected wild-type, IL-15−/− or IL-15Rα−/− mice or in wild-type mice after treatments with Tm-β1 or with IL-15 super-agonist. Results are representative of 5 mice each. (B) The average percentage of H2-M3 tetramer-binding among all splenic CD8+ T cells (left) and average numbers of H2-M3 tetramer-binding splenic CD8+ T cells (right) + SD, n=5 for each condition, is shown for the various mice. *, p<0.05; **, p<0.01 compared with infected, untreated wild-type mice. (C) H2-M3 peptide-directed cytotoxicity. Splenic CD8+ T cells from d5-infected wild-type and IL-15−/− mice were negatively sorted and co-incubated for 20 h with H2-M3 peptide-coated EL-4 target cells at various effector-to-target ratios (E:T). Solid diamonds depict uncoated EL-4 cells co-incubated with CD8+ T cells from infected wild-type mice, n=3 for each point. Despite the presence of approximately 10 % of H2-M3 tetramer-binding CD8+ T cells in IL-15−/− when compared with wild-type mice, their cytotoxic activity did not reach levels of wild-type CD8+ T cells if E:T ratios were increased approximately tenfold (compare the highest E:T ratio in IL-15−/− samples with the lowest in wild-type samples).

Some MHC class Ib-restricted CD8+ T cell responses are IL-15-independent

Since H2-M3 responses appeared to be affected in mice with reduced or absent IL-15 signaling, we tested whether this IL-15 effect applied to all MHC class Ib-restricted CD8 responses. For this we analyzed splenic CD8+ T cells 5 days after LM infection by 6-hour in vitro re-stimulations with BMDCs that differed in their ability to present epitopes via MHC class I molecules and that had been infected with the same bacteria. Re-stimulations with wild-type versus class Ia-deficient (Kb−/−Db−/−) dendritic cells should permit estimations of the extent of class Ia- versus class Ib-restricted responses. To verify the use of these cells, we initially analyzed responses to LM 5 and 8 days post infection in wild-type mice (Fig. S3). To include the analyses of specific MHC class Ia responses, mice were immunized with recombinant LM strains that expressed antigenic ovalbumin or LCMV-gp protein. While IFN-γ single-producing CD8+ T cells had high backgrounds, the induction of IFN-γ/TNF-α double-producers appeared to strictly depend on the presence of bacterial antigen. We therefore used the number of IFN-γ/TNF-α double-producing CD8+ T cells as a measure of the CD8 response amplitudes. Comparing IFN-γ/TNF-α production that had been induced by LM-infected wild-type or class Ia−/− BMDCs confirmed that the majority of CD8+ T cells at day 5 responded to class Ib-presented epitopes while less class Ib responses were found at day 8. We also noted that the majority of responding CD8+ T cells at day 8 were recognizing recombinant LCMV-gp or ovalbumin proteins, respectively. Together these data appear to justify the use of LM-infected BMDCs for analyses.

We then used the same BMDCs to compare early CD8+ T cell responses to LM between wild-type and IL-15-deficient mice. Representative panels in Fig. 2A show that IFN-γ/TNF-α was induced in a portion of CD8+ T cells from both infected wild-type and IL-15-deficient mice when co-incubated with LM-infected BMDCs that lacked class Ia molecules. These inductions appeared to be specific since few IFN-γ/TNF-α double-producing CD8+ T cells were detected if re-stimulations were done with infected BMDCs that lacked MHC class I entirely (β2M−/−) or with uninfected wild-type BMDCs. While the percentages of CD8+ T cells responding to MHC class Ia or class Ib molecules were similar between wild-type and IL-15-deficient mice, their total number differed due to lower total CD8+ T cell numbers in IL-15−/− mice (Fig. 2B). Thus, while IL-15 signaling appears to be necessary for the expansion of H2-M3-restricted CD8+ T cells in response to LM infection, the cytokine may not affect all class Ib-restricted CD8 responses.

Figure 2. Determination of comprehensive CD8 responses in spleens 5 days after LM infections.

Figure 2

Open in a new tab

BMDCs from wild-type, MHC class Ia−/− (Kb−/−Db−/−) or MHC class I−/− (β2M−/−) mice that had been infected with LM were used to analyze the total or MHC class Ib-restricted CD8 responses in splenic CD3+/CD8+ T cells from wild-type or IL-15−/− mice. Responses were measured by the percentage of CD8+ T cells that expressed IFN-γ/TNF-α 6 h after co-incubations. (A) Representative panels. The right panels show control CD8+ T cells after 6-h co-incubations with uninfected wild-type BMDCs. (B) The average percentages (left) and numbers (right) of splenic CD3+/CD8+ T cells from wild-type or IL-15−/− mice that respond to LM-infected wild-type or MHC class Ia−/− BMDCs is shown. n, number of mice analyzed for each value.

Survival and proliferation of activated H2-M3-restricted CD8+ T cells are IL-15-independent

IL-15 could affect H2-M3-restricted CD8+ T cell responses by acting directly on CD8+ T cells or by influencing proximal pathways such as antigen uptake and presentation. To investigate we immunized mice with wild-type BMDCs that had been coated with H2-M3 peptide, thus circumventing potential IL-15 effects on antigen presentation. Fig. 3A shows that this immunization induced clear H2-M3 responses in IL-15-deficient mice but its amplitude was still reduced when compared with wild-type mice as measured both by the percentage of H2-M3 tetramer-binding among all CD8+ T cells and by their total numbers (Fig. 3A, left panels and left graph). This supports the concept that IL-15 may influence H2-M3-restricted CD8+ T cells directly.

Figure 3. Expansion of H2-M3-restricted CD8+ T cells by injections of H2-M3 peptide-coated BMDCs.

Figure 3

Open in a new tab

Wild-type and IL-15−/− mice were injected with one million peptide-coated DCs each at days 0 and 2 and were analyzed at day 5. (A). Representative panels and graphs show that this treatment induced expansions of H2-M3-restricted CD3+/CD8+ T cells in both strains of mice to different amplitudes (left panels and left graph) but that their growth rate as measured by BrDU incorporation was similar (right panels and right graph). (B) The survival of H2-M3-restricted CD3+/CD8+ T cells is IL-15-independent after activation. Total splenic CD3+/CD8+ T cell populations from wild-type mice at day 5 after treatments with H2-M3-coated BMDCs (left panel) were sorted and transferred into wild-type or IL-15−/− recipient mice that had received peptide-coated wild-type or IL-15−/− BMDCs, respectively four hours prior. Analysis 24 h later revealed that among the transferred Ly5.1+ donor cells, only H2-M3-restricted CD3+/CD8+ T cells had survived (right panels). The number of H2-M3-restricted CD3+/CD8+ donor T cells that was recovered from recipient mice did not depend on the IL-15-proficiency of the recipients (graph). n, number of mice analyzed for each value.

IL-15 may act on CD8+ T cells before activation by affecting proliferation and survival of naïve cells. Alternatively, post-activation effects of the cytokine could explain differences in response amplitudes. We addressed the latter by measuring post-activation proliferation rates after immunizations with H2-M3 peptide-coated BMDCs since clear H2-M3 tetramer-binding CD8+ T cell populations were detected after this treatment in both wild-type and in IL-15-deficient mice. Fig. 3A (right panels and right graph) shows that post-activation proliferation rates of H2-M3 tetramer-binding CD8+ T cells as measured by BrDU incorporation were similar in both wild-type and in IL-15-deficient mice.

To measure post-activation survival we isolated total splenic CD8+ T cell populations from donor mice 5 days after BMDC-peptide immunizations and injected them into congenic recipient mice (Ly5. 1−) that had been immunized 4 hours prior with BMDC-peptide. To assure a complete lack of IL-15, immunizations of IL-15−/− recipients were done with IL-15−/− BMDCs whereas wild-type BMDCs were used for wild-type recipients. Analyses 24 h later (Fig. 3B) revealed that although both H2-M3 tetramer-positive and -negative CD8+ T cells had been transferred, the only surviving donor CD8+ T cells (Ly5.1+) recognized H2-M3 tetramer. The number of surviving donor cells in spleens did not depend on the presence of IL-15. Together these data suggest that the proliferation and survival of H2-M3-restricted CD8+ T cells are IL-15-independent after activation.

Most but not all naïve MHC class Ib-restricted CD8+ T cells are IL-15-dependent

An alternative approach to assess MHC class Ib-restricted CD8+ T cell responses includes the use of mice that are deficient in MHC class Ia (Kb−/−Db−/−). All CD8+ T cell responses in these mice are by definition restricted by MHC class Ib molecules. In naïve MHC class Ia-deficient mice, an additional lack of IL-15 reduced the number of CD8+ single-positive thymocytes predominantly at the mature CD24low stage (Fig. 4A). IL-15-deficiency also reduced the number of peripheral CD8+ T cells in spleens and lymph nodes (Fig. 4B). Additional analyses revealed that the remaining IL-15-independent and MHC class Ib-restricted CD8+ T cell population was phenotypically distinct by expressing low levels of CD122, CD44, Ly6C, CD11a and high levels of Integrinβ7 (Fig. 4C).

Figure 4. Effect of IL-15 on CD8+ T cell populations in naïve MHC class Ia-deficient mice.

Figure 4

Open in a new tab

(A) Analyses of thymi from 4-week old mice revealed that on an MHC class Ia−/− (Kb−/−Db−/−) background, IL-15-deficiency caused a reduction of CD8+ single-positive thymocytes (left panels and graph) mainly at the mature CD24low stage (right panels, numbers depict percentages of CD24low, CD24int and CD24high populations). (B) IL-15-deficiency similarly affected the numbers of CD3+/CD8+ T cells in spleens (top) and lymph nodes (bottom) in 6-week old naïve mice. (C) The IL-15-independent CD3+/CD8+ T cell population in MHC class Ia−/− mice is phenotypically distinct based on the expressions of the markers shown. (D). Tetramer analyses revealed that IL-15-deficiency severely decreased both the percentages and numbers of H2-M3-restricted CD3+/CD8+ T cells in spleens (top) and lymph nodes (bottom) from naïve MHC class Ia−/− mice (left four panels and graphs). IL-15 deficiency also affected IFN-γ production that was induced in naïve splenic CD8+ T cells by 6-h exposures to H2-M3 peptide (right panels). n, number of mice analyzed for each value.

If IL-15 affects the survival of H2-M3-restricted CD8+ T cells before activation, their numbers should be different in naïve mice. Fig. 4D shows that MHC class Ia-deficient mice harbor a small population of CD8+ T cells that bind H2-M3 tetramer in both lymph nodes and spleens. Ex vivo exposures of these naïve CD8+ T cells to H2-M3 peptide induced rapid IFN-γ production (Fig. 4D, shown for spleen). The percentages of CD8+ T cells that bound H2-M3 tetramer or generated IFN-γ in response to H2-M3 peptide were significantly reduced by an additional lack of IL-15. Thus, the survival of most naive MHC class Ib- including H2-M3-restricted peripheral CD8+ T cells depends on IL-15.

Mice deficient in MHC class Ia and IL-15 mount class Ib-restricted CD8 responses

We next investigated the IL-15-dependence of CD8+ T cell responses to LM infection in MHC class Ia-deficient mice. If all MHC class Ib-restricted responses would depend on IL-15, little CD8+ T cell activation should occur in MHC class Ia/IL-15-deficient mice. Fig. 5A shows that similar percentages of CD8+ T cells had acquired activated phenotypes (CD25+, CD69+) 5 days after LM infection irrespective of IL-15-proficiency indicating the existence of IL-15-independent MHC class Ib-restricted CD8+ T cells that respond to LM infection. In contrast, H2-M3 tetramer-binding activity was almost absent in infected MHC class Ia/IL-15-deficient mice. CD8+ T cells from MHC class Ia−/− mice could be re-stimulated by the DC line DC2.4 that had been infected with LM regardless of IL-15-proficiency of immunized mice (Fig. 5B) further supporting the presence of LM-responding, IL-15-independent and MHC class Ib-restricted CD8+ T cells. IL-15-deficiency also reduced the percentage and number of H2-M3 tetramer-binding CD8+ T cells after immunizations with H2-M3 peptide-loaded BMDCs of MHC class Ia−/− mice (Fig. 5C, left panels and left graph). Post-activation proliferation rates were not affected by IL-15 (right panels and right graph). Analyses of the survival of activated H2-M3-restricted CD8+ T cells in naïve mice revealed a strong decline in that only approximately 1 % of transferred H2-M3-restricted CD8+ T cells were recovered in spleens 24 h and none 48 h after transfers (Fig. 5D and not shown). H2-M3-restricted CD8+ T cells had a survival advantage over other CD8+ T cells since the transferred samples contained only approximately 30% H2-M3 tetramer-binding CD8+ T cells while all surviving donor cells bound the tetramer (not shown). The survival of activated H2-M3-restricted CD8+ T cells in naïve mice did not depend on the IL-15 proficiency of the recipient hosts (Fig. 5D). Together these data show that only a portion of MHC class Ib-restricted CD8+ T cell responses depends on IL-15.

Figure 5. IL-15-dependence of CD8 responses to listeria infection in MHC class Ia-deficient mice.

Figure 5

Open in a new tab

(A) Analyses five days post infection show activations of similar percentages of splenic CD3+/CD8+ T cells as measured by the expression of CD25 and CD69 regardless of IL-15-proficiency (left, left panels represent cells from naïve mice). Panels and graph on the right show that IL-15-deficiency caused a decrease of H2-M3-tetramer-binding among CD3+/CD8+ T cells. (B) An IL-15-deficiency of infected hosts had little effect on the percentages of CD3+/CD8+ T cells from mice 5 days post infection that expressed IFN-γ/TNF-α after re-activations with LM-infected DC2.4 cells for six hours in vitro (right panels and graph, left panels depict re-activations with uninfected DC2.4 cells. (C) Injections with peptide-coated BMDCs as described in Fig. 3 caused expansions of H2-M3-restricted CD3+/CD8+ T cells to amplitudes that were lower in IL-15−/− mice (left panels and left graph). The proliferation rates of H2-M3-restricted CD3+/CD8+ T cells analyzed by BrDU incorporation was not affected by the IL-15-proficiency of the infected hosts (right panels and right graph). (D) Transfers of CFSE-labeled CD8+ T cells from H2-M3-peptide/BMDC-immunized Kb−/−Db−/− mice into naïve Kb−/−Db−/− and Kb−/−Db−/− IL-15−/− mice revealed similar survival rates regardless of IL-15 proficiency. n, number of mice analyzed for each value.

Thymic selection on hematopoietic cells generates both IL-15-dependent and -independent MHC class Ib-restricted CD8+ T cells

As described above, one mechanism that appears to be affected in IL-15-deficient mice is the frequency of naïve H2-M3-restricted CD8+ T cells. This possibility would suggest that thymic selection is able to distinguish thymocytes based on their TCR affinity to peptides presented by H2-M3 or other class I molecules. This could be achieved by the nature of the accessory cell type responsible for selection in that H2-M3-binding thymocytes may be selected by the interaction with only a subset of antigen-presenting cells that may also confer IL-15-dependence. Using bone marrow-chimeric mice, Urdahl et al. (15) had shown that MHC class Ib- including H2-M3-restricted CD8+ T cells are thymically selected on hematopoietic cells and do not require MHC class I expression on epithelial cells. We investigated whether a selection on antigen-presenting cells of hematopoietic origin is associated with IL-15-dependence. Thymic hematopoietic cell selection was forced by transferring wild-type and IL-15−/− bone marrows into irradiated MHC class I−/− (β2M−/−) mice. It has been well established that IL-15-dependent CD8+ T cells require the cytokine being trans-presented by monocytes/dendritic cells (23, 24). Therefore, if IL-15-dependent CD8+ T cells were generally selected on hematopoietic cells, no CD8+ T cells should survive under the condition of hematopoietic IL-15-deficiency. Fig. 6 shows that most hematopoietic cell-selected CD8+ T cells depended on IL-15. Similar to the MHC class Ia/IL-15+ mice however, a small CD122low population did not depend on IL-15. These data suggest that thymic hematopoietic cell selection causes IL-15-dependence in most CD8+ T cells. However, an IL-15-independent subpopulation of CD8+ T cells exists that undergoes hematopoietic cell selection and that correlates with the IL-15-independent MHC class Ib-restricted CD8+ T cell subpopulation that participates in the response to LM infection.

Figure 6. IL-15-dependence of MHC-class Ib-restricted CD3+/CD8+ T cells after thymic selection on hematopoietic cells.

Figure 6

Open in a new tab

Bone marrows from wild-type or IL-15−/− mice were transferred into irradiated wild-type or β2M−/− mice and analyzed 12 weeks later. The effect of IL-15-deficiency on CD3+/CD8+ T cell numbers in spleens and lymph nodes is stronger when thymic CD8 selection was limited to cells of hematopoietic origin (compare top and bottom halves). However, thymic hematopoietic cell selection generated a small subset of IL-15-independent CD3+/CD8+ T cells that expressed low amounts of CD122 (bottom panels). n, number of mice analyzed for each value.

Discussion

IL-15 affects the survival and activation of CD8+ T cells in multiple ways (25–27). Here we describe data suggesting a new effect of IL-15 in that the cytokine is required for the development of CD8+ T cells that recognize epitopes presented by some MHC class Ib molecules including H2-M3, but also that IL-15 is not required for other epitope-specific primary responses that include some mediated by both MHC class Ia and class Ib. These conclusions are based on data showing that IL-15-deficiency caused significant decreases in the number of H2-M3-specific CD8+ T cells and affected their cytolytic activity after LM infections while class Ia- and some class Ib-restricted responses were less affected. In addition, BMDC-mediated immunizations also resulted in smaller amplitudes of H2-M3 responses under IL-15-deficiency.

An IL-15-dependence of MHC class Ib-restricted CD8+ T cells had been suggested by previous publications (15, 16, 28). Thymic hematopoietic cell selection results in mostly CD122high CD8+ T cells that respond to MHC class Ib-presented epitopes (15), and IL-15-deficiency leads to a loss of CD122high CD8+ T cells (16). A recent publication also describes IL-15-dependent CD8+ T cells with suppressor function that are restricted by another MHC class Ib molecule, Qa-1 (28). We confirmed the connection between MHC class Ib-restriction and IL-15-dependence for most CD8+ T cells. Surprisingly however, our data also identify a previously unrecognized MHC class Ib-restricted subpopulation of CD8+ T cells that is selected on thymic cells of hematopoietic origin, that participates in responses to LM, but that remains IL-15-independent. The existence of such a population is supported by the following findings: A. CD8+ T cells from LM-infected IL-15−/− mice could be reactivated with LM-infected class Ia-deficient BMDCs. B. A phenotypically distinct CD8+ T cell population exists in naïve mice that lack both class Ia and IL-15. C. CD8+ T cell activation was observed in mice deficient in both class Ia and IL-15. D. MHC class Ib-restricted CD8+ T cells are selected on hematopoietic cells in the thymus. A portion of these CD8+ T cells remains IL-15-independent.

The results suggest that thymic selection is able to recognize TCRs that bind to H2-M3-presented peptides and affect the cytokine dependence of selected thymocytes. One way to achieve this would be if H2-M3-mediated selection in the thymus were done by certain cell types only. As mentioned above, only hematopoietic cells are necessary for the thymic selection of H2-M3-restricted CD8+ T cells (15). In addition, CD8+ T cells that are present in mice deficient for the kinase ITK are IL-15-dependent and are also selected on hematopoietic cells (29–31). Our data show that the majority of CD8+ T cells in mice that do not express MHC class I on non-hematopoietic cells are IL-15-dependent. However similar to mice deficient in both class Ia and IL-15, a small CD8+ T cell population of unknown function with a non-activated phenotype persists that appears to be class Ib-restricted, IL-15-independent and does not require MHC class I expression on non-hematopoietic cells possibly caused by the existence of distinct cell types within the thymic hematopoietic antigen-presenting cells. Nevertheless, our data suggest that the interaction of H2-M3 restriction-committed thymocytes with hematopoietic cells during selection is associated with IL-15-dependence.

The extent of CD8 responses to bacterial infections is affected by multiple mechanisms that include activities proximal to antigen presentation such as mouse strain-dependent antibacterial activities, survival and activation of antigen-presenting cells, among others (5, 6). These effects are difficult to control and may in part explain variations that we observed in the effect of IL-15 on distinct class Ia-restricted responses. On the other hand, BMDC-mediated immunizations should allow better conclusions about CD8+ T cell dynamics since no variation proximal of antigen presentation should exist. In addition, CD4 cell help should also not be a factor when responses are induced with BMDCs whose only foreign antigen is an MHC class I-binding peptide. Therefore, the differences in the extent of H2-M3 responses that we observed between wild-type and IL-15−/− mice after immunizations with peptide-coated BMDCs suggest that IL-15 has direct effects on H2-M3-restricted CD8+ T cells. The finding that the number of H2-M3-restricted CD8+ T cells is affected by IL-15-deficiency in naïve mice points to a defect during their development. In contrast, neither survival nor proliferation of H2-M3-restricted CD8+ T cells are compromised by the lack of IL-15 during the peak of their response suggesting that TCR engagement or other co-stimulatory signals can overcome cytokine dependence for these cells. A similar finding was reported for NK cells that expand significantly during a viral infection in IL-15−/− mice despite their absolute IL-15-dependence in uninfected mice (32).

We observed that the IL-15 effect on H2-M3 responses induced by infections with LM exceeded the differences observed after BMDC immunizations. This suggests that mechanisms upstream of antigen presentation contribute to the IL-15 effect on H2-M3 responses. A potential cause is IL-15-mediated changes in antigen-presenting cells (6). This would necessitate that different cells present H2-M3- and class Ia-restricted peptides since IL-15 effects were less evident on class Ia-restricted responses. Another cause could be activities by NK cells on CD8 responses (33) that are missing in IL-15−/− mice.

It is believed that thymic T cell development starts with the entry of a common pre-thymocyte that randomly re-arranges its TCR-encoding genes. A TCR affinity-based separation into CD4+ and CD8+ single-positive thymocytes is well described (34). The data presented here suggest a second TCR affinity-based mechanism in that thymic development induces IL-15-dependence in thymocytes that recognize epitopes restricted by the unconventional MHC class Ib molecule H2-M3 and possibly by others.

Supplementary Material

1
2
3
4

Acknowledgments

We would like to thank Drs. E. Pamer, R. Ahmed and J. Yewdell, NIAID, NIH for providing materials. We are indebted to the group of Dr. Pamer for invaluable advice.

Footnotes

The authors have no conflicting financial interests.

1

This research is supported by the intramural research program of the National Cancer Institute, NIH.

3

Abbreviations used: LM, Listeria monocytogenes; BMDC, bone marrow-derived dendritic cells

References

  • 1.Waldmann TA. IL-15 in the life and death of lymphocytes: immunotherapeutic implications. Trends Mol Med. 2003;9:517–521. doi: 10.1016/j.molmed.2003.10.005. [DOI] [PubMed] [Google Scholar]
  • 2.Lodolce JP, Boone DL, Chai S, Swain RE, Dassopoulos T, Trettin S, Ma A. IL-15 receptor maintains lymphoid homeostasis by supporting lymphocyte homing and proliferation. Immunity. 1998;9:669–676. doi: 10.1016/s1074-7613(00)80664-0. [DOI] [PubMed] [Google Scholar]
  • 3.Becknell B, Caligiuri MA. Interleukin-2, interleukin-15, and their roles in human natural killer cells. Advances in immunology. 2005;86:209–239. doi: 10.1016/S0065-2776(04)86006-1. [DOI] [PubMed] [Google Scholar]
  • 4.Dubois S, Mariner J, Waldmann TA, Tagaya Y. IL-15Ralpha recycles and presents IL-15 In trans to neighboring cells. Immunity. 2002;17:537–547. doi: 10.1016/s1074-7613(02)00429-6. [DOI] [PubMed] [Google Scholar]
  • 5.Orinska Z, Maurer M, Mirghomizadeh F, Bulanova E, Metz M, Nashkevich N, Schiemann F, Schulmistrat J, Budagian V, Giron-Michel J, Brandt E, Paus R, Bulfone-Paus S. IL-15 constrains mast cell-dependent antibacterial defenses by suppressing chymase activities. Nat Med. 2007;13:927–934. doi: 10.1038/nm1615. [DOI] [PubMed] [Google Scholar]
  • 6.Dubois SP, Waldmann TA, Muller JR. Survival adjustment of mature dendritic cells by IL-15. Proc Natl Acad Sci U S A. 2005;102:8662–8667. doi: 10.1073/pnas.0503360102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Becker TC, Wherry EJ, Boone D, Murali-Krishna K, Antia R, Ma A, Ahmed R. Interleukin 15 is required for proliferative renewal of virus-specific memory CD8 T cells. J Exp Med. 2002;195:1541–1548. doi: 10.1084/jem.20020369. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Schluns KS, Williams K, Ma A, Zheng XX, Lefrancois L. Cutting edge: requirement for IL-15 in the generation of primary and memory antigen-specific CD8 T cells. J Immunol. 2002;168:4827–4831. doi: 10.4049/jimmunol.168.10.4827. [DOI] [PubMed] [Google Scholar]
  • 9.Diab A, Cohen AD, Alpdogan O, Perales MA. IL-15: targeting CD8+ T cells for immunotherapy. Cytotherapy. 2005;7:23–35. doi: 10.1080/14653240510018037. [DOI] [PubMed] [Google Scholar]
  • 10.Hryniewicz A, Price DA, Moniuszko M, Boasso A, Edghill-Spano Y, West SM, Venzon D, Vaccari M, Tsai WP, Tryniszewska E, Nacsa J, Villinger F, Ansari AA, Trindade CJ, Morre M, Brooks D, Arlen P, Brown HJ, Kitchen CM, Zack JA, Douek DC, Shearer GM, Lewis MG, Koup RA, Franchini G. Interleukin-15 but not interleukin-7 abrogates vaccine-induced decrease in virus level in simian immunodeficiency virus mac251-infected macaques. J Immunol. 2007;178:3492–3504. doi: 10.4049/jimmunol.178.6.3492. [DOI] [PubMed] [Google Scholar]
  • 11.Mueller YM, Do DH, Altork SR, Artlett CM, Gracely EJ, Katsetos CD, Legido A, Villinger F, Altman JD, Brown CR, Lewis MG, Katsikis PD. IL-15 treatment during acute simian immunodeficiency virus (SIV) infection increases viral set point and accelerates disease progression despite the induction of stronger SIV-specific CD8+ T cell responses. J Immunol. 2008;180:350–360. doi: 10.4049/jimmunol.180.1.350. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Pamer EG. Immune responses to Listeria monocytogenes. Nat Rev Immunol. 2004;4:812–823. doi: 10.1038/nri1461. [DOI] [PubMed] [Google Scholar]
  • 13.Rodgers RJ, Cook RG. MHC class Ib molecules bridge innate and acquired immunity. Nat Rev Immunol. 2005;5:459–471. doi: 10.1038/nri1635. [DOI] [PubMed] [Google Scholar]
  • 14.Sullivan LC, Hoare HL, McCluskey J, Rossjohn J, Brooks AG. A structural perspective on MHC class Ib molecules in adaptive immunity. Trends Immunol. 2006;27:413–420. doi: 10.1016/j.it.2006.07.006. [DOI] [PubMed] [Google Scholar]
  • 15.Urdahl KB, Sun JC, Bevan MJ. Positive selection of MHC class Ib-restricted CD8(+) T cells on hematopoietic cells. Nat Immunol. 2002;3:772–779. doi: 10.1038/ni814. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Kennedy MK, Glaccum M, Brown SN, Butz EA, Viney JL, Embers M, Matsuki N, Charrier K, Sedger L, Willis CR, Brasel K, Morrissey PJ, Stocking K, Schuh JC, Joyce S, Peschon JJ. Reversible defects in natural killer and memory CD8 T cell lineages in interleukin 15-deficient mice. J Exp Med. 2000;191:771–780. doi: 10.1084/jem.191.5.771. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Vugmeyster Y, Glas R, Perarnau B, Lemonnier FA, Eisen H, Ploegh H. Major histocompatibility complex (MHC) class I KbDb −/− deficient mice possess functional CD8+ T cells and natural killer cells. Proc Natl Acad Sci U S A. 1998;95:12492–12497. doi: 10.1073/pnas.95.21.12492. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Busch DH, Vijh S, Pamer EG. Animal model for infection with Listeria monocytogenes. Curr Protoc Immunol. 2001;Chapter 19(Unit 19):19. doi: 10.1002/0471142735.im1909s36. [DOI] [PubMed] [Google Scholar]
  • 19.Earl PL, Cooper N, Wyatt LS, Moss B, Carroll MW. Preparation of cell cultures and vaccinia virus stocks. Curr Protoc Protein Sci. 2001;Chapter 5(Unit5):12. doi: 10.1002/0471140864.ps0512s13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Moutaftsi M, Peters B, Pasquetto V, Tscharke DC, Sidney J, Bui HH, Grey H, Sette A. A consensus epitope prediction approach identifies the breadth of murine T(CD8+)-cell responses to vaccinia virus. Nat Biotechnol. 2006;24:817–819. doi: 10.1038/nbt1215. [DOI] [PubMed] [Google Scholar]
  • 21.Dubois S, Patel HJ, Zhang M, Waldmann TA, Muller JR. Preassociation of IL-15 with IL-15R alpha-IgG1-Fc enhances its activity on proliferation of NK and CD8+/CD44high T cells and its antitumor action. J Immunol. 2008;180:2099–2106. doi: 10.4049/jimmunol.180.4.2099. [DOI] [PubMed] [Google Scholar]
  • 22.Rubinstein MP, Kovar M, Purton JF, Cho JH, Boyman O, Surh CD, Sprent J. Converting IL-15 to a superagonist by binding to soluble IL-15R{alpha} Proc Natl Acad Sci U S A. 2006;103:9166–9171. doi: 10.1073/pnas.0600240103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Sandau MM, Schluns KS, Lefrancois L, Jameson SC. Cutting edge: transpresentation of IL-15 by bone marrow-derived cells necessitates expression of IL-15 and IL-15R alpha by the same cells. J Immunol. 2004;173:6537–6541. doi: 10.4049/jimmunol.173.11.6537. [DOI] [PubMed] [Google Scholar]
  • 24.Schluns KS, Klonowski KD, Lefrancois L. Transregulation of memory CD8 T-cell proliferation by IL-15Ralpha+ bone marrow-derived cells. Blood. 2004;103:988–994. doi: 10.1182/blood-2003-08-2814. [DOI] [PubMed] [Google Scholar]
  • 25.Schluns KS, Lefrancois L. Cytokine control of memory T-cell development and survival. Nat Rev Immunol. 2003;3:269–279. doi: 10.1038/nri1052. [DOI] [PubMed] [Google Scholar]
  • 26.Surh CD, Boyman O, Purton JF, Sprent J. Homeostasis of memory T cells. Immunol Rev. 2006;211:154–163. doi: 10.1111/j.0105-2896.2006.00401.x. [DOI] [PubMed] [Google Scholar]
  • 27.Williams MA, Holmes BJ, Sun JC, Bevan MJ. Developing and maintaining protective CD8+ memory T cells. Immunol Rev. 2006;211:146–153. doi: 10.1111/j.0105-2896.2006.00389.x. [DOI] [PubMed] [Google Scholar]
  • 28.Kim HJ, Verbinnen B, Tang X, Lu L, Cantor H. Inhibition of follicular T-helper cells by CD8(+) regulatory T cells is essential for self tolerance. Nature. 2010;467:328–332. doi: 10.1038/nature09370. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Atherly LO, Lucas JA, Felices M, Yin CC, Reiner SL, Berg LJ. The Tec family tyrosine kinases Itk and Rlk regulate the development of conventional CD8+ T cells. Immunity. 2006;25:79–91. doi: 10.1016/j.immuni.2006.05.012. [DOI] [PubMed] [Google Scholar]
  • 30.Broussard C, Fleischacker C, Horai R, Chetana M, Venegas AM, Sharp LL, Hedrick SM, Fowlkes BJ, Schwartzberg PL. Altered development of CD8+ T cell lineages in mice deficient for the Tec kinases Itk and Rlk. Immunity. 2006;25:93–104. doi: 10.1016/j.immuni.2006.05.011. [DOI] [PubMed] [Google Scholar]
  • 31.Dubois S, Waldmann TA, Muller JR. ITK and IL-15 support two distinct subsets of CD8+ T cells. Proc Natl Acad Sci U S A. 2006;103:12075–12080. doi: 10.1073/pnas.0605212103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Sun JC, Ma A, Lanier LL. Cutting edge: IL-15-independent NK cell response to mouse cytomegalovirus infection. J Immunol. 2009;183:2911–2914. doi: 10.4049/jimmunol.0901872. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Zitvogel L, Terme M, Borg C, Trinchieri G. Dendritic cell-NK cell cross-talk: regulation and physiopathology. Curr Top Microbiol Immunol. 2006;298:157–174. doi: 10.1007/3-540-27743-9_8. [DOI] [PubMed] [Google Scholar]
  • 34.Germain RN. T-cell development and the CD4-CD8 lineage decision. Nat Rev Immunol. 2002;2:309–322. doi: 10.1038/nri798. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS351644-supplement-1.eps (651KB, eps)
2
NIHMS351644-supplement-2.eps (946.2KB, eps)
3
NIHMS351644-supplement-3.eps (1.3MB, eps)
4
NIHMS351644-supplement-4.doc (41.5KB, doc)

RESOURCES