Abstract
In vitro and recent in vivo studies have identified protein kinase Cθ (PKCθ) as an important intermediate in signaling pathways leading to T cell activation, proliferation, and cytokine production. However, the importance of PKCθ to many T cell-driven inflammatory responses has not been demonstrated. In this study we show that although PKCθ is required for the development of a robust lung inflammatory response controlled by Th2 cells, it plays a lesser role in the development of a similar lung inflammatory response controlled by Th1 cells. PKCθ-deficient mice were strongly compromised in generating Th2 cells and exhibited reduced airway eosinophilia and Th2 cytokine production in lungs. PKCθ was required for the initial development of Th1 cells, with these cells exhibiting delayed kinetics of differentiation and accumulation. However, with recall Ag challenge via the airways, this defect was overcome, and lung infiltration and Th1 cytokine production were largely unimpaired in PKCθ-deficient animals. These data suggest that PKCθ can play roles in aspects of both Th2 and Th1 responses, but lung inflammation induced by Th2 cells is more dependent on this protein kinase than lung inflammation induced by Th1 cells.
Tcell activation requires the concerted activation of TCR- and CD28-mediated intracellular signal transduction pathways. Growing evidence has suggested that protein kinase C θ (PKCθ),4 a Ca2+-independent isoform of the PKC family of intracellular serine/threonine kinases, may be the major signaling moiety linking TCR- and CD28-specific signals necessary for efficient T cell activation (1, 2). PKCθ is more strongly expressed in T cells than other cell types and is translocated to the site of contact between T cells and APCs, where it colocalizes with the TCR/CD3 complex in the central core of the immunological synapse after TCR/CD28 ligation (3, 4, 5, 6). Furthermore, studies using pharmocological inhibitors and PKCθ antisense RNA and overexpression studies using dominant negative and constitutively active mutant proteins have established a role for PKCθ in activation of AP-1 (7) and NF-κB (8, 9). PKCθ also synergizes with calcineurin to activate NFAT and IL-2 transcription (10, 11), and with Vav to mediate IL-4 gene expression in response to CD28 costimulation (12).
In further support of its important role in T cell activation, peripheral T cells isolated from PKCθ deficient mice (PKC−/−) demonstrate profound defects in TCR/CD28-induced IL-2 production and proliferation (4, 13). TCR/CD28-induced NF-κB and AP-1 activation is significantly reduced in PKC-deficient T cells (4, 13), and NFAT activation is abrogated (13).
Although these in vitro data strongly support the idea of a critical function for PKCθ in T cell responses, there are few studies to date showing the importance of this molecule in T cell-driven immune responses in vivo. A recent report in PKCθ knockout mice demonstrated normal generation of lymphocytic choriomeningitis virus (LCMV)-specific CD8 T cells and normal Ab responses after vesicular stomatitis virus (VSV) infection, but that PKCθ-deficient CD8 T cells were impaired in generating CTL activity after peptide immunization (14). This suggests that there may be differential requirements for PKCθ depending on the subset of T cells responding or the inflammatory conditions that accompany T cell priming. In the present study we used PKCθ-deficient mice and determined their susceptibility to lung inflammation controlled by either Th2 or Th1 cells. Our results clearly demonstrate that PKCθ is required for optimal Th2 responses in vivo, including the production of high levels of IL-4 and IL-5, the recruitment of large numbers of eosinophils to the airways, goblet cell hyperplasia, and mucus production. Development of Th1 cells was impaired and delayed in kinetics, but neutrophilia and production of IFN-γ in lungs were relatively normal after recall Ag challenge. These data provide clear evidence that this enzyme plays a major role in the development of Th2 cells and allergic asthmatic reactions, but is not absolutely essential for Th1 cells to be generated. This suggests that a requirement for PKCθ may be overcome in certain situations that involve strong Th1-inducing stimuli.
Materials and Methods
Mice
The studies reported here conform to the principles outlined by the animal Welfare Act and the National Institutes of Health guidelines for the care and use of animals in biomedical research. PKCθ-deficient mice (PKC−/−; originally on a 129-C57BL/6J background) were a gift from Dr. D. Littman (Howard Hughes Medical Institute, Skirhall Institute of Biomolecular Medicine, New York, NY;4) and were backcrossed for more than five generations onto the C57BL/6 background. These mice have no defect in thymic selection and bear normal numbers of peripheral T cells. Wild-type (wt; PKC+/+) C57BL/6 mice were used as controls.
Induction of airway inflammation
Mice were sensitized and challenged as described previously (15) to induce a Th2 response or with a modification described below to induce a Th1 response. Briefly, mice were either sensitized by i.p. injection of 20 μg of OVA protein (chicken egg albumin; Sigma-Aldrich, St. Louis, MO) adsorbed to 2 mg of aluminum hydroxide (alum; Pierce) in PBS on day 0 for a Th2 response or sensitized by s.c. injection of OVA (50 μg/ml) in CFA at the base of the tail for a Th1 response. Nonsensitized (control) mice received 2 mg of alum in PBS or CFA alone. On day 25 all mice were challenged via the airways with OVA (5 mg/ml in 15 ml of PBS) for 30 min, once a day for 4 consecutive days, by ultrasonic nebulization. Mice were killed 18–24 h after the last aerosol challenge and were assessed for inflammation of the lung.
Measurement of airway hyper-responsiveness (AHR)
AHR was measured in vivo 1–3 h after the last aerosolized OVA exposure by recording respiratory curves by whole-body plethysmography (Buxco Electronics, Wilmington, NC) in response to inhaled methacholine (1.25–10 mg/ml; Aldrich, Milwaukee, WI) as described previously (15).
Bronchoalveolar lavage (BAL) measurements
Inflammation was assessed by BAL. Lungs were lavaged five times with 1-ml aliquots of PBS containing 2% BSA. The recovered BAL fractions were centrifuged; cells were counted, then pelleted onto glass slides by cytocentrifugation. Differential cell counts were performed on Hema 3 (Fisher Scientific, Pittsburgh, PA)-stained cytospins using standard morphology and staining characteristics. BAL was assessed for cytokine content by standard ELISA protocols using commercially available Abs or those produced in-house (15
Characterization of lung histology and cellular infiltrates
Lung sections were prepared and analyzed as described previously (15). Lungs were removed and postfixed in 10% zinc-buffered formalin (Protocal; Fisher Diagnostics, Middletown, VA) and paraffin-embedded, and sections (5 μm) were stained with H&E for quantitation of inflammatory infiltrates or with periodic acid-Schiff (PAS) for quantitation of bronchial mucus gland hypertrophy. A semiquantitative scoring system was used to grade the size of lung infiltrates, where +5 signified a large (more than three cells deep) widespread infiltrate around the majority of vessels and bronchioles, and +1 signified a small number of inflammatory foci. To ensure comparable analysis between different groups, we analyzed six to eight randomly selected airways per lung section per animal, and an average inflammatory score was obtained for the entire lung tissue.
Assay of T cell function ex vivo
Peribronchial lymph nodes (PBLNs) were collected at the time of lung harvest. Lavaged lungs were digested with HBSS (Invitrogen Life Technologies, Gaithersburg, MD) supplemented with 3 mg/ml collagenase (type V; Roche, Indianapolis, IN), 0.1 mg/ml DNase (Sigma-Aldrich), 100 μg/ml streptomycin (Invitrogen Life Technologies, Carlsbad, CA), and 100 U/ml penicillin (Invitrogen Life Technologies) for 60 min at 37°C. After lysing RBCs with ACK lysis buffer, PBLN and lung cells were resuspended in RPMI 1640 medium (Invitrogen Life Technologies) supplemented with 10% FCS (Omega Scientific, Tarzana, CA), 1% l-glutamine (Invitrogen Life Technologies), 100 μg/ml streptomycin, 100 U/ml penicillin, and 50 μM 2-ME (Sigma-Aldrich). Lung cells (8 × 105 cells/well), or PBLN cells (2 × 105 cells/well) were plated in round-bottom, 96-well microtiter plates in 200 μl with increasing concentrations of OVA (10–100 μg/ml) for 96 h at 37°C. After 84 h, 1 μCi of [3H]thymidine (Valeant Pharmaceuticals, Costa Mesa, CA) in 10 μl was added to each well. The cells were harvested 12 h later, and thymidine incorporation was measured using a Betaplate scintillation counter (Tomtec, Hamden, CT). Each in vitro stimulation was performed in quadruplicate. Supernatants were harvested after 96 h for cytokine analysis.
Phenotypic and intracellular staining of T cells
Cell surface molecules were analyzed by FACS. After blocking FcRs with excess anti-Fc (24.G2), cells were incubated with fluorescently conjugated Abs against CD4 (FITC or allophycocyanin), CD3 (PE), CD8 (FITC), OX40 (PE), CD25 (PE), or CD69 (PerCP). Flow cytometric measurement of cytokine production in T cells was performed according to standard protocols, with some modifications. Briefly, after lysing RBCs, PBLN or splenocytes were resuspended in RPMI 1640 medium (Invitrogen Life Technologies) supplemented with 10% FCS (Omega Scientific), 1% l-glutamine (Invitrogen Life Technologies), 100 μg/ml streptomycin, 100 U/ml penicillin, and 50 μM 2-ME (Sigma-Aldrich). Lung cells (8 × 105 cells/well) or lymph node cells (2 × 105 cells/well) were plated in round-bottom, 96-well microtiter plates in 200 μl with increasing concentrations of OVA (500 μg/ml) for 90 h at 37°C. GolgiPlug (BD Biosciences) was added to the cultures according to the manufacturer’s instructions, and the incubation was continued for 8 h. At the end of the culture, cells were harvested, washed twice in PBS plus 2% BSA buffer, stained anti-CD4, and followed by fixation with Cytofix-Cytosperm (BD Biosciences) for 20 min at 4°C. Fixed cells were washed twice in 1× BD Perm/Wash solution (BD Biosciences) and subjected to intracellular cytokine staining in Perm/Wash buffer for 30 min at 4°C. Anti-IL-2 (FITC and PE), anti-IL-4 (PE and allophycocyanin), anti-IL-5 (PE and allophycocyanin), and anti-IFN-γ (FITC and allophycocyanin) were all obtained from BD Pharmingen (San Diego, CA) and used at a 1/50 dilution. Samples were analyzed for their proportion of cytoplasmic cytokines after gating on CD4+ T cells by FACSCalibur flow cytometer using CellQuest (BD Biosciences) and FlowJo software (Tree Star, San Carlos, CA).
In vitro T cell differentiation
CD4+ T cells were purified from spleen and lymph nodes by passing over nylon columns, followed by complement lysis using Abs to CD8 (3.155), heat-stable Ag (J11D), class II MHC (M5/114, Y17, and CA-4.A12), B cells (RA3.6B2), macrophage (M1/70), NK cells (PK136), and dendritic cells (33D1). T cells (5 × 105 cells/ml) were stimulated with plate-bound anti-CD3 (145-2C11; 5 μg/ml) and soluble anti-CD28 (37N51; 5 μg/ml) with or without IL-2 (30 ng/ml). After 7 days, live cells were restimulated with anti-CD3 or anti-CD3/CD28, and supernatants were harvested 24 h after secondary stimulation for assay of cytokine content by ELISA.
Statistics
Statistical significance was analyzed by Student’s t test. Unless otherwise indicated, data represent the mean ± SD, with p < 0.05 considered statistically significant.
Results
PKCθ-deficient mice exhibit reduced lung inflammation driven by Th2 cytokines
AHR is a cardinal feature of a pulmonary allergic response and has been linked with the development of Th2 cells. To determine whether a deficiency in PKCθ has a direct effect on the development of airway dysfunction, wt (PKC+/+) and PKCθ-deficient (PKC−/−) mice were sensitized with OVA in alum to generate a Th2 response and subsequently challenged by inhalation of OVA several weeks later. Wild-type mice developed AHR in response to methacholine challenge. In contrast, PKC−/− mice showed attenuation in methacholine sensitivity, as characterized by a reduction in AHR (Fig. 1⇓A). Airway reactivity to methacholine reflects a combination of increased smooth muscle sensitivity, due to inflammatory mediators such as histamine and leukotrienes, and airway narrowing, due to inflammation (16). To assess whether PKCθ deficiency influences airway inflammation, lung lavages were performed on mice killed 24 h after the last Ag challenge, and the BAL was examined for influx of total leukocytes. In OVA-primed PKC+/+ mice, the total number of BAL leukocytes dramatically increased after repeated exposure to OVA, compared with unprimed challenged mice, which was characterized by a predominant population of eosinophils (Fig. 1⇓B). In contrast, the number of total leukocytes, including eosinophils and lymphocytes, was reduced by ∼ 70% in the airways of PKC−/− mice.
Ag-induced AHR, eosinophilia, and Th2 cytokine production is reduced in the absence of PKCθ. Groups of PKCθ-deficient (PKC−/−) and wt (PKC+/+) mice were immunized i.p. with OVA in alum or injected with alum alone (control). Twenty-five days later, all mice were challenged by inhalation of nebulized OVA. A, One to 3 h after the last aerosol challenge, individual mice were assessed for AHR by barometric plethysmography. Results are the mean percent change in enhanced pause levels above baseline (saline-induced AHR), after exposure to increasing concentrations of inhaled methacholine. Values were calculated from four mice in each group per experiment. B, Total leukocyte numbers were counted in BAL recovered from PKCθ-deficient (PKC−/−; n = 10) and wt (PKC+/+; n = 6) mice 24 h after the last OVA challenge. Total numbers of eosinophils (Eosin), lymphocytes (Lymph), and monocytes (Mono) were calculated from differently stained BAL cytospins. C–H, Twenty-four hours after the final Ag challenge, lung tissue sections from PKC+/+ (C) or PKC−/− (D) mice were stained with H&E (magnification, ×200) for quantitation of inflammatory infiltrates and PAS (magnification, ×200; dark staining) to highlight the mucus-secreting cells (F and G). Lung sections were graded by light microscopy on an arbitrary scale from 0–5 for inflammation severity (E) and by percentage of PAS+ epithelial cells for mucus production (H). I–L, One day after the last challenge, BAL fluid from PKC−/− and PKC+/+ mice was assessed for Th2 cytokines, IL-4 (I), IL-5 (J), IL-13 (K), and IFN-γ (L) by ELISA. Results are the mean ± SEM from six to 10 mice per group. ∗, p < 0.0001; #, p < 0.03 (vs wt mice immunized and challenged with OVA).
The attenuated severity of airway inflammation observed in PKC−/− mice was also supported by histological analysis of lung sections. As shown in Fig. 1⇑, C–E, wt mice developed a prominent inflammatory infiltrate, characterized by eosinophilic and mononuclear inflammation in perivascular and peribronchiolar areas, along with extensive goblet cell hypertrophy and mucus (PAS+) production. In contrast, lung sections from PKC−/− mice had far fewer infiltrating cells around the bronchioles and blood vessels and relatively normal bronchial epithelium and reduced mucus (Fig. 1⇑, F–H).
AHR, eosinophila, and mucus production are controlled by Th2 cytokines, such as IL-4, IL-5, and IL-13 (17, 18, 19, 20). To determine whether the absence of PKCθ impaired Th2 cytokine production in the lungs, BAL was assessed for IL-4, IL-5, and IL-13 24 h after the last OVA aerosol challenge. In PKCθ−/− animals, the levels of IL-4 (Fig. 1⇑I), IL-5 (Fig. 1⇑J), and IL-13 (Fig. 1⇑K) were reduced by an average of 60, 82, and 73%, respectively, compared with wt controls. IFN-γ, which was induced at only low levels above background, was not significantly different in PKCθ−/− mice (Fig. 1⇑L). Because we found no evidence for elevated levels of IFN-γ or neutrophilia (not shown), which can accompany production of this cytokine (21), this suggests that there was not strong skewing to development of Th1 cells when PKCθ was absent, which is of significance when considering the results below. Collectively, these results show that PKCθ was required for promoting the asthmatic lung inflammation driven by Th2 cells.
PKCθ-deficient mice are largely unimpaired in generating lung inflammation driven by Th1 cytokines
To determine whether PKCθ also has a critical role in Th1 responses that result in lung inflammation, wt (PKC+/+) and PKCθ-deficient (PKC−/−) mice were sensitized with OVA in CFA to generate a Th1 response and subsequently challenged by inhalation of OVA several weeks later. In contrast to the Th2 response, lung inflammation driven by Th1 cells involves neutrophils, rather than eosinophils, and little/no AHR or mucus production (Fig. 2⇓ and data not shown) (21). As demonstrated in Fig. 2⇓, A, C, and E, wt mice developed a prominent inflammatory infiltrate that was somewhat more dispersed than in the Th2 response, but was still largely in perivascular and peribronchiolar areas. In contrast to the results with Th2-driven inflammation, lung sections from PKC−/− mice had similar infiltration to that observed in wt mice (Fig. 2⇓, B, D, and E). Analyses of BAL showed marked and similar elevations in total leukocyte numbers in both wt and PKC−/− mice and almost identical numbers of lymphocytes, although a statistically significant reduction in neutrophils was consistently evident in the absence of PKCθ (Fig. 2⇓F). Again, in contrast to the Th2 response, the signature cytokine of the Th1 response, IFN-γ, was not reduced in PKC−/− BAL compared with wt mice (Fig. 2⇓G). These data show that PKCθ did not play a dominant role in promoting lung inflammation under the recall Th1-priming conditions used.
PKCθ-deficient mice are able to mount a Th1-mediated lung inflammatory response. Groups of PKCθ-deficient (PKC−/−; n = 4) and wt (PKC+/+; n = 4) mice were immunized s.c. with OVA in CFA or injected with CFA alone (control). Twenty-five days later, all mice were challenged by inhalation of nebulized OVA. A–E, Twenty-four hours after the final Ag challenge, lung tissue sections from PKC+/+ (magnification: A, ×40; C, ×100) or PKC−/− (magnification: B, ×40; D, ×100) mice were stained with H&E for quantitation of inflammatory infiltrates. Lung sections were graded by light microscopy on an arbitrary scale from 0–5 for inflammation severity (E). F, Total leukocyte numbers were counted in BALF recovered from PKCθ-deficient (PKC−/−) and wt (PKC+/+) mice 24 h after the last OVA challenge. Total numbers of neutrophils (Neut), lymphocytes (Lymph), and monocytes (Mono) were calculated from differently stained BAL cytospins. G, One day after the last challenge, BAL from PKC−/− and PKC+/+ mice was assessed for IFN-γ by ELISA. Results are the mean ± SEM from four mice per group. One representative of two independent experiments is shown. ∗, p < 0.02, (vs wt mice immunized and challenged with OVA).
PKCθ controls accumulation of Ag-induced CD4 cells during Th2, but not Th1, lung inflammation
The data from lung analyses suggested that priming of Th2 cells, but not Th1 cells, might be controlled by PKCθ. In the absence of functional reagents to allow tracking of OVA-reactive CD4 cells in PKC−/− mice, we performed a number of phenotypic analyses to quantitate the extent of T cell priming that accompanied the different lung inflammatory responses (Fig. 3⇓). In the Th2 response, analyses of percentages of CD4 cells in the CD3 compartment in lung draining lymph nodes at the time of inflammation showed no difference in wt mice after Ag priming and airway challenge (Fig. 3⇓A, top). However, when converted into total CD4 T cell numbers, a great elevation was observed in wt mice compared with nonprimed mice, but a much more modest increase was observed in PKC−/− mice (Fig. 3⇓A, top right). Analyses of CD44 expression, a marker of Ag experience, on CD4 cells showed a significantly greater percentage of high expressors in wt mice which translated into far more primed T cells than in PKC−/− mice (Fig. 3⇓A, middle). Lastly, analyses of OX40-expressing cells within the CD4/CD44high population also showed a marked reduction in the number induced in PKC−/− mice after Ag priming (Fig. 3⇓A, bottom). OX40 is an inducible costimulatory molecule that we have previously shown to be a good marker of Ag-responding T cells (22). It is involved in both Th2 (22) and Th1 (S. Salek and M. Croft, unpublished observations) lung inflammatory reactions. Collectively, these data suggest that under the conditions that resulted in Th2 cytokine-driven inflammation, there were far fewer Ag-induced CD4 cells present during the lung response when PKCθ was absent.
PKCθ is required for the accumulation of activated CD4 T cells during Th2 but not Th1 induced lung inflammation. Groups of PKCθ-deficient (PKC−/−; n = 4) and wt (PKC+/+; n = 4) mice were immunized i.p. with OVA in alum (A) or s.c. with OVA in CFA (B). Control mice were immunized with either alum or CFA only. Twenty-five days later, all mice were challenged by inhalation of nebulized OVA. One day after the last challenge, lung draining lymph nodes were collected individually, single-cell suspensions were made, and total cell numbers were counted. A and B, Top panel, Percentage and absolute numbers of CD4+ T cells. Lymph node cells were triple-stained for CD3, CD4, and CD8. Absolute numbers of CD3+CD4+ cells per organ are shown in the right column. Middle panel, Percentage (left) and absolute numbers (right) of CD44high-expressing cells CD3+CD4+ T cells. ▦, Cells stained with isotype-matched control mAb. The numbers in each histogram indicate the percentage of CD3+CD4+ cells that express high levels of CD44. Bottom panel, Percentage (left) and absolute numbers (right) of OX40+ cells gating on CD4+CD44high T cells. Quadrant settings, distinguishing positive from background fluorescence, were determined by staining with isotype-matched control mAbs (not shown). ∗, p < 0.0001 (vs wt mice immunized and challenged with OVA). Similar results were obtained in two independent experiments.
In contrast to these data, when the same parameters were measured at the time of the Th1 lung response, analyses of Ag-induced CD4 numbers (Fig. 3⇑B, top), CD4/CD44high numbers (Fig. 3⇑B, middle), and the number of CD4/CD44high cells that expressed OX40 (Fig. 3⇑B, bottom) showed no significant difference compared with wt mice.
To further delineate the extent of Th2 and Th1 priming during the opposing lung responses, we performed in vitro recall responses, restimulating draining lymph node T cells with OVA. As is typical of these divergent responses, Th2 cells in wt mice proliferated well, whereas Th1 cells displayed less proliferation based on a dose response (Fig. 4⇓A). Proliferative responses from PKC−/− mice were strongly reduced compared with wt mice in the Th2 response, whereas they were minimally impaired in the Th1 response (Fig. 4⇓A, left and right, respectively). Similarly, analyses of cytokine production by intracellular staining showed a marked reduction in the percentage of IL-4-producing (p < 0.05) and IL-5-producing (p < 0.01) CD4 cells in PKC−/− mice, but no difference or even enhanced percentages of IFN-γ-producing cells (Fig. 4⇓B). Similar results were obtained when supernatants were examined for cytokine production by ELISA (data not shown). Thus, the combined data from Figs. 3⇑ and 4⇓ suggested that at the time of the Th2 lung inflammatory response, fewer Ag-reactive T cells accumulated, and they displayed a partial hyporesponsive phenotype in the absence of PKCθ. In contrast, at the time of the Th1 lung inflammatory response, there was little evidence for defective T cell priming with normal numbers of Ag-reactive T cells that were capable of making IFN-γ.
Reduced Ag-specific T cell reactivity in PKCθ-deficient mice during Th2-mediated, but not Th1-mediated, inflammatory responses. Mice were immunized with OVA under Th2 (alum) or Th1 (CFA)-inducing conditions and subsequently challenged with aerosolized OVA as in Fig. 1⇑. One day after the last challenge, lung draining lymph nodes from PKC−/− (○; n = 4) and PKC+/+ (•; n = 4) mice were tested in vitro for T cell reactivity to OVA. A, Proliferation in lung draining lymph nodes after 96 h of stimulation with increasing concentrations of OVA as indicated. Results are the mean ± SEM from quadruplicate cultures, using cells pooled from four to seven mice per group. B, Lung draining lymph node cells were restimulated in vitro with OVA for 96 h. Cells were harvested and stained for CD4 and intracellular IL-4, IL-5, or IFN-γ. The numbers represent the percentage of IL-4-, IL-5-, IFN-γ-positive cells gated on CD4+ T cells. Quadrant settings, distinguishing positive from background fluorescence, were determined by staining with isotype-matched control mAbs (not shown). Similar results were obtained in two independent experiments.
PKCθ has an early role in Th1 as well as Th2 responses
Because PKCθ has been thought to control initial T cell activation, and the lung inflammatory responses were induced by Ag rechallenge of mice several weeks after immunization, we assessed earlier phenotypic and functional markers of priming to determine whether the defect in Th2 responses was a result of impaired early priming, and whether any defect might be evident in the Th1 response. Analyses were conducted on draining lymph node T lymphocytes similar to those during the lung response (see Fig. 3⇑), but on days 3 and 7 after immunization, molecules associated with T cell activation, CD25, CD69, and OX40 (Fig. 5⇓) and production of cytokines (Fig. 6⇓) were assessed. The number of CD4/CD44high-expressing cells that were positive for CD25, CD69, or OX40 was strongly reduced in PKC−/− mice under Th2-priming conditions even by day 3, and this phenotype was maintained through to day 7 (Fig. 5⇓). Expression levels of these molecules on the few PKC−/− T cells that were positive were fairly comparable to those on wt T cells (data not shown), suggesting that a small number of Th2 cells was responding normally.
PKCθ is required for early accumulation of Th2 and Th1 cells during the primary immune response. Groups of PKCθ-deficient (PKC−/−; n = 4) and wt (PKC+/+; n = 4) mice were immunized i.p. with OVA in alum (Th2), or s.c. with OVA in CFA (Th1). Inguinal and lumbar lymph nodes from PKC−/− (○; n = 4) and PKC+/+ (•; n = 4) mice were collected individually at the indicated days postimmunization with Ag and total cell numbers counted. Cells were stained with CD4 plus CD44 and CD25, CD69, or OX40. The absolute numbers of CD25+ (A), CD69+ (B), or OX40+ (C) on CD4/CD44high-expressing T cells were determined and are presented as the mean ± SD of four animals per group. Data are expressed as cell numbers per organ. One representative of three independent experiments is shown. ∗, p < 0.0001; ∗∗∗, p < 0.001; #, p < 0.02 (vs wt mice immunized and challenged with OVA).
PKCθ-deficient CD4 T cells exhibit impaired early Th2 and Th1 differentiation in the primary immune response. Groups of PKCθ-deficient (PKC−/−; n = 4) and wt (PKC+/+; n = 4) mice were immunized i.p. with OVA in alum (Th2) or s.c. with OVA in CFA (Th1). After 3 or 7 days, inguinal and lumbar lymph nodes from individual mice were collected, and single-cell suspension were prepared. Total lymph node cells were restimulated in vitro with OVA for 96 h, after which CD4 T cells were examined for expression of IL-4 (A), IL-5 (A), IL-2 (B), and IFN-γ (C). The numbers represent the percentage of IL-4-, IL-5-, IL-2-, IFN-γ-positive cells gated on CD4+ T cells. Quadrant settings, distinguishing positive from background fluorescence, were determined by staining with isotype-matched control mAbs. One representative of three independent experiments is shown.
Significantly and in contrast to data in recall responses, under Th1 conditions a defect in Th1 cell priming was also apparent in PKC−/− mice, with lower numbers of CD25-, CD69-, or OX40-expressing CD4/CD44high cells being detected (Fig. 5⇑). This was more pronounced on day 3 than day 7, suggesting that the kinetics of Th1 development were delayed in the absence of PKCθ. This was substantiated by analyzing production of Th1 and Th2 cytokines at these time points (Fig. 6⇑). IL-2 (p < 0.046), IFN-γ (p < 0.034), and IL-5 (p < 0.013) production were impaired on day 3, whereas by day 7, IL-2 and IFN-γ were equivalently produced in the absence of PKCθ.
To extend these data, we performed an in vitro differentiation assay under neutral conditions, where wt vs PKC−/− CD4 cells were stimulated over a period of 1 wk with anti-CD3 and anti-CD28 in the presence or the absence of IL-2 (Fig. 7⇓). Upon restimulation of equivalent numbers of effector T cells generated with CD3/CD28 as well as a pronounced defect in IL-4 production, PKC−/− T cells were also impaired in secreting IFN-γ, albeit to a lesser extent. Addition of IL-2 largely overcame the IFN-γ defect, but not the IL-4 defect, supporting the idea that the strength of signaling might compensate for a dominant role for PKCθ in Th1 differentiation. Thus, PKCθ is not simply involved in Th2 responses, but can play a role in some aspects of the Th1 response and, based on our in vivo and in vitro data, can control the kinetics of differentiation into the Th1 phenotype as well as the kinetics of accumulation of Th1 cells.
PKCθ-deficient CD4 T cells are impaired in Th1 as well as Th2 differentiation in vitro. Purified CD4 T cells from PKC−/− and PKC+/+ mice were stimulated in vitro under neutral conditions with anti-CD3 and anti-CD28 in the absence (A and C) or presence (B and D) of IL-2. After 1 wk, an equivalent number of live T cells were restimulated with anti-CD3, and IL-4 (A and B) and IFN-γ (C and D) production was measured. Similar results were seen in an additional experiment where T cells were restimulated with anti-CD3 and anti-CD28 (not shown). Identical data were obtained in two experiments. Approximately 10-fold lower viable PKC−/− (1.07 × 106 cells/ml) T cells were recovered at the end of the primary stimulation (day 7) with anti-CD3 and anti-CD28 compared with PKC+/+ (11.04 × 106 cells/ml) cells. The mean (±SD) percent reductions in IL-4 (A) and IFN-γ (C) from two independent experiments were 95.5 ± 0.7 and 59 ± 14.6%, respectively.
Discussion
Collectively, our data demonstrate that PKCθ is integral to the development of the Th2 response that results in asthma-like lung inflammation, whereas Th1 cells that infiltrate and initiate lung inflammation do not exhibit a profound requirement for PKCθ to be expressed. In contrast, during primary responses, both the number of T cells generated under Th1 and Th2 inflammatory conditions and the production of Th1 and Th2 cytokines were reduced in the absence of PKCθ, suggesting that a role in Th1 responses might be subdominant and overcome with time and/or depending on the level of inflammation. These results extend many in vitro findings on the importance of PKCθ to T cell priming, but also highlight the fact that certain T cell responses can develop relatively normally in the absence of this enzyme.
Many previous reports from in vitro systems have suggested that T cell activation is dependent on PKCθ (reviewed in Ref.1); in particular, PKCθ-deficient T cells were almost completely impaired in expressing CD25 and CD69 and proliferating in response to anti-CD3 and anti-CD28 (4). Additionally, a compromised in vitro recall proliferative response was reported in the latter study when assayed 2 wk after immunization of PKCθ-deficient mice with keyhole limpet hemocyanin in alum (4). Although our data assessing primary Th1 and Th2 responses correlated with these earlier reports on the importance of PKCθ, it is very interesting that lung inflammation induced by recall challenge of mice primed under Th1-polarizing conditions was only mildly affected in its absence. There was little evidence of a long term impairment of Th1 development, even though analysis of primary responses showed reduced accumulation of Th1 cells and delayed kinetics of differentiation into the Th1 phenotype. Similarly, recent data from Berg-Brown et al. (14) also showed that CTL responses to LCMV infection and CD4-dependent Ab responses to VSV infection were not reduced in PKCθ-deficient mice. In contrast, Berg-Brown et al. (14) did show strongly impaired CD8 responses and a form of CTL anergy in the absence of PKCθ when soluble peptide Ag was administered in vivo. While this manuscript was in the process of revision, Marsland et al. (23) reported defects in Th2-controlled lung inflammation in PKCθ-deficient mice similar to those reported in this study, after either immunization with OVA or infection with Nippostrongylus brasiliensis. Furthermore, they demonstrated that PKCθ-deficient mice on both the C57BL/6 background and the susceptible BALB/c background were able to mount a normal protective Th1 response against Leishmania major infection. This is also in agreement with our data showing that Th1 recall responses in the lung were minimally impaired in PKCθ-deficient mice that were immunized and subsequently challenged via the airways with OVA. The main difference in our data is the finding that initial priming of Th1 cells was defective, whereas these investigators did not determine whether early in vivo priming of Th1 cells was also compromised in PKCθ-deficient mice infected with L. major. Thus, PKCθ appears to have a subtler role in controlling T cell responses than previously appreciated.
Why, then, does PKCθ appear to be essential for some T cell responses and not others? One possibility is a differential role in CD4 vs CD8 T cells, but our data combined with those reported by Marsland et al. (23) and Berg-Brown et al. (14) argue against this. Another idea promoted by the combined results is that Th1 and Th2 cells or the generation of Th1 and Th2 phenotypes are differentially controlled by PKCθ. The finding of normal IgM and IgG responses to VSV, which induces Th1 immunity, are in line with this, and the finding of normal CTL responses to LCMV also loosely correlates with this idea, because such responses are type 1 (cytotoxic T cell (Tc1) and Th1) cytokine dominated. However, it needs to be tested whether every Th2- or Tc2-driven response will be critically dependent on PKCθ, and whether other Th1/Tc1-controlled responses can occur in its absence, before any rigid conclusion can be drawn. Additionally, it needs to be determined molecularly whether molecules such as GATA-3, c-Maf, and T-bet, which control transcription of Th2 and Th1 cytokines, are differentially controlled by PKCθ to fully understand the significance of the in vivo phenotypes.
In contrast to differential control of subsets of T cells, another idea regarding what dictates the use of PKCθ is that it is controlled by either the participation of costimulatory molecules such as CD28 or it is determined by the overall level of inflammatory signals, perhaps bridging innate and adaptive immunities, that are encountered during naive T cell priming. The latter could include production of cytokines such as IL-12 that drive Th1/Tc1 generation or cytokines using receptors with the common γ-chain that can control the extent of T cell clonal expansion.
The possibility that the participation of CD28 in a T cell response might be crucial to the involvement of PKCθ is supported by previous in vitro data centrally linking it to CD28 signals (8, 24, 25) and data showing that PKCθ-deficient CD8 cells behave almost identically to CD28-deficient CD8 cells (14). In relation to our work in this study and that of Marsland et al. (23) showing defective Th2-driven lung inflammation, the connection is further supported by earlier studies demonstrating that CD28/B7 interactions are also critical to similar lung responses controlled by Th2 cells (26, 27, 28). However, other data appear to argue against the simple concept that only CD28 activates PKCθ, and hence, a role for PKCθ is apparent only when CD28 is engaged. Although we do not know whether CD28 is required for the Th1 lung inflammatory response in our studies, we have unpublished data showing the involvement of two other costimulatory molecules in this response, namely OX40 and 4-1BB. Both molecules strongly synergize with CD28 (29, 30), so it is likely that CD28 may be engaged in this study even though little or no role for PKCθ was ultimately found in the lung response. Additionally, previous studies have shown a crucial role for CD28 in Ab production after VSV infection (31), which contrasts strongly with the unimpaired responses observed in PKCθ-deficient mice (14). Thus, perhaps the overall level of signaling at the time of naive T cell encounter with Ag is more crucial to whether PKCθ plays a dominant or a subdominant role in T cell priming rather than the involvement of CD28. Again, this might be dictated by the presence or the absence of inflammatory cytokines such as IL-12, which could dictate not only the threshold level of signals, but also the Th1/Th2 balance. In the responses where PKCθ has a dominant role (to soluble peptide, immunization in alum, and Nippostrongylus), there are probably major differences in the quantity and type of innate cell-derived cytokines produced compared with those responses where PKCθ has little apparent role (to LCMV and VSV, immunization in CFA, and Leishmania). Our data, showing that a Th1 defect can be overcome, even though the initial differentiation and accumulation of Th1 cells are impaired in the absence of PKCθ, are in line with the idea that the strength or duration of signaling might determine the requirement for PKCθ.
In summary, the results in this study show that PKCθ can play roles in the initial priming of Th1 and Th2 cells, but that it can have a restricted role in T cell responses as they develop with an apparent bias to being far more essential for Th2 than for Th1 inflammatory reactions. Analysis of PKCθ involvement in other diseases states that are driven by Th1 cells, and understanding which membrane-expressed cytokine and costimulatory receptors activate PKCθ and which might activate alternate pathways that bypass a requirement for PKCθ will be important for determining whether targeting PKCθ has only a limited therapeutic potential for immune diseases.
Acknowledgments
We thank Dr. Dan Littman for originally making available the PKCθ-deficient mice.
Footnotes
-
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
-
↵1 This work was supported by National Institutes of Health Grant AI50498 (to M.C.) and CA35299 and AI49888 (to A.A.). This is manuscript 589 from the La Jolla Institute for Allergy and Immunology.
-
↵2 A.A. and M.C. share senior authorship.
-
↵3 Address correspondence and reprint requests to Dr. Michael Croft, La Jolla Institute for Allergy and Immunology, Division of Molecular Immunology, 10355 Science Center Drive, San Diego, CA 92121. E-mail address: mick{at}liai.org
-
↵4 Abbreviations used in this paper: PKCθ, protein kinase Cθ; AHR, airway hyper-reactivity; BAL, bronchoalveolar lavage; LCMV, lymphocytic choriomeningitis virus; PAS, periodic acid-Schiff; PBLN, peribronchial lymph node; VSV, vesicular stomatitis virus; wt, wild-type mice; Tc, cytotoxic T cells.
- Received November 5, 2003.
- Accepted September 3, 2004.
- Copyright © 2004 by The American Association of Immunologists
References
In this issue
