It is well known that protein kinase C (PKC) plays an important role in regulation of TCR cell surface expression levels. However, eight different PKC isotypes are present in T cells, and to date the particular isotype(s) involved in TCR down-regulation remains to be identified. The aim of this study was to identify the PKC isotype(s) involved in TCR down-regulation and to elucidate the mechanism by which they induce TCR down-regulation. To accomplish this, we studied TCR down-regulation in the human T cell line Jurkat, in primary human T cells, or in the mouse T cell line DO11.10 in which we either overexpressed constitutive active or dominant-negative forms of various PKC isotypes. In addition, we studied TCR down-regulation in PKC knockout mice and by using small interfering RNA-mediated knockdown of specific PKC isotypes. We found that PKCα and PKCθ were the only PKC isotypes able to induce significant TCR down-regulation. Both isotypes mediated TCR down-regulation via the TCR recycling pathway that strictly depends on Ser126 and the di-leucine-based receptor-sorting motif of the CD3γ chain. Finally, we found that PKCθ was mainly implicated in down-regulation of directly engaged TCR, whereas PKCα was involved in down-regulation of nonengaged TCR.
The TCR is a multisubunit receptor complex composed of the Ag recognizing TCRαβ dimer (1, 2), and the CD3γε, CD3δε, and TCRζζ dimers responsible for signal delivery to the interior of the cell (3, 4). Cell surface expression of TCR is a dynamic process. During steady state, the amount of surface-expressed TCR is dependent on protein turnover (5) as well as constitutive cycling of the receptor (6, 7). At least two different, but interrelated pathways of TCR down-regulation are induced when the TCR is engaged by specific MHC/peptide complexes, anti-TCR Abs, or other ligands (8). One pathway depends on tyrosine kinase activity, Cbl, and ubiquitin, and leads to TCR degradation in the lysozymes (9, 10, 11, 12, 13, 14). The other pathway is dependent on protein kinase C (PKC)3-mediated activation of the CD3γ dileucine-based (diL) receptor-sorting motif and leads to TCR recycling (7, 15). In addition to down-regulation of directly engaged TCR, a second mechanism of TCR down-regulation exists. This process involves down-regulation of nonengaged TCR and is called TCR comodulation (16, 17, 18, 19, 20). It has been shown recently that TCR comodulation is dependent on PKC (21).
To date, nine distinct PKC isotypes have been described. Of these, six have a binding site for phorbol ester/1,2-diacylglycerol (DAG) and are present in T cells (PKCα, -β, -δ, -ε, -η, and -θ) (22). In the majority of studies describing the role of PKC in TCR down-regulation phorbol esters, like phorbol 12,13-dibutyrate (PDB) and PMA, have been used to activate PKC. With the present knowledge on phorbol ester substrates, it is most likely that the six different PKC isotypes as well as other proteins, for instance, chimaerin (23) and Ras exchange factor RasGRP3 (24), became activated by the phorbol esters applied in these studies. In several PKC studies, commercial PKC inhibitors (e.g., rottlerin and members of the Ro, Bis, and Gö series of inhibitors) were also applied. However, these inhibitors have been found to inhibit a broad spectrum of kinases other than PKC (25, 26), and hence, they are not reliable as tools for determining the involvement of specific PKC isotypes in cellular processes. In addition to the six PKC isotypes with binding sites for phorbol ester/DAG, T cells express two PKC isotypes (PKCζ and -ι) without this binding site. Thus, the specific role in TCR down-regulation of each of the eight PKC isotypes expressed in T cells remains to be determined.
The aim of this study was to identify the PKC isotypes involved in TCR down-regulation and to elucidate the mechanism by which they induce TCR down-regulation.
Materials and Methods
Cells and Abs
The human Jurkat T cell line JTag, the B cell line Raji (American Type Culture Collection), and a stable transfectant of the mouse T cell hybridoma DO11.10 expressing two different TCR comprising either Vβ2 or Vβ8 (21) were cultured in complete medium (C medium): RPMI 1640 medium (Sigma-Aldrich) supplemented with 10% FBS (Invitrogen Life Technologies), 2 × 105 U/L penicillin (Leo Pharmaceuticals Products), and 50 mg/ml streptomycin (Merck) at 37°C in 5% CO2. Primary human Vβ8+ T cells were generated from freshly isolated PBL and expanded in culture, as previously described (27). Mouse lymph node cells from 8-wk-old wild-type (WT), PKCα (−/−), and PKCθ (−/−) knockout (KO) mice were expanded in culture for 3 days at 1 × 106 cells/ml in C medium supplemented with 3.2 ng/ml murine rIL-2 (R&D Systems), 0.5 μM 2-ME, and 4 × 105 mouse CD3-CD28 expander beads/ml (Dynal Biotech). Production and genotyping of the PKCθ (28) and PKCα (29) KO are described elsewhere. Abs against the TCR included mAb F101.01 (anti-TCR) (30), mAb UCHT1 (anti-CD3ε; DakoCytomation), pAb A452 (anti-CD3ε; DakoCytomation), mAb BV8 (anti-Vβ8; BD Biosciences), mAb BV3 (anti-Vβ3; BD Biosciences), mAb Vβ2 and Vβ8 (BD Biosciences), mAb 6B10.2 (anti-TCRζ; Santa Cruz Biotechnology), and H57-597 (anti-TCRβ; BD Biosciences). Other Abs used included mAb against PKCα, PKCθ, PKCβ, PKCδ, PKCε, PKCζ, and PKCι (BD Biosciences); W6β2 (anti-panMHC-I, a gift from S. Buus, University of Copenhagen, Copenhagen, Denmark); mAb GFP (BD Clontech); mAb CD28 (BD Biosciences); mAb fyn (Santa Cruz Biotechnology); PE-conjugated goat anti-mouse IgG, Cy5-conjugated donkey anti-mouse IgG, and FITC-conjugated goat anti-Armenian hamster IgG (Jackson ImmunoResearch Laboratories); and HRP-conjugated swine anti-rabbit Ig and HRP-conjugated rabbit anti-mouse Ig (DakoCytomation). Beads coated with Vβ-specific mAb in combination with CD28 were prepared from M-450 Epoxy Dynabeads (Dynal Biotech). A total of 20 μg of Vβ-specific mAb and 10 μg of anti-CD28 was bound to 6 × 107 M-450 Epoxy Dynabeads, according to the protocol of the manufacturer.
Constructs and transfection
The pPKC constructs included the constitutive active (A/E) and the dominant-negative (K/R) forms of the various PKC isotypes cloned into the pEF vector. Using the pEGFP-N1 vector (BD Clontech), various CD3γ-GFP variants were constructed. CD3γWT, -S126A, -S123A, and -LLAA inserts were obtained by PCR using CD3γ sequences from already existing and published constructs (pTβ-CD3γ-Fneo) as templates (31). The PCR products were cut with EcoRI and BamHI, cloned into the EcoRI-BamHI fragment of the expression vector pEGFP-N1, and confirmed by complete DNA sequencing. For transient transfection of JTag cells, 3 μg of pPKC and 1 μg of pEGFP (or pEGFP-CD3γ) were used with the lipid DMRIE-C reagent (Invitrogen Life Technologies), according to the manufacturer. For transient transfection of activated primary human Vβ8+ T cells, Nucleofector technology (Amaxa Biosystems) was used. A total of 1 × 106 Vβ8+ T cells was mixed with Nucleofector solution from the Human T Cell Nucleofector kit (Amaxa Biosystems) and 3 μg of pPKC and 1 μg of pEGFP. The program T-14 was used for transfection. The transfected cells were transferred to C medium and incubated for 24 h at 37°C, 5% CO2.
Cell stimulation and TCR down-regulation
Jurkat and mouse lymph node T cells were washed twice in C medium and adjusted to 0.5 × 106 cells/ml. A total of 20 nM PDB (Sigma-Aldrich) or CD3-CD28 expander beads (Dynal Biotech) at a cell:bead ratio of 1:2 for Jurkat cells and 1:1 for mouse T cells was added to the cell suspensions and incubated for the time indicated at 37°C, 5% CO2. The beads were removed, and the cells were stained with PE-UCHT1 in case of Jurkat cells and with H57-597 plus FITC-conjugated goat anti-Armenian hamster IgG in the case of mouse T cells. For Staphylococcus aureus enterotoxin E (SEE) superantigen (Sigma-Aldrich) stimulation of human primary Vβ8+ T cells, Raji cells were either unpulsed or pulsed with 2 ng/ml SEE and mixed with the Jurkat cells at a ratio of 2:1. The cells were incubated for 20 min at 37°C, 5% CO2, and stained with PE-BV8. The mean fluorescence intensity (MFI) of the stained TCR was determined by flow cytometry and used to calculate the percentage of TCR down-regulation as (1 − (MFI2 ng/ml SEE/MFI0 ng/ml SEE))× 100%. As control, cells were stained for surface expression of MHC class I (MHC-I).
TCR recycling and confocal microscopy analyses
JTag cells were cotransfected with pEGFP and either pEF, pPKCαA/E, pPKCθA/E, or pPKCθmyr. Forty-eight hours posttransfection, 1 × 106 transfected cells were washed twice and resuspended in C medium. The cell suspension was incubated for 2 h at 12°C with PE-UCHT1, and the temperature was then switched to 37°C. At the time points indicated, cell samples were transferred to a precooled solution of FACS medium (PBS with 2% FCS, 0.1% NaN3). The cells were washed in FACS medium and subjected to flow cytometric analysis. Confocal microscopy analyses were performed on a Zeiss LSM510 connected to a Zeiss Axiovert 100 M microscope (Carl Zeiss). GFP and Cy5 fluorescence were detected using band pass filter BP 505–550 and long pass filter LP 650, respectively.
Biotinylation and TCR degradation
Cells were washed twice in PBS and resuspended in 0.5 mg/ml sulfosuccinimidyl-6-(biotinamido)hexanoate/PBS (Pierce). The cells were incubated for 30 min on ice, washed, and resuspended in C medium to a concentration of 1 × 106 cells/ml. The biotinylated cells (4 × 106 cells/sample) were stimulated for 6 h with anti-TCR mAb or PDB and lysed in lysis buffer (50 mM Tris-base (pH 7.5), 150 mM NaCl, and 1 mM MgCl2) supplemented with 1 mg/ml Pefabloc SC, 10 μg/ml leupeptin, 10 μg/ml pepstatin, 5 mM EDTA, 10 mM NaF, 1 mM Na3VO4, and 1% Triton X-100. Cell lysates were precleared with protein A-agarose (Kem-En-Tec), and biotinylated proteins were precipitated with streptavidin-agarose beads (Pierce), followed by Western blotting using A452 (anti-CD3ε) or 6B10.2 (anti-TCRζ) Abs.
Small interfering RNA (siRNA) and comodulation
A total of 1 × 106 Jurkat cells was cotransfected with 5 μM nonsense-siRNA (silencer negative control 1 siRNA; Ambion), 5 μM PKCα-siRNA (ID301; Ambion), or 2 μM PKCθ-siRNA (ID782 and ID783; Ambion) together with 1 μg of pEGFP and 4 μg of pcDNA6-HA1.7 TCRβ-5 (construct expressing the TCRβ variant Vβ3) (21). The amount of siRNA used was selected to ensure proper knockdown of target mRNA without targeting other PKC isotype mRNAs. For transfection, Nucleofector technology was used with the Cell Line Nucleofector kit V and program G-10 (Amaxa Biosystems). The transfected cells were transferred to C medium at 37°C, 5% CO2. Twenty-four hours posttransfection, the cells were retransfected with siRNA and pcDNA6-HA1.7 TCRβ-5, and incubated for additional 24 h. The cells were washed, adjusted to 0.5 × 106 cells/ml in C medium, and incubated in flat-bottom Nunclon Microwell plates (Invitrogen Life Technologies) for 20 min at 37°C with anti-Vβ3/anti-CD28 or anti-Vβ8/anti-CD28 Ab-conjugated beads at a cell:bead ratio of 1:5 and 1:10, respectively. The beads were removed from the cell samples, and the cells were stained with PE-BV3 and analyzed by flow cytometry. TCR double-positive DO11.10 cells were transfected with pEGFP and dominant-negative pPKCαK/R, pPKCθK/R, or pEF; stimulated with anti-Vβ2/anti-CD28 or anti-Vβ8/anti-CD28 Ab-conjugated beads at a cell:bead ratio of 1:10; and subsequently analyzed, as described above.
Identifying PKCα and PKCθ as the PKC isotypes involved in TCR down-regulation
As a first approach to identify the PKC isotypes involved in TCR down-regulation, we transfected the human T cell line JTag with constructs expressing the constitutive active (A/E) forms of all PKC isotypes identified in T cells (PKCα, -β, -δ, -ε, -ζ, -η, -θ, and -ι) (22). The pPKCA/E constructs were cotransfected with pEGFP to identify cells with high plasmid expression levels. TCR cell surface expression levels were subsequently determined in this cell population by flow cytometry. Of the constitutive active PKC isotypes, only PKCα and PKCθ induced significant TCR down-regulation (Fig. 1,A). Furthermore, a myristylated form of WT PKCθ (PKCθmyr) also induced TCR down-regulation, illustrating the importance of membrane recruitment for activation of this kinase. In addition, a minor effect of the constitutive active PKCβI on TCR down-regulation was observed, and hence, we cannot completely rule out a role of this particular PKC isotype in TCR down-regulation. Cell surface expression of MHC-I was not affected by the various constitutive active PKC isotypes, indicating that the effect of constitutive active PKCα and PKCθ on TCR expression was specific and not due to a general influence on receptor expression (Fig. 1,A). The expression levels of GFP and the various PKCA/E isotypes were comparable (Fig. 1, B and C). Furthermore, the constitutive activity of the various PKCA/E isotypes has previously been confirmed (32, 33, 34), excluding the possibility of artifacts caused by deficient kinase activity.
To further investigate whether both PKCα and PKCθ were involved in TCR down-regulation, the dominant-negative (K/R) forms of these kinases (34) were transfected into JTag cells. The transfected cells were stimulated with CD3-CD28 expander beads or with PDB, and the TCR surface expression levels were determined by flow cytometry. Both the dominant-negative PKCα and PKCθ partially inhibited TCR down-regulation in response to the applied stimuli (Fig. 2, A–E).
We next extended the study to primary human T cells. For this, we purified human PBL from which the Vβ8+ T cells were isolated and subsequently expanded in culture (27). The Vβ8+ T cells were transiently transfected with PKCαA/E, PKCθA/E, PKCαK/R, PKCθK/R, or the empty vector pEF. As seen for JTag cells, PKCαA/E and PKCθA/E induced TCR down-regulation in primary T cells (Fig. 2,F). Furthermore, PKCαK/R and PKCθK/R partially inhibited TCR down-regulation in primary T cells stimulated with SEE-pulsed Raji cells (Fig. 2 F).
To study the role of PKCα and PKCθ in T cells from genetically deficient mice, we next performed TCR down-regulation experiments on T cells from PKC KO mice. Lymph node cells from WT, PKCα, and PKCθ KO mice were isolated and expanded in culture. The cells were then stimulated with CD3-CD28 expander beads or with PDB, and the cell surface expression level of the TCR was determined by flow cytometry. In agreement with the results obtained for transfected cells, we found impaired TCR down-regulation in PKCα and PKCθ KO mice in response to both stimuli (Fig. 2, G–L).
From these experiments, we concluded that both PKCα and PKCθ are involved in down-regulation of the TCR. At first sight, this seemed surprising as PKCα and PKCθ belong to two different PKC subfamilies, and hence depend on partially different cofactors and lipid metabolites for their activation. Possible explanations could be that PKCα and PKCθ use different mechanisms to induce TCR down-regulation or that they are involved in different types of TCR down-regulation.
PKCα and PKCθ both induce TCR down-regulation via the CD3γ diL motif-dependent recycling pathway
Previous studies have demonstrated that CD3γ of the TCR contains a phosphoserine-dependent diL-based receptor-sorting motif involved in TCR down-regulation (31). Following TCR triggering, PKC becomes activated. Activated PKC phosphorylates Ser126 of CD3γ, which makes the diL motif accessible to binding by AP-2. Binding of AP-2 to CD3γ subsequently leads to TCR endocytosis (31, 35, 36). To examine whether both PKCα- and PKCθ-induced TCR down-regulation were dependent on the CD3γ diL motif, we studied Jurkat cells expressing TCR with mutations in the CD3γ chain. We cotransfected JTag cells with constitutive active PKC and various mutated forms of CD3γ-GFP. The TCR surface expression levels of the individual transfected cells were analyzed by flow cytometry (Fig. 3, A, C, and D). Cotransfection of the constitutive active PKCα or PKCθ with CD3γWT-GFP or CD3γS123A-GFP (Fig. 3, A, ▪ and dark gray, and C) led to the same degree of TCR down-regulation as seen in cells with native TCR (Fig. 1,A). In contrast, cotransfection of the constitutive active PKCα or PKCθ with either CD3γLLAA-GFP or CD3γS126A-GFP (Fig. 3, A, □ and light gray, and D) led to an almost complete block in TCR down-regulation. The expression levels of the various PKC isotypes and CD3γ-GFP molecules were similar (Fig. 3,B). Furthermore, all CD3γ-GFP variants were expressed at similar levels on the cell surface (Fig. 3 E). These results indicated that both PKCα- and PKCθ-induced TCR down-regulation depend on the phosphoacceptor Ser126 and the diL-based receptor-sorting motif of the CD3γ chain.
Earlier studies have indicated that TCR down-regulation induced by PKC leads to increased recycling of the TCR (7, 21, 37). Therefore, we investigated whether both PKCα and PKCθ induced TCR down-regulation through the recycling pathway. JTag cells were transfected with constitutive active PKC, and surface-expressed TCR were stained with PE-UCHT1 at 12°C for 2 h. By using this condition, TCR trafficking was blocked and complete labeling of the cell surface-expressed TCR was obtained. The temperature was then shifted to 37°C in the continuous presence of PE-UCHT1. After ∼20 min, the entire pool of cycling TCR was labeled, and at this time point we found no significant difference in the level of labeled TCR in PKCA/E and mock-transfected cells, indicating that both PKCα and PKCθ induced TCR down-regulation by increased internalization and localization of the TCR in the endocytotic recycling compartments (Fig. 4, A–C). Confocal microscopy analyses confirmed that a higher fraction of TCR was found in vesicles in cells transfected with constitutive active PKC compared with mock-transfected cells (Fig. 4, D and E). This supported that PKC activation shifts the ratio between cell surface-expressed TCR and TCR located in endocytotic recycling compartments in favor of the recycling compartments (7). Although these results indicated that TCR degradation did not take place in cells transfected with constitutive active PKC, we further examined whether PKC activation induces TCR degradation. For this, Jurkat cells were surface biotinylated and stimulated for 6 h with either PDB or by TCR triggering that is known to induce dose-dependent TCR degradation (9, 11, 27, 38). Following stimulation, the biotinylated proteins were precipitated and subjected to SDS-PAGE, followed by Western blotting using anti-CD3ε (data not shown) or anti-TCRζ Abs (Fig. 4 F). TCR triggering led to a dose-dependent TCR degradation, as expected, whereas PDB stimulation and hence PKC activation did not induce TCR degradation, confirming the results described above.
Taken together, these results demonstrated that both PKCα and PKCθ induced TCR down-regulation via the CD3γ diL motif-dependent recycling pathway.
Additive effect of PKCα and PKCθ in TCR down-regulation
Up to this point, our study did not explain why T cells make use of two different PKC isotypes in TCR down-regulation. One possibility was that the two PKC isotypes operate in an additive way to induce TCR down-regulation. To investigate this hypothesis, JTag cells were transfected with either PKCαK/R, PKCθK/R, or a combination of both PKCαK/R and PKCθK/R. The transfected cells were stimulated for 20 min with PDB, and the TCR surface expression levels were subsequently determined. Expression of either PKCαK/R or PKCθK/R led to a ∼50% reduction in TCR down-regulation as compared with the control cells transfected with pEF. Simultaneous expression of both PKCαK/R and PKCθK/R caused an additional 50% reduction in TCR down-regulation as compared with cells transfected with only one of the dominant-negative isotypes (Fig. 5,A). The observed additive effect of PKCαK/R and PKCθK/R was not caused by a higher total expression level of PKCK/R in the double-transfected cells (Fig. 5 B). These experiments could suggest that PKCα and PKCθ acted on TCR located in different compartments of the plasma membrane.
PKCθ down-regulates engaged TCR, whereas PKCα down-regulates nonengaged TCR
Triggering of the TCR leads to internalization not only of engaged, but also of nonengaged receptors (16, 17, 18). We have demonstrated recently that down-regulation of nonengaged TCR (TCR comodulation) is dependent on PKC (21). We therefore wanted to investigate the role of PKCα and PKCθ in down-regulation of engaged vs nonengaged TCR. The Jurkat cell line expresses the TCRβ chain Vβ8. To obtain cell lines with two different Vβ chains and reduced PKC isotype-specific activity, we transfected Jurkat cells with constructs expressing Vβ3, GFP, and the dominant-negative forms of PKC. However, this approach did not result in significant expression of Vβ3 and PKC, and we therefore shifted to an experimental design using siRNA. The PKCα- and PKCθ-siRNA sequences used (PRKCA 301 + PRKCQ 783) almost exclusively target the RNA of interest according to the manufacturer. In agreement, we found that PRKCA 301 + PRKCQ 783 specifically targeted PKCα and PKCθ, respectively (Fig. 6,A). A siRNA experiment was also performed using a different PKCθ-siRNA (PRKCQ 782) from which we obtained similar results (data not shown). Cells transfected with Vβ3, GFP, and siRNA were stimulated with anti-Vβ3/anti-CD28 or anti-Vβ8/anti-CD28 Ab-conjugated beads. Following stimulation, the beads were removed and the level of Vβ3 down-regulation was measured in the Vβ3/Vβ8 double-positive cell population by flow cytometry. Stimulation with either Vβ3 beads or Vβ8 beads of cells transfected with nonsense-siRNA led to down-regulation of the Vβ3-TCR complex (Fig. 6 B). This demonstrated that comodulation of the nontriggered Vβ3-TCR took place following stimulation with anti-Vβ8. In contrast, in cells transfected with PKCα-siRNA, only stimulation with Vβ3 beads led to optimal Vβ3-TCR down-regulation, and anti-Vβ8-induced comodulation of Vβ3-TCR was severely inhibited. Furthermore, in cells transfected with PKCθ-siRNA, only stimulation with Vβ8 beads led to Vβ3-TCR down-regulation, whereas anti-Vβ3-induced down-regulation of Vβ3-TCR was inhibited. Collectively, these experiments indicated that PKCθ is involved in down-regulation of engaged TCR, whereas PKCα is implicated in down-regulation of nonengaged TCR.
The differential use of PKC isotypes in down-regulation of engaged and nonengaged TCR would allow T cells to selectively regulate trafficking of these TCR subpopulations during an immune response. To substantiate the results obtained by using siRNA, we next studied a stable transfectant of the mouse T cell line DO11.10 that expresses two distinct TCR comprising either Vβ2 or Vβ8 (21). The double-TCR-positive DO11.10 cells were transfected with pEGFP, and the dominant-negative PKCαK/R, PKCθK/R, or pEF and expression of the transfected PKC isotypes were analyzed by Western blotting (Fig. 6,C). The cells were subsequently stimulated with anti-Vβ2/anti-CD28 or anti-Vβ8/anti-CD28 Ab-conjugated beads. Following stimulation, the beads were removed and the level of Vβ2 down-regulation was determined. Stimulation with either Vβ2 or Vβ8 beads of mock-transfected cells resulted in down-regulation of the Vβ2-TCR complex (Fig. 6 D). This demonstrated that comodulation of the nontriggered Vβ2-TCR took place following stimulation with anti-Vβ8. In contrast, in cells transfected with PKCαK/R, only stimulation with Vβ2 beads led to optimal Vβ2-TCR down-regulation, and anti-Vβ8-induced comodulation of Vβ2-TCR was inhibited. Furthermore, in cells transfected with PKCθK/R, only stimulation with Vβ8 beads led to Vβ2-TCR down-regulation, whereas anti-Vβ2-induced down-regulation of Vβ2-TCR was inhibited. Thus, this experiment supported that PKCθ is involved in down-regulation of engaged TCR, whereas PKCα is implicated in down-regulation of nonengaged TCR.
Despite the large amount of information on the critical role of PKC in TCR down-regulation (6, 7, 35, 39, 40, 41), the identification of the specific PKC isotypes involved in this process still remained to be determined. In the present study, we systematically addressed the function of T cell-expressed PKC isotypes in TCR down-regulation, using in vitro studies with human and mouse T cell lines as well as human and mouse primary T cells. Of the PKC isotypes studied, only the constitutive active forms of PKCα and PKCθ induced significant TCR down-regulation. In line with this, overexpression of dominant-negative PKCα and PKCθ, as well as PKCα and PKCθ gene ablation by siRNA and PKC KO, reduced TCR down-regulation significantly. Thus, our study defined PKCα and PKCθ as the two major PKC isotypes involved in TCR down-regulation. This finding is in agreement with an earlier study showing that PKCα and PKCθ are the main PKC isotypes recruited to the plasma membrane in Jurkat cells within the first 15 min of TCR triggering (42). The kinetics of PKC recruitment described by Szamel et al. (42) correlates with the kinetics of PKC-induced TCR down-regulation in Jurkat cells.
The observation that PKCθ is involved in TCR down-regulation is in accordance with the established role of PKCθ as a regulator of T cell activation. PKCθ is known to translocate to the center of the immunological synapse, where it colocalizes with the TCR. Once activated, PKCθ plays a critical role in TCR signaling (43, 44). In mice lacking PKCθ, T cells are partially unresponsive to TCR triggering and deficient in IL-2 production due to impaired NF-κB, AP-1, and NF-AT activation (28, 45). In addition, PKCθ is required for the development of a robust immune response controlled by Th2 cells (46), and lack of PKCθ leads to induction of T cell anergy (47). PKCα also has critical roles in T cell activation and T cell immunity. Thus, PKCα plays a key role in TCR-induced IFN-γ production and Th1-dependent immune responses (G. Baier, unpublished observations).
We were surprised to find that PKC isotypes belonging to two different PKC subfamilies were involved in TCR down-regulation. PKCα is a member of the conventional PKC subfamily, and PKCθ is a member of the novel PKC subfamily. Both subfamilies contain a binding site for phorbol ester/DAG, but only conventional PKC contain a functional binding site for Ca2+ (48, 49). The activity of PKCα can be modified by the intracellular Ca2+ level of the cell, whereas the activity of PKCθ is not dependent on the Ca2+ level (50). Early tyrosine phosphorylation events induced by TCR triggering recruit PLCγ and PI3K to the membrane. Activated PLCγ generates DAG and inositol 1,4,5-triphosphate that leads to increased intracellular Ca2+ concentrations. Activated PI3K generates second messengers that are capable of activating several PKC isotypes via 3′-phosphoinositide-dependent kinase-1 or related enzymes (51, 52). Thus, the activity of PKCα and PKCθ is differently regulated following TCR triggering. We investigated whether PKCα and PKCθ used different mechanisms to induce TCR down-regulation. We found that both PKCα and PKCθ induced TCR down-regulation by the TCR recycling pathway in which the phosphoacceptor group Ser126 and the diL-based receptor-sorting motif of the CD3γ chain were indispensable. These results were in agreement with previous studies on substrate requirements for PKC during TCR down-regulation in general (31, 35).
Interestingly, we observed an additive effect of PKCα and PKCθ in TCR down-regulation. Previous studies have shown that PKCθ is recruited to the contact zone between the T cell and the APC following TCR triggering, whereas PKCα is recruited to the entire plasma membrane (53, 54). Giving the different membrane localizations of PKCα and PKCθ in response to T cell activation, it was reasonable to consider whether these kinases cooperated in TCR down-regulation of engaged and nonengaged TCR. The theory of TCR comodulation states that down-regulation of nonengaged TCR is initiated through intracellular signaling involving tyrosine kinases and PKC rather than by direct physical interaction between the TCR and MHC/peptide complexes (16, 17, 18, 20). In accordance, we have demonstrated recently that TCR comodulation is dependent on PKC (21). In the present study, we observed that Vβ3-TCR in Vβ3/Vβ8 double-positive T cells was down-regulated following both Vβ3- and Vβ8-TCR triggering, confirming the occurrence of TCR comodulation. By using siRNA to knockdown specific PKC isotypes, we found that PKCα was mainly involved in down-regulation of nonengaged TCR, whereas PKCθ was mainly involved in down-regulation of engaged TCR. To validate this intriguing observation, we examined the effect on TCR down-regulation of dominant-negative forms of PKCα and PKCθ in stable double-TCR-positive DO11.10 cells. By using this model, we confirmed that PKCα was mainly involved in down-regulation of nonengaged TCR, whereas PKCθ was mainly involved in down-regulation of engaged TCR. These results are in agreement with our results obtained with T cells from PKC KO mice. Thus, the reduced TCR down-regulation observed in PKCα KO T cells following TCR triggering most likely represented impaired comodulation of nonengaged TCR, whereas down-regulation of directly triggered TCR was intact. Likewise, the reduced TCR down-regulation observed in PKCθ KO T cells following TCR triggering most likely represented impaired down-regulation of directly triggered TCR, whereas down-regulation of nonengaged TCR was intact in these mice. If directly triggered TCR are down-regulated by PKCθ and nontriggered TCR are down-regulated/comodulated by PKCα, this would also explain why TCR down-regulation was not completely blocked in neither PKCα nor PKCθ KO mice.
Furthermore, we observed that siRNA knockdown of PKCθ or overexpression of the dominant-negative form of PKCθ did not interfere with PKCα-induced down-regulation of nonengaged TCR, suggesting that PKCα-induced TCR down-regulation was independent of PKCθ activity. Likewise, siRNA knockdown of PKCα or overexpression of the dominant-negative form of PKCα did not interfere with PKCθ-induced down-regulation of engaged TCR, suggesting that PKCθ-induced TCR down-regulation was independent of PKCα activity. In a recent study, Alarcón and coworkers (19) showed that two different pathways of internalization are involved in down-regulation of engaged and nonengaged TCR, and that they take place from different compartments of the plasma membrane. Thus, down-regulation of engaged TCR depended upon lipid rafts, whereas down-regulation of nonengaged TCR was independent of rafts and instead mediated by clathrin-coated pits. This might explain why two different PKC isotypes are involved in down-regulation of engaged and nonengaged TCR, and it could be suggested that the key to understand how T cells distinguish between the two pathways of TCR down-regulation might be found in different cellular localizations and requirements for activation of the two PKC isotypes. From the present study and a recent study demonstrating that recycling TCR are transported to the contact site between the T cell and the APC (20), we suggest that the primary role of PKCα in TCR trafficking is to induce endocytosis of nonengaged TCR that subsequently recycle to the contact zone between the T cell and the APC, whereas the primary role of PKCθ is to induce endocytosis of directly triggered TCR at the contact zone (Fig. 7).
We cannot exclude that PKCα- and PKCθ-mediated TCR down-regulation represents a negative feedback mechanism to terminate TCR signaling as described in TCR down-regulation-deficient Cbl KO mice (12). However, the observed activation deficiencies in T cells from PKCα and PKCθ KO mice indicate that PKCα- and PKCθ-mediated TCR down-regulation might be required for optimal TCR signaling. Along this line, mutation in the PKC-dependent diL-based receptor-sorting motif of the CD3γ chain leads to T cell activation deficiency in a mouse model (C. M. Bonefeld, M. C. Haks, B. L. Nielsen, M. W. Nielsen, L. Boding, J. P. Christensen, A. R. Thomsen, P. Krimpenfort, M. von Essen, A. M. Kruisebeek, and C. Geisler, unpublished observations). Collectively, these observations may indicate that deficient PKC/CD3γ-dependent TCR down-regulation results in T cell activation deficiency and subsequent immunosuppression, whereas deficient Cbl-dependent TCR down-regulation results in increased T cell responsiveness and subsequent autoimmunity. One possible explanation is that a lower number of activated TCR signaling complexes is generated in T cells with deficient PKC/CD3γ-dependent TCR down-regulation due to deficient endocytosis and subsequent recruitment of nonengaged TCR to the contact zone between the T cell and the APC. As a consequence, the sustained signaling threshold required for induction of proliferation of naive T cells may not be obtained in such cells. Thus, a role of PKC-mediated TCR down-regulation could be to establish and maintain the critical signaling threshold required for optimal TCR-mediated T cell expansion and differentiation.
The technical help of Bodil Nielsen and Christina Lutz is gratefully acknowledged.
The authors have no financial conflict of interest.
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.
This work was supported in part by grants from the Danish Medical Research Council, the Novo Nordisk Foundation, the Austrian Science Foundation FWF-P16229-B07, the A.P. Møller Foundation for Advancement of Medical Sciences, the Agnes and Poul Friis Foundation, the Danish Medical Association Research Fund, and the Astrid Thaysen Foundation for Basic Medical Sciences. M.v.E. was supported by a Ph.D. scholarship from Faculty of Health Sciences, University of Copenhagen. L.B. was supported by a scholarship from the Novo Nordisk Foundation.
Abbreviations used in this paper: PKC, protein kinase C; DAG, 1,2-diacylglycerol; diL, dileucine; KO, knockout; MFI, mean fluorescence intensity; MHC-I, MHC class I; PDB, phorbol 12,13-dibutyrate; SEE, Staphylococcus aureus enterotoxin E; siRNA, small interfering RNA; WT, wild type.