Cutting Edge: Requirement for TRAF6 in the Induction of T Cell Anergy¹

Ahallmark of the immune system is its ability to distinguish self from non-self, thereby achieving self-tolerance while retaining the capacity to respond to infectious insult. Peripheral tolerance in CD4⁺ T cells is maintained by several mechanisms including deletion of autoreactive T cells, suppression by regulatory T cells, and functional inactivation or anergy (1, 2). Although it is clear that breakdown of any of these mechanisms can lead to the development of autoimmune disease, the molecular pathways associated with T cell peripheral tolerance are not well understood. Normal T cell activation requires integrated signaling from both TCR and CD28. CD28 costimulation acts by amplifying TCR-mediated signaling events and is thought to lower the threshold required for a productive T cell response (3, 4). In contrast, TCR ligation in the absence of CD28 costimulation results in anergy characterized by hypoproliferation and impaired IL-2 production in response to subsequent stimulation (3, 5). Recently, E3 ubiquitin ligases including Cbl-b, Itch, and Grail have been implicated in the development and maintenance of T cell anergy (6, 7, 8, 9, 10). Anergized T cells express higher levels of these E3 ligases, and T cells deficient in these E3 ligases are resistant to anergy induction. In addition, single deletion of either Cbl-b or Itch results in enhanced susceptibility to or development of autoimmune disease (11, 12, 13). Coordinated activity of these E3 ubiquitin ligases is postulated to down-regulate T cell activation by proteasomal degradation or altered subcellular compartmentalization of T cell signaling components (14).

TNFR-associated factor 6 (TRAF6)³ is an adaptor protein that can act as an E3 ubiquitin ligase to mediate signaling downstream from both TNFR superfamily members as well as IL-1/TLR family members (15, 16, 17, 18). Signaling via TRAF6 results in the activation of transcription factors including NFκB, NFAT, Akt, and MAP kinases (16). Unlike conventional E3 ligases that conjugate K48-linked ubiquitination, TRAF6 mediates K63 ubiquitination resulting in substrate activation as opposed to proteasomal degradation (14, 18). Recently, our group identified a T cell-intrinsic role for TRAF6 in the maintenance of peripheral tolerance (19). T cell-specific deletion of TRAF6 (TRAF6-ΔT) resulted in the resistance of T cells to suppression by control CD4⁺CD25⁺ regulatory T cells and the subsequent development of multiorgan inflammatory disease. TRAF6-ΔT T cells efficiently proliferated in response to stimulation with anti-CD3 alone and exhibited enhanced activation of PI3K and Akt, indicating independence from the normal requirement for CD28 costimulation. Together, these data suggested a negative regulatory role for TRAF6 in costimulation-dependent T cell activation. Given the role of costimulatory signals in preventing the development of anergy both in vitro and in vivo, we investigated the susceptibility of TRAF6-ΔT T cells to anergy induction. We now report that TRAF6 deficiency relieves the requirement for costimulation to elicit productive T cell responses and renders T cells resistant to anergizing signals. Resistance to anergy is correlated with the dysregulation of E3 ubiquitin ligase Cbl-b. Taken together with the previously re-ported role for TRAF6 in rendering T cells susceptible to suppression by CD4⁺CD25⁺ regulatory T cells, these data place TRAF6 at a central point of convergence for the interpretation of tolerizing signals.

TRAF6-ΔT and TRAF6-ΔT OT-II mice have been previously described (19). B6.PL-Thy1a/CyJ (Thy1.1) congenic and CD28^−/− mice were obtained from The Jackson Laboratory. Mouse care and experimental procedures were performed in accordance with protocols from Institutional Animal Care and Use Committees of the University of Pennsylvania (Philadelphia, PA).

The following reagents were purchased from BD Pharmingen: purified anti-CD3 and anti-CD28, CTLA4-Ig, anti-CD3-FITC, anti-CD62L-PE, anti-Vβ5-PE, anti-CD4-PerCP, anti-CD44-allophycocyanin anti-Thy1.2-allophycocyanin, biotinylated anti-CD8, anti-B220, anti-IA^b, anti-Gr1, and an OptEIA ELISA kit for IL-2. Anti-biotin magnetic beads were purchased from Miltenyi Biotec. CFSE was purchased from Molecular Probes. Staphylococcal enterotoxin B (SEB) was purchased from Toxin Technologies, and OVA peptide was purchased from AnaSpec. Anti-Cbl-b was purchased from Santa Cruz Biotechnology and anti-actin was purchased from Sigma-Aldrich.

Single cell suspensions from spleen and lymph nodes were incubated with biotin anti-B220, biotin anti-IA^b, biotin-anti-Gr1, and biotin-anti-CD8 followed by streptavidin microbeads. CD4⁺ T cells were negatively selected after separation from bead-bound cells over a magnetic column. CD4⁺ T cell purity was checked by FACS and typically ranged between 85 and 95%. Where indicated, naive (CD62L^highCD44^low) CD4⁺ T cells were isolated by cell sorting on a FACSDiVa flow cytometer (BD Biosciences).

Cells were cultured in RPMI 1640, 5% FCS, 10 mM HEPES, 5 × 10⁻⁵ M 2-ME, 100 IU/ml penicillin, and 100 μg/ml streptomycin. Purified CD4⁺ T cells were stimulated with the indicated dose of plate-bound anti-CD3. For anergy experiments, purified CD4⁺ T cells were stimulated with anti-CD3 (1 μg/ml) in the presence of irradiated APCs with or without soluble CTLA4-Ig (15 μg/ml). For Cbl-b expression levels, purified T cells were stimulated with plate-bound anti-CD3 (5 μg/ml) and soluble CTLA4-Ig (15 μg/ml).

For OVA tolerance induction, gender- and age-matched control and TRAF6-ΔT OT-II transgenic mice were injected i.v. with either PBS or 500 μg of OVA peptide on days 0 and 3. One week later, splenic and lymph node CD4⁺Vβ5⁺ T cells were isolated and CFSE labeled. T cells were then subcutaneously injected into congenic B6 Thy1.1 recipients followed by immunization with OVA peptide (50 μg) in CFA. Three days later, lymph nodes were collected and Thy1.2 Vβ5⁺ positive cells were analyzed for CFSE dye dilution.

For SEB tolerance induction, control and TRAF6-ΔT mice were injected with 75 μg of SEB. On days 0, 3, and 7 after injection, CD4, Vβ8, and Vβ6 expression was monitored by FACS analysis of PBMC. After 7 days, CD4⁺Vβ8⁺ T cells were purified and stimulated in vitro with the indicated dose of SEB and irradiated APCs.

Cells were washed with ice-cold PBS and lysed in buffer containing 150 mM NaCl, 20 mM HEPES (pH 7.0), 10% glycerol, 0.75% Triton X-100, and protease and phosphatase inhibitors. SDS gel-loading buffer was added to cell lysates and samples were boiled and subjected to SDS-PAGE and Western blot analysis. Relative expression was determined with a Macintosh computer using the public domain NIH Image program (developed at the U.S. National Institutes of Health and available on the internet at http://rsb.info.nih.gov/nih-image/.

TRAF6 deficiency uncouples the normal requirement for CD28 costimulation Previous work in our laboratory demonstrated that TRAF6-ΔT T cells efficiently proliferate in response to stimulation with anti-CD3 alone (19). Because normal T cell activation requires cooperative signals from both TCR and CD28, these findings suggested removal of the requirement for CD28 costimulation. To definitively test this, we crossed TRAF6-ΔT mice with CD28^−/− mice. No abnormalities in T cell development or peripheral T cell populations were observed in TRAF6-ΔT CD28^−/− mice. However, similar to TRAF6-ΔT mice, 8- to 10-wk-old mice doubly deficient for TRAF6 and CD28 exhibited the accumulation of activated (CD69⁺) and effector/memory (CD44^highCD62^low) T cells in vivo (Fig. 1,a). CD28^−/− deficiency in T cells results in profoundly reduced proliferation and a complete defect in IL-2 production following CD3 stimulation (19) and, as expected, sorted naive T cells from control CD28^−/− mice responded poorly to stimulation with anti-CD3 in vitro (Fig. 1, b and c). In contrast, TRAF6-ΔT CD28^−/− T cells proliferated and produced IL-2 at levels similar to those of control CD28-sufficient T cells (Fig. 1, b and c). These data indicate that the absence of TRAF6 restores the ability of CD28^−/− T cells to respond efficiently to CD3 stimulation.

TRAF6-deficient CD4⁺ T cells resist anergy induction both in vitro and in vivo. TRAF6-ΔT T cell independence from CD28 costimulation suggested that TRAF6-ΔT T cells might also be resistant to anergy induction, a normal result of TCR ligation in the absence of costimulation. To test this, we stimulated TRAF6-ΔT and control T cells with anti-CD3 in the presence or absence of soluble CTLA4-Ig and irradiated APCs. T cell responses were measured following secondary stimulation with plate-bound anti-CD3. Although anergy was efficiently induced in control T cells, the costimulatory blockade in TRAF6-ΔT T cells did not impair their ability to proliferate or produce IL-2 (Fig. 2, a and b). Because CD28 signals are known to promote survival in addition to preventing anergy, we next examined the viability of control and TRAF6-ΔT T cells following in vitro stimulation. No differences in survival were observed between control and TRAF6-ΔT T cells (Fig. 2 c), indicating that TRAF6-ΔT T cell resistance to anergy induction was not due to enhanced survival.

To examine T cell anergy induction in vivo, we used an adoptive transfer model with OT-II TCR-transgenic mice (19). Administration of high-dose Ag has been shown to induce a hyporesponsive state in Ag-specific T cells (7, 20). Control OT-II and TRAF6-ΔT OT-II mice were injected with OVA peptide or PBS as a control. Ten days after the initial injection, CD4⁺Vβ5⁺ T cells were isolated, CFSE labeled, and injected into congenic recipients followed by immunization with the OVA peptide in CFA. As expected, T cells from both PBS-treated control and TRAF6-ΔT mice proliferated in response to OVA challenge (Fig. 3,a). This proliferation was abrogated in OVA-tolerized control T cells. In contrast, T cells from OVA-injected TRAF6-ΔT mice maintained their ability to proliferate, indicating the resistance of TRAF6-ΔT T cells in anergy induction in vivo (Fig. 3,a). Consistent with these data, T cells from OVA-tolerized wild-type mice failed to produce IL-2 in response to Ag restimulation in vitro, while OVA tolerization failed to anergize TRAF6-ΔT T cells restimulated under the same conditions (Fig. 3 b).

We next extended our findings to an alternative method of in vivo anergy by examining the role of TRAF6 in superantigen-induced tolerance. Injection of mice with SEB results in the selective expansion of Vβ8⁺ T cells followed by their contraction (21). Remaining Vβ8⁺ T cells are unresponsive to subsequent SEB stimulation and are thus rendered functionally anergic (21). As expected, 7 days after injection with SEB, CD4⁺Vβ8⁺ T cells from control mice exhibited impaired in vitro proliferation in response to stimulation with SEB (Fig. 3,c). In contrast, CD4⁺Vβ8⁺ from SEB-injected TRAF6-ΔT mice proliferated strongly at levels comparable to those of PBS-injected mice (Fig. 3,c). The enhanced proliferation observed by TRAF6-ΔT T cells was not due to an abnormal deletion of Vβ8⁺ T cells, as both control and TRAF6-ΔT T cells underwent similar expansion and contraction phases in response to SEB (Fig. 3 d). Thus, consistent with our in vitro findings, TRAF6 deficiency results in impaired anergy induction in vivo.

Independence from CD28 costimulation and impaired anergy induction in TRAF6-ΔT T cells suggested that the negative regulation of costimulatory pathways might be defective in TRAF6-ΔT T cells. Consistent with this, previous work in our laboratory showed that TRAF6-ΔT T cells exhibit enhanced activation of PI3K and Akt (22). CD28 ligation results in the recruitment of p85, the regulatory subunit of PI3K, and leads to downstream activation of Akt (3). Cbl-b acts as a negative regulator of this pathway by ubiquitinating p85 and preventing its association with CD28 (23). Because Cbl-b^−/− T cells can also respond to CD3 in the absence of CD28 and have recently been shown to resist anergy induction (6, 7), we next examined Cbl-b regulation in TRAF6-ΔT T cells. Surprisingly, while control CD4⁺ T cells stimulated under anergizing conditions strongly up-regulated Cbl-b mRNA and protein expression, TRAF6-ΔT T cells exhibited delayed and reduced expression of Cbl-b, a difference that was maintained at both early and later time points (Fig. 4). These data indicate that the absence of TRAF6 results in improper regulation of Cbl-b. However, expression of other E3 ubiquitin ligases reported to be important in anergy induction, including Itch and Grail, were not altered in TRAF6-ΔT T cells (data not shown).

The data presented here conclusively demonstrate a role for TRAF6 in the induction of T cell anergy. Loss of TRAF6 uncouples the normal requirement for costimulation to elicit functional T cell responses and results in dysregulation of Cbl-b, an E3 ubiquitin ligase required for anergy induction. Cbl-b has been previously shown to negatively regulate costimulatory signals by preventing PI3K association with CD28 (23). Indeed, previous work in our laboratory showed that TRAF6-ΔT T cells exhibit enhanced activation of PI3K and Akt, both signaling mediators downstream from CD28 ligation (19). In addition, T cells deficient for PTEN (phosphatase and tensin homologue deleted on chromosome 10), a negative regulator of PI3K activation, are also resistant to anergy induction, and mice with T cell-specific deletion of PTEN develop lymphoproliferative and autoimmune disease (24, 25). Thus, it is possible that decreased Cbl-b expression in TRAF6-ΔT T cells is responsible for the hyperactivation of PI3K/Akt, which in turn renders TRAF6-ΔT T cells resistant to anergy induction. However, as Cbl-b has also been implicated in phospholipase Cγ and protein kinase Cθ ubiquitination and degradation following anergy induction (6, 7), it will be important to examine the expression patterns of these and other known Cbl-b targets in TRAF6-ΔT T cells. Taken together with our previous finding that TRAF6-ΔT T cells are resistant to suppression by CD4⁺CD25⁺ regulatory T cells, the data presented here suggest that signals initiated by distinct T cell extrinsic factors, such as tolerizing APCs or CD4⁺CD25⁺ regulatory T cells, result in the activation of common molecular signaling components, and place TRAF6 at a central point of convergence for the interpretation of these tolerizing signals.

We thank the University of Pennsylvania Cancer Center Flow Cytometry Core for expert assistance with FACS.

The authors have no financial conflict of interest.