Involvement of Tyrosine Kinase-2 in Both the IL-12/Th1 and IL-23/Th17 Axes In Vivo

Various combinations of Jak family members selectively associate with cytokine receptors to transmit signals that are involved in various cellular events (1). In the case of tyrosine kinase-2 (Tyk2), it is activated in response to various cytokines, including IFNs, IL-6, IL-10, IL-12, IL-13, and IL-23 (2–7). However, Tyk2 is dispensable for IL-6– and IL-10–mediated signaling in mice (8, 9). We and other investigators reported that Tyk2 is required for IFN-α/β–mediated signals to suppress hematopoietic cell growth but not for the signals that induce antiviral activities (10, 11). Thus, the involvement of Tyk2 in IFN-α/β–mediated signaling is restricted. In contrast, IL-12–mediated signals, especially those for IFN-γ production by T cells, are highly dependent on Tyk2 (8, 9, 12). Because Tyk2 is recruited to IL-12Rβ1, IL-23–mediated signaling also probably involves Tyk2 (7). Consequently, experiments using Tyk2-deficient cells revealed that different levels of dependence on Tyk2 are evident among several cytokines.

IL-12, whose receptor is associated with Tyk2 and Jak2 and mainly activates the transcription factor STAT4, is the lineage-specific cytokine responsible for Th1 generation (5, 13). Phosphorylation of STAT4, in conjunction with signals derived from TCR-mediated stimuli, induces the expression of T-bet, a master transcriptional regulator of IFN-γ–producing Th1 cells (14). Although IL-23 was initially shown to induce the differentiation of Th17 cells, it is now generally accepted that the differentiation of Th17 cells is dependent on TGF-β and IL-6 and that IL-23 is instead involved in the expansion, maintenance, and functional maturation of Th17 cells (15). IL-17 is now believed to play essential roles in the pathogenesis of chronic inflammatory disorders, as well as in the host defenses against various pathogens (15, 16). It is noteworthy that IL-12 and IL-23 have common features. As heterodimeric cytokines, they share the p40 subunit, and their receptors share the IL-12Rβ1 subunit, which associates with Tyk2. Thus, Tyk2 seems to be indispensable for the IFN-γ/Th1 axis but is also involved in the immune responses mediated by IL-17–producing Th17 cells. Because Th1 and Th17 are both mainly involved in the proinflammatory immune responses that are controlled by Tyk2, mutations leading to loss of function of Tyk2 can be expected to lead to striking immunological phenotypes.

Experimental allergic encephalomyelitis (EAE), which is induced by immunization with myelin Ags or by adoptive transfer of myelin-specific CD4 effector cells, is an animal model of multiple sclerosis (17). Notably, recent studies demonstrated that Th17 cells are responsible for the development of EAE (18). Indeed, IL-23p19– and IL-12/IL-23 p40-deficient mice are resistant to EAE (19, 20). Tyk2-deficient mice also show lower clinical scores and inflamed CNS areas in this model (21). Moreover, the involvement of Tyk2 was confirmed by experiments using mice carrying different Tyk2 polymorphisms (22). B10.D1 mice, which express the Tyk2A allele, are resistant to EAE development and can be compensated by one copy of the Tyk2G allele from B10.Q/Ai mice. In addition to the EAE model, mice carrying Tyk2 polymorphisms exhibit susceptibility in a model for collagen-induced arthritis (CIA) (23). B10.Q/Ai mice are highly susceptible to CIA, whereas B10.D1 mice are resistant. These studies suggested that deficiency of Tyk2 results in defined clinical disorders. Recently, a patient with Tyk2 deficiency was reported (24). The patient experienced increased susceptibility to viral and mycobacterial infections, atopic dermatitis, and elevated IgE levels, thereby exhibiting broader and more profound immunological defects than expected from studies of Tyk2-deficient mice. The different dependencies on Tyk2 between human and mice could arise from species specificity. Alternatively, precise analyses of Tyk2-deficient mice may reveal new aspects of Tyk2 functions in vivo.

In the current study, we investigated the in vivo involvement of Tyk2 in several IL-12/IL-23–dependent models of experimental diseases, namely delayed-type hypersensitivity (DTH), imiquimod (IMQ)-induced psoriasis-like skin inflammation, and dextran sulfate sodium (DSS)- or 2,4,6-trinitrobenzene sulfonic acid (TNBS)-induced colitis. Our results indicated a somewhat broader contribution of Tyk2 to immune systems than previously expected and suggested that Tyk2 may represent an important candidate for drug development by targeting both the IL-12/Th1 and IL-23/Th17 axes.

B10.D1-H2q/SgJ (B10.D1) mice bearing the Tyk2A allele and B10.Q/Ai mice with the Tyk2G allele were purchased from The Jackson Laboratory (Bar Harbor, ME) and Taconic Farms (Germantown, NY), respectively. Gene-targeted Tyk2-deficient mice were backcrossed for at least eight generations onto BALB/c mice (8, 25). Mice were kept under specific pathogen-free conditions and provided with food and water ad libitum. All experiments were performed according to the guidelines of the Institutional Animal Care and Use Committee of Hokkaido University and Daiichi-Sankyo.

The proliferation of viable splenocytes after Con A treatment was measured using a WST-8 assay (Cell Counting Kit-8; Dojindo Laboratories, Kumamoto, Japan) (26). Briefly, 10 μl WST-8 solution was added to the cells in each well and incubated for 3 h. Absorbance was measured at a test wavelength of 450 nm, and IL-2 in culture supernatants was measured by specific ELISA (Abcam, Cambridge, U.K.). To quantify IFN-γ and IL-17 production, splenocytes were stimulated with murine IL-2, IL-12, or IL-23 for 48 h or 72 h, and each culture supernatant was measured by specific ELISA (R&D Systems, Minneapolis, MN), according to the manufacturers’ instructions.

Wild-type and Tyk2-deficient splenocytes were positively selected based on CD4 expression with Dynabeads CD4⁺ T cell positive selection (Invitrogen, Carlsbad, CA), and the bead-bound cells were detached using DETACHaBEAD (Invitrogen, Carlsbad, CA). The CD4⁺ T cells were subsequently separated based on CD62L expression with MACS CD62L MicroBeads and MACS separation columns (Miltenyi Biotec, Auburn, CA). The separated naive CD4⁺CD62L⁺ T cells were activated by plate-bound anti-CD3 and soluble anti-CD28 (BD Pharmingen, San Diego, CA) and cultured in the presence of murine IL-6 (50 ng/ml; PeproTech, Rocky Hill, NJ) and human TGF-β (5 ng/ml; R&D Systems) with murine IL-23 (10 ng/ml; R&D Systems) for Th17 cell, IL-12 (1 ng/ml; PeproTech) for Th1 cell, and TGF-β (5 ng/ml) for regulatory T cell (Treg) differentiation (27). After 3 d in culture, cells were restimulated with PMA and ionomycin for 4 h, followed by addition of GolgiPlug (BD Pharmingen). Cells were permeabilized and fixed with Cytofix/Cytoperm (BD Pharmingen), according to the manufacturer’s instructions. Detection of IFN-γ– and IL-17–producing cells was determined by intracellular cytokine staining with anti–IFN-γ–FITC or anti–IL-17–PE (BD Pharmingen). Cells were acquired on a FACSCalibur flow cytometer (BD Biosciences, San Jose, CA) and analyzed using CellQuest (BD Biosciences) and FlowJo (TreeStar, Ashland, OR) software.

Six mice per group were injected s.c. with 250 μg methylated BSA (mBSA) (Sigma-Aldrich, St. Louis, MO) at two sites in the abdomen in a combined total of 100 μl a 1:1 emulsion of CFA (BD Biosciences, San Diego, CA) and saline (28). On day 7 following immunization, the mice were challenged by injection of 50 μl 0.5 mg/ml mBSA in saline into one rear footpad, whereas the other rear footpad received 50 μl PBS. Measurements of footpad swelling were taken at 24 h after challenge, using a thickness gauge (Mitutoyo, Kanagawa, Japan). The magnitude of the DTH responses was determined from differences in footpad thickness between the Ag-and saline-injected footpads.

At 8–11 wk of age, mice received a daily topical dose of 5 mg commercially available IMQ cream (5%) (Beselna Cream; Mochida Pharmaceuticals, Tokyo, Japan) on a side of both ears for 6 consecutive days (29). To score the severity of inflammation of the ear skin, both affected ear thickness and ear tissue weight were measured. At the days indicated, the ear thickness of both ears was measured using the thickness gauge (Mitutoyo) and averaged. Also, after application on 4 consecutive days, ears were collected for quantitative PCR analysis, and ear-draining lymph node cells were collected for FACS analysis. In FACS analysis, cells were stimulated for 4 h by PMA/ionomycin with GolgiPlug and stained with anti-CD4 Ab conjugated with Alexa Fluor 647 (BD Pharmingen) and anti–IFN-γ or anti–IL-17 Ab, as indicated.

Colitis was induced by means of drinking water supplemented with 3% DSS (45,000–50,000 m.w.; MP Biomedicals, Santa Ana, CA), as described previously (30). Control mice were treated in a similar manner with drinking water without DSS. The disease activity index (DAI) and histological score were assessed in accordance with established criteria (30), which combined scores of weight loss, consistency, and bleeding divided by 3, and acute clinical symptoms with diarrhea and/or extremely bloody stools.

TNBS-induced colitis has been described (31). Mice were immunized with 3.5 mg TNBS (Sigma-Aldrich) in 40% ethanol by intrarectal administration. Each experimental group contained six 6–9-wk-old female mice.

RNA was extracted from ears and colons using ISOGEN (Nippon Gene, Tokyo, Japan). Using 5 μg total RNA template, cDNA was prepared using a High Capacity cDNA Reverse Transcription kit (Applied Biosystems, CA). Quantitative real-time PCR analyses of the respective gene, as well as the control GAPDH mRNA transcripts, were carried out using TaqMan Gene Expression assay probe/primer mixture and TaqMan Gene Expression Master Mix. PCR amplification and evaluation were performed using Applied Biosystems 7900HT Fast Real-Time PCR System. The reverse transcription and PCR conditions were in accordance with the manufacturer’s instructions, and PCR was performed for 40 cycles.

Sections from formalin-fixed, paraffin-embedded ears were stained with H&E, and epidermal thickness was measured using a BIOREVO BZ-9000 microscope and measurement module BZ-H1ME (KEYENCE, Tokyo, Japan). Using this software, the epidermal thickness was measured at six positions per section and averaged.

All unpaired data were analyzed by an F test to evaluate the homogeneity of variance. If the variance was homogeneous, the Student t test was applied. If the variance was heterogeneous, the Welch t test was performed. For paired comparisons, the Student paired t test was performed. In other cases, the Wilcoxon rank-sum test, for scoring data, and the Tukey test, for the comparison of Tyk2^+/+, Tyk2^+/−, and Tyk2^−/− mice, were performed. The log-rank test was performed in survival experiments. A value of p < 0.05 was chosen as an indication of statistical significance. A statistical comparison was performed using statistical software (SAS System Release 8.2; SAS Institute, Cary, NC).

To clarify and confirm the roles of Tyk2 in effector T cell functions, we used splenocytes derived from wild-type (Tyk2^+/+), Tyk2-heterodeficient (Tyk2^+/−), and Tyk2-deficient (Tyk2^−/−) mice. These three types of splenocytes exhibited similar responses to the T cell mitogen Con A, resulting in the production of IL-2 and cell growth (Fig. 1A). These cells also showed no significant differences in their IFN-γ production levels after IL-2 stimulation (Fig. 1B). However, the splenocytes from Tyk2^−/− mice did not produce IFN-γ in response to IL-12, whereas those from Tyk2^+/+ and Tyk2^+/− mice did (Fig. 1C). The impaired response to IL-12 was also observed in splenocytes from Tyk2-mutant mice (B10.D1) (Fig. 1D). Notably, the Tyk2-deficient mice used in this study have a BALB/c background, whereas B10.D1 mice have a C57BL/10SnSg background. In a previous report, we demonstrated unresponsiveness to IL-12 using Tyk2-deficient mice with a mixed background of 129/SV and C57BL/6 (8). Thus, the involvement of Tyk2 in IL-12–induced IFN-γ production is general and independent of the genetic background.

IL-12 and IL-23 are heterodimeric cytokines composed of a common p40 subunit and a p35 or p19 subunit, respectively. Unlike IL-12, IL-23 promotes a distinct CD4⁺ T cell phenotype characterized by the production of IL-17, denoted Th17 cells (15). IL-23 enhances Th17 function and survival by acting on differentiated Th17 cells that express the IL-23R (15). Because the IL-12Rβ1 chain is also part of IL-23R and associates with Tyk2, we examined the effect of Tyk2 on IL-23–induced IL-17A production. As shown in Fig. 2A, IL-23–mediated stimuli induced IL-17 production in the splenocytes from Tyk2^+/+ and B10.Q/Ai control mice in dose-dependent manners. The splenocytes from Tyk2^−/− and B10.D1 mice completely failed to produce IL-17 in response to IL-23–mediated stimuli. Notably, the splenocytes from Tyk2^+/− mice exhibited a moderate decrease in IL-17 production after stimulation with all of the tested concentrations of IL-23. Therefore, IL-23–induced IL-17 production is strongly dependent on the Tyk2 protein context and is distinct from the effects of Tyk2 on IL-12–induced IFN-γ production. To further examine the involvement of Tyk2 in Th17 differentiation, we cultured CD4⁺CD62L⁺ naive T cells in the presence of TGF-β and IL-6. The cultures of Tyk2^+/+ cells contained significantly higher proportions of IL-17–producing cells than did those of Tyk2^−/− cells, although both cultures induced IL-17–producing cells (Fig. 2B). The subsequent addition of IL-23 increased the numbers of IL-17–producing cells in the Tyk2^+/+ cell cultures but not in the Tyk2^−/− cell cultures. Therefore, Tyk2 influences Th17 induction or maintenance by interacting with IL-23 signaling. The number of IFN-γ–producing cells by IL-12 stimulation was also affected in Tyk2^−/− cell culture, as we described previously (8). We also examined IL-10–producing cells by intracellular staining in the same situation (data not shown). However, we could not detect any IL-10–producing cells as a result of the lower expression level of IL-10, indicating that IL-10 is not involved in Th17 differentiation in this situation. We further examined whether Tyk2 regulates Treg differentiation by TGF-β–mediated Foxp3 expression. As shown in Fig. 2C, TGF-β–induced differentiation into Foxp3⁺ Tregs was normal in Tyk2^−/− cells, suggesting that Tyk2 is not involved in TGF-β–mediated signaling for Treg differentiation. Taken together, Tyk2 is related to pathogenic Th1 and Th17 but not Treg differentiation and maintenance in vivo.

The data obtained in the above-described in vitro experiments encouraged us to examine Tyk2^−/− mice using IL-12– and/or IL-23–related experimental models. DTH responses are strongly T cell dependent and were reported to be defective in IL-12p40– and IL-23p19–deficient mice (28, 32). To evaluate DTH responses in the absence of Tyk2, we sensitized groups of Tyk2^+/+, Tyk2^+/−, and Tyk2^−/− mice with mBSA in CFA and then elicited DTH responses after 7 d by injection of mBSA or saline (negative control) into the footpad. As shown in Fig. 3, specific footpad swelling was observed in the three types of mBSA-injected mice, but the degrees of swelling paralleled the levels of Tyk2 protein (83.9 ± 9.1 × 10⁻² mm for Tyk2^+/+ mice, 46.0 ± 30.5 × 10⁻² mm for Tyk2^+/− mice, and 18.3 ± 11.5 × 10⁻² mm for Tyk2^−/− mice). Therefore, Tyk2 is crucial for DTH.

Recent studies suggested that IL-23 is functionally involved in the pathogenesis of psoriasis (33, 34). Expression of IL-23 is increased in psoriatic lesional skin, and intradermal injection of IL-23 into mouse skin results in erythema, a mixed inflammatory infiltrate, and epidermal hyperplasia (33). The IL-23/IL-17 axis also plays an important role in the development of IMQ-induced skin inflammation, as another model of psoriasis (29). Therefore, we investigated IMQ-induced skin inflammation in Tyk2-deficient mice. The ear swelling reaction was determined by the ear thickness and ear tissue weight. As shown in Fig. 4A, IMQ treatment induced markedly enhanced ear thickness in Tyk2^+/+ mice; however, a significant reduction in IMQ-induced ear thickness was observed in Tyk2^−/− mice. Similarly, the increase in ear tissue weight caused by IMQ treatment was also reduced in Tyk2^−/− mice compared with Tyk2^+/+ mice (Fig. 4B). Histological analysis also revealed that ears injected with IMQ developed epidermal hyperplasia with inflammatory cellular infiltration (Fig. 4C). We next examined expression levels of the related cytokines in ear skin 24 h after IMQ was applied for 4 consecutive days. As shown in Fig. 4D, expression of IMQ-induced Th17-related cytokines was significantly reduced in Tyk2^−/− mice; interestingly, IL-12 (p35) expression was also reduced in these mice. We further investigated the number of CD4⁺IL-17⁺ or CD4⁺IFN-γ⁺ T cells in draining lymph nodes from Tyk2^+/+ and Tyk2^−/− mice with or without IMQ treatment. As shown in Fig. 4E, the number of both CD4⁺IL-17⁺ and CD4⁺IFN-γ⁺ T cells after IMQ treatment decreased markedly in Tyk2^−/− mice. These results indicated that both Th1 and Th17 cells are involved in IMQ-induced skin inflammation through Tyk2. Therefore, Tyk2 is highly involved in the skin inflammation induced by IMQ treatment.

We further investigated whether Tyk2 plays a critical role in the pathogenesis of colitis, because two IL-12 family members (IL-12 and IL-23) were shown to play central roles in mediating intestinal inflammation (35–37). We first used DSS-induced colitis, a model of human Crohn’s disease. All mice survived until sacrifice on day 7 (i.e., after 7 d of DSS treatment). Tyk2^+/+ and Tyk2^−/− mice that received untreated water did not show any clinical signs (diarrhea, fecal occult blood, perianal bleeding, rectal prolapse, or weight loss) of spontaneous intestinal inflammation. In Tyk2^+/+ mice, DSS treatment produced experimental colitis, as assessed by the DAI, which began to increase on day 2. Tyk2^−/− mice receiving DSS-containing water showed slower (beginning on day 5) and lower disease activity than did Tyk2^+/+ mice (Fig. 5A). The significant reduction in the severity of DSS-induced colitis in Tyk2^−/− mice was confirmed by evaluation of body weight loss and colon shortening. As shown in Fig. 5B, Tyk2^+/+ mice with DSS treatment started to exhibit decreased body weight on day 3 and lost ∼20% of their body weight by day 7. In Tyk2^−/− mice, DSS treatment had no effect on the body weight. Similarly, DSS treatment induced more severe shortening of the colon length in Tyk2^+/+ mice compared with Tyk2^−/− mice (Fig. 5C, 5D). We next examined expression levels of the related cytokines in colon tissue. As shown in Fig. 5E, expression of DSS-induced Th1- or Th17-related cytokines was reduced in colon tissue from Tyk2^−/− mice. Therefore, Tyk2 controls the disease activity in DSS-induced colitis through controlling the Th1/Th17 axis.

The involvement of Tyk2 in experimental colitis was also examined using the haptenating reagent TNBS. TNBS-induced colitis is dependent on T cells and is believed to be a model of human ulcerative colitis (31). TNBS treatment induced massive colitis, and all of the treated Tyk2^+/+ mice died within 3 d (Fig. 6A). In contrast, half of the Tyk2^−/− mice treated with TNBS survived (Fig. 6A), and the body weight of the survivors returned to the normal range after they recovered from diarrhea (Fig. 6B). Therefore, Tyk2 seems to be a key molecule for controlling the development of experimental colitis in mice.

Tyk2 plays essential roles in IL-12– and/or IL-23–mediated signaling. Tyk2 deficiency resulted in impaired IFN-γ secretion by IL-12–stimulated splenocytes, as well as IL-17 secretion by IL-23–stimulated splenocytes. However, Tyk2-deficient splenocytes showed normal responses to Con A and IL-2. In this study, we showed that both Th1 and Th17 cell differentiation from naive CD4⁺ cells are affected in Tyk2-deficient mice. Furthermore, Tyk2 was not involved in Treg differentiation from naive CD4⁺ cells by TGF-β. We also found that Tyk2 participates in the development of a variety of experimental diseases, including DTH-induced footpad thickness, IMQ-induced psoriasis-like skin inflammation, and DSS- or TNBS-induced colitis. Taken together, our data strongly suggested that Tyk2 plays important roles in the development of various immune or inflammatory diseases by controlling the Th1/Th17 axis.

Recently, a patient homozygous for a Tyk2 mutation, which resulted in a premature termination codon and the absence of Tyk2 protein, was identified (24). The signaling defects in this Tyk2-deficient patient led to a complex clinical picture, including hyper-IgE syndrome and susceptibility to multiple infectious pathogens. Moreover, Tyk2 was recently identified as a strong multiple sclerosis-susceptibility gene by a genome-wide association study (38); in addition, Tyk2 polymorphisms were found to be associated with an increased risk for systemic lupus erythematosus (39). In the case of Tyk2-deficient mice, Tyk2 was reported to be a critical genetic regulator of EAE (21, 22), which is a model of multiple sclerosis. Although B10.Q/Ai mice, which express the Tyk2G allele, are susceptible to CIA, a Tyk2A mutation renders B10.D1 mice resistant to CIA (23, 40). These phenotypes caused by Tyk2 deficiency are likely to be mediated, in part, by alterations to the Th1/Th2 ratio and by impairment of Th17 cells. Our in vivo experiments further suggested the possible involvement of Tyk2 in immune diseases, which are mainly related to IL-12 and/or IL-23, in line with the above studies. DTH responses were reported to be defective in IL-12p40– and IL-23p19–deficient mice (28, 32). Skin inflammation model induced by IMQ, TLR7/8 ligands, is a new model for human psoriasis, and IL-23p19– and IL-17A–deficient mice showed lower scores for erythema, scaling, and thickness after IMQ treatment (29). In fact, mRNA expression of IMQ-induced Th17-related cytokines was significantly reduced in Tyk2^−/− mice (Fig. 4D). However, mRNA expression of Th1-related cytokines was also reduced in Tyk2^−/− mice. Furthermore, the number of CD4⁺IL-17⁺ or CD4⁺IFN-γ⁺ T cells in draining lymph nodes after IMQ treatment decreased markedly in Tyk2^−/− mice (Fig. 4E). Therefore, both Th1 and Th17 cells are involved in IMQ-induced skin inflammation through Tyk2. Moreover, the DSS-induced colitis model is a little more complicated. Inflammatory bowel disease is characterized by sustained intestinal mucosa inflammation, which is mainly caused by excessive macrophage activation and Th1 and/or Th17 immune responses. However, it also occurs in SCID mice, which lack lymphocytes (41). Oral DSS activates intestinal macrophages, leading to massive production of inflammatory cytokines and chemokines. Subsequently, a number of lymphocytes are recruited to the inflamed sites, resulting in Th1 and Th17 responses. During this inflammatory process, Tyk2 could regulate the Th1 and Th17 responses, whereas the effects of Tyk2 on macrophages are unclear. Indeed, mRNA expression of DSS-induced Th1- or Th17-related cytokines was reduced in colon tissue from Tyk2^−/− mice (Fig. 5E). One report described that Tyk2-deficient macrophages lack NO production upon stimulation with LPS (42), suggesting the possible involvement of Tyk2 in macrophage functions in vivo. In addition, Tyk2^−/− dendritic cells were reported to be defective in IL-12 and IL-23 production upon stimulation with CpG oligodeoxynucleotide (43). Thus, dendritic cells/macrophages may also play an important role in the pathogenesis of these diseases, because our results suggested that Tyk2 deficiency may affect both pathogenic IFN-γ and IL-17. Further experiments are still required in this regard.

The first in vivo evidence that Jaks are critical for cytokine signaling came from studies of a group of human disorders termed SCID. Jak3 selectively associates with the common γ cytokine receptor chain, and mutations in its receptor are known to cause SCID. Therefore, mutations of Jak3 were sought and found to underlie some cases of autosomal recessive SCID (1, 44). Recently, a patient with Tyk2 deficiency was reported to exhibit impaired immune system functions (24). Regarding mutations in Jaks, the somatic Jak2 valine-to-phenylalanine (V617F) mutation has been identified; it is detected in up to 90% of patients with polycythemia and in a sizeable proportion of patients with other myeloproliferative disorders, such as essential thrombocythemia and idiopathic myelofibrosis (1, 45). Recently, activating point mutations in the JAK1 gene were identified in patients with acute lymphoblastic leukemia, as well as rarely in acute myeloid leukemia patients (46). Since the discovery of the Bcr-Abl kinase inhibitor imatinib, great advances have been made in developing kinase inhibitors with exquisite selectivities and potencies (47). Therefore, Jak inhibitors may confer great therapeutic benefits through disease control in patients with autoimmune diseases and leukemia, which presumably result from high levels of circulating cytokines that signal through Jak enzymes. Our results further suggest that Tyk2 could represent a target molecule for the treatment of immune abnormalities.

We thank Dr. K. Ishibashi (Daiichi-Sankyo) for help in breeding Tyk2^−/− mice.

The authors have no financial conflicts of interest.