IL-4/IL-13 Signaling Inhibits the Potential of Early Thymic Progenitors To Commit to the T Cell Lineage

Bone marrow (BM)-derived thymic settling progenitors (TSPs) (1) undergo a maturation process to give rise to a massive number of young thymocytes. Early on, TSPs were considered early T cell lineage progenitors destined to give rise mostly to T cells (2). Later, however, these progenitors were found to give rise to lymphoid and myeloid cells (3, 4) and were referred to as early thymic progenitors (ETPs) to accommodate their multipotent attribute (3). Although the maturation process of ETPs is relatively well defined (5–7), the environmental trigger for ETP commitment remains largely unknown. Recent studies identified ETP subsets that could only differentiate to one specific lineage (8–10). A common feature associated with these “unipotent” subsets is expression of a cytokine receptor. For instance, we have previously reported that the unipotent attribute of an ETP subset identified in the thymus is tied to expression of the IL-13Rα1 chain (9), which is known to associate with IL-4Rα to form a functional IL-4Rα/IL-13α1 heteroreceptor (HR) through which IL-4 and IL-13 can signal (11–13). This HR⁺ ETP subset is restricted to the myeloid lineage and gives rise to CD11b⁺ cells in vitro when cultured on stromal cells and in vivo when injected intrathymically into HR-deficient (HR^−/−) mice (9). However, HR⁺ ETPs do not give rise to T cells in vitro or in vivo upon intrathymic transfer (9). These observations point to a link between the HR and restriction of commitment to the myeloid lineage as the HR offers a responsive element to the thymic environment that could be triggered by IL-4 and IL-13 cytokines. Given that cytokine signaling through the HR has been shown to play a role in the death of neonatal Th1 cells (12), the function of dendritic cells (DCs) (14, 15), and the differentiation of macrophages (13), we postulate that the HR on ETPs plays an active role in their commitment to a specific lineage. Specifically, environmental IL-4 and IL-13 could trigger HR signaling and guide commitment to the myeloid lineage. This indeed proved to be correct because HR⁺ ETPs display an active form of STAT6 transcription factor that plays a critical role in antagonizing Notch1 expression and commitment to the T cell lineage. Interference with Notch1 enacted the myeloid pathway, hence commitment of the ETPs to CD11b myeloid cells. These observations point to a new role that environmental IL-4/IL-13 and their HR plays in ETP maturation, which would impact central tolerance and T cell development.

All animal experiments were done according to protocols approved by the University of Missouri Animal Care and Use Committee. C57BL/6 mice were purchased from The Jackson Laboratory (Bar Harbor, ME). IL-13Rα1^+/+–GFP and IL-13Rα1^−/− C57BL/6 mice were described previously (9). Only female mice were used throughout the study. Animals were typically 6–8 wk old at the time that experiments were performed. All animals were maintained under specific pathogen–free conditions in individually ventilated cages and kept on a 12-h light–dark cycle with access to food and water ad libitum.

Anti-CD3 (145-2C11), anti-CD4 (RM4-5), anti-CD8 (53-6.7), anti-CD25 (7D4), anti-CD44 (IM7), anti-CD45 (30-F11), anti-CD11b (M1/70), anti-CD117 (2B8), anti-CD127 (SB/199), anti-Id3 (S30-778), anti–p-STAT6^Y641 (J71-773.58.11), and anti-Tcf1 (S33-966) Abs were purchased from BD Biosciences (San Jose, CA). Anti-Notch1 Ab (22E5) and anti–p-ERK1/2^T202/Y204 (MILAN8R) were purchased from eBioscience (San Diego, CA). Anti-Hes1 (7H11) and anti-C/EBPα (EP709Y) Abs were from Abcam (Cambridge, MA). Anti–IL-13Rα1 Ab (1G3-A7), produced in our laboratory, was previously described (13).

This kit, which was purchased from Miltenyi Biotec, includes Abs against CD4 (L3T4), CD8α (Ly-2), CD11b (Mac-1), CD11c, CD19, B220 (CD45R), CD49b (DX5), CD105, MHCII⁺, Ter-119⁺, and TCRγ/δ.

Abs were directly conjugated to FITC, PE, PE-Cy5, PE-Cy5.5, PerCP-Cy5.5, PE-Cy7, allophycocyanin, allophycocyanin-Cy7 (or allophycocyanin–eFluor 780), or biotin. Biotinylated Abs were revealed with Streptavidin PE.

Sample analysis used a Beckman Coulter CyAn (Brea, CA), and data were analyzed using FlowJo version 10 (TreeStar). Dead cells were excluded using 7-aminoactinomycin D (7-AAD; EMD Biosciences) or Fixable Viability Dye eFluor 780 (eBioscience).

ETPs were isolated as previously described (9). In brief, thymi were harvested from IL-13Rα1^+/+–GFP or IL-13Rα1^−/− C57BL/6 mice after perfusion with PBS, and CD4⁺ cells were eliminated by MACS using anti-CD4 MicroBeads. ETPs were isolated after depletion of Lin⁺ (CD8α⁺, CD11b⁺, CD11c⁺, CD19⁺, B220⁺, CD49b⁺, CD105⁺, MHCII⁺, Ter-119⁺, TCRγ/δ⁺) thymic cells. HR⁺ ETPs (c-Kit⁺CD44⁺CD25⁻) were sorted from Lin⁻ thymic cells of IL-13Rα1^+/+–GFP reporter mice on the basis of GFP (IL-13Rα1) expression. HR^−/P ETPs represent GFP⁻ cells of Lin⁻c-Kit⁺CD44⁺CD25⁻ thymic cells; these cells have the genetic potential for receptor expression because they are isolated from IL-13Rα1^+/+–GFP reporter mice. HR^−/− ETPs were sorted from Lin⁻ thymic cells of IL-13Rα1^−/− mice on the basis of CD44, c-Kit, and CD25 (c-Kit⁺CD44⁺CD25⁻).

Thymic cells from IL-13Rα1–GFP reporter mice were depleted of Lin⁺ cells, and the HR⁺ double-negative (DN)1c population was sorted as GFP⁺CD44⁺CD25⁻CD24⁺c-Kit^int cells. The HR⁻ DN1c population was sorted as CD44⁺CD25⁻CD24⁺c-Kit^int cells from Lin⁻ thymic cells of IL-13Rα1^−/− mice.

Sorting was performed on a Beckman Coulter MoFlo XDP cell sorter. Cell purity was routinely checked, and only sorts with a purity of ˃95% were used in this study.

OP9 and OP9-DL1 cultures were used as previously described (16), with slight modifications. Briefly, OP9 and OP9-DL1 stromal cells were plated 2 d before initiation of cultures at a concentration of 20,000 cells per milliliter in 24-well plates. Progenitors were added at the indicated cell number per well. IL-7 was used at a final concentration of 1 ng/ml, Flt3 ligand (Flt3L) was used at 5 ng/ml, GM-CSF and IL-4 were used at 10 ng/ml, and IL-13 was used at 20 ng/ml. Under these conditions, lymphoid progeny were evident at days 3–10 of OP9-DL1 cell culture, and myeloid progeny were evident at day 3 of OP9 cell culture.

Total RNA was isolated from gut epithelial cells for IL-13Rα1 and from thymocytes for Tcf-1, and cDNA was made using a RevertAid First Strand cDNA Synthesis Kit (Thermo Scientific), according to the manufacturer’s protocol. cDNA was used as a template to amplify IL-13Rα1 or Tcf-1 using gene-specific primers. The amplified products were then cloned into NotI/ClaI restriction sites of the MSCV–IRES–Thy1.1 (Empty-RV) (17) vector to generate MSCV–IL-13Rα1–IRES–Thy1.1 (HR-RV) and MSCV–Tcf-1–IRES–Thy1.1 (Tcf-1–RV), respectively. The primers used to clone IL-13Rα1 and Tcf-1 were IL-13R1, sense 5′-TAGTAGGCGGCCGCACCATGGCGCGGCCAGCGCTGCTGGGCGAG-3′ and antisense 5′-TAGTAGATCGATCCATCAAGGAGCTGCTTTCTTCAG-3′ and Tcf-1, sense 5′-TAGTAGGCGGCCGCACCATGTACAAAGAGACTGTCTACT-3′ and antisense 5′-AATAGTAGATCGATCCACTAGAGCACTGTCATCGGAAGG-3′.

The constructs containing IL-13Rα1 or Tcf-1 genes were verified by automated sequencing.

Retroviral packaging was performed as described, with slight modifications (17). Briefly, 293FT cells (Invitrogen) were transfected with HR-RV, Tcf-1–RV, or control vector Empty-RV, along with the retroviral packaging vector, by Lipofectamine 2000. After 48 h, viral supernatants were mixed with Polybrene (1 μg/ml) and used to transduce ETPs. Retrovirus-infected ETPs were suspended for 48 h in stimulation DMEM mixture containing 1% penicillin/streptomycin, 20% FCS, l-glutamate (2 mM), IL-3 (10 ng/ml), IL-6 (10 ng/ml), SCF (20 ng/ml), and Flt3L (20 ng/ml) before sorting.

ETPs (DN1a,b population) sorted from HR^−/− mice were transduced with HR-RV or Empty-RV and used in single-cell culture on OP9-DL1 stromal cells. Single cells were dispensed in 96-well plates containing OP9-DL1 stromal cells using a Beckman Coulter MoFlo XDP cell sorter. The culture was supplemented with 1 ng/ml IL-7 and 5 ng/ml Flt3 ligand, with or without 10 ng/ml IL-4 and 20 ng/ml IL-13. After 10 d, the resulting cultures were analyzed for the expression of CD25 T cell lineage marker by CD45⁺ live cells.

Sorted HR⁺ ETPs (DN1c), HR^−/P ETPs, or HR^−/− DN1c thymocytes were used to isolate RNA by TRIzol extraction and isopropanol precipitation. Reverse transcription and DNA amplification were performed on a StepOnePlus system using a Power SYBR Green RNA-to-C_T 1-Step Kit (both from Applied Biosystems), according to the manufacturer’s instructions. RT-PCR was done with primers specific for Tcf-1, sense 5′-CCAGTGTGCACCCTTCCTAT-3′ and antisense 5′-AGCCCCACAGAGAAACTGAA-3′; Hes1, sense 5′-CGGCATTCCAAGCTAGAGAAGG-3′ and antisense 5′-GGTAGGTCATGGCGTTGATCTG-3′; C/EBPα, sense 5′-AGCAACGAGTACCGGGTACG-3′ and antisense 5′-GTTTGGCTTTATCTCGGCTC-3′; Notch1, sense 5′-GGACATGCAGAACAACAAGG-3′ and antisense 5′-CAGTCTCATAGCTGCCCTCA-3′; Id3, sense 5′-AGCTTAGCCAGGTGGAAATCCT-3′ and antisense 5′-TCAGCTGTCTGGATCGGGAG-3′; Deltex1, sense 5′-GAGGATGTGGTTCGGAGGTA-3′ and antisense 5′-CCCTCATAGCCAGATGCTGT-3′; IL-7Rα, sense 5′-AGTCCGATCCATTCCCCATAA-3′ and antisense 5′-ATTCTTGGGTTCTGGAGTTTCG-3′; Rag2, sense 5′-CACATCCACAAGCAGGAAGTACAC-3′ and antisense 5′-GGTTCAGGGACATCTCCTACTAAG-3′; Ptcrα, sense 5′-CTGGCTCCACCCATCACACT-3′ and antisense 5′-TGCCATTGCCAGCTGAGA-3′; CD25, sense 5′-AACCATAGTACCCAGTTGTCGG-3′ and antisense 5′-TCCTAAGCAACGCATATAGACCA-3′; Gata3, sense 5′-GAGGTGGTGTCTGCATTCCAA-3′ and antisense 5′-TTTCACAGCACTAGAGACCCTGTTA-3′; Lat, sense 5′-CTGTTGTCTCCTCTGCTCCTGT-3′ and antisense 5′-CTCACTCTCAGGAACATTCACG-3′; Lck, sense 5′-CTAGTCCGGCTTTATGCAGTG-3′ and antisense 5′-CCGAGGGAGTCTTGAGAAAAT-3′; Egr1, sense 5′-GAGGAGATGATGCTGCTGAG-3′ and antisense 5′-TGCTGCTGCTGCTATTACC-3′; and GAPDH, sense 5′-AACTTTGGCATTGTGGAAGG-3′ and antisense 5′-GGATGCAGGGATGATGTTCT-3′.

Relative transcript abundance was determined by the comparative threshold cycle method using StepOne software (Applied Biosystems) normalization with GAPDH. All samples were run in triplicate.

SCH772984 ERK1/2 inhibitor (Selleck Chemicals) and AS1517499 STAT6 inhibitor (Axon Medchem) were dissolved in 50 μl DMSO/PBS (1/1 v/v) and administered to mice i.p. SCH772984 (12.5 mg/kg) was injected twice a day for 10 d, and AS1517499 (10 mg/kg) was given once a day for 5 d. DMSO/PBS with no inhibitor was used as control. For in vitro inhibition, STAT6 and ERK1/2 inhibitors were used at 200 nM and 10 μM, respectively.

HR⁻ ETPs (Lin⁻CD4⁻CD8⁻CD25⁻CD44⁺c-Kit⁺) transduced with HR-RV were cultured for 48 h in stimulation DMEM mixture and sorted on the basis of IL-13Rα1 and Thy1.1 expression. Transduced (HR⁺) cells were then cultured in the presence or absence of different cytokines (10 ng/ml IL-4, 20 ng/ml IL-13, 10 ng/ml IFN-γ, or 10 ng/ml IL-12) for 1 h (for phosphorylation of intracellular signaling molecules) or for 24 h (for expression of transcription factors). STAT6 and ERK1/2 inhibitors were used as above.

Data were analyzed using an unpaired two-tailed Student t test or one-way ANOVA, as indicated. All statistical analyses were performed using Prism software version 4.0c (GraphPad).

In our previous study, a short-term culture on OP9-DL1 stromal cells indicated that HR⁻ ETPs from HR^+/+ mice, referred to as HR^−/P because of their genetic potential for receptor expression, differentiate into T lineage cells, whereas HR⁺ ETPs do not (9). Because the kinetics of ETP maturation are flexible (3, 18), the question remains whether the HR influences ETP maturation in a time-dependent manner. To test this premise, ETPs were sorted into HR^−/P and HR⁺ ETPs (Supplemental Fig. 1), cocultured on OP9-DL1 cells for extended time periods, and assessed for commitment to the T cell lineage. The results show that, although HR^−/P ETP maturation on OP9-DL1 stromal cells, which support lymphoid differentiation (16), yielded T lineage cells that reached optimal numbers by day 10, HR⁺ ETPs were unable to mature through the T lineage pathway throughout the 10-d culture period (Fig. 1A). The percentage and cell number data compiled from several experiments confirm that HR^−/P ETPs, but not HR⁺ ETPs, give rise to T cells (Fig. 1A). However, culture on OP9 stromal cells, which support myeloid differentiation (16), revealed that HR⁺ ETPs and HR^−/P ETPs commit to the myeloid lineage in a short 3-d time period (Fig. 1B). Also, the findings are reproducible, because similar results were obtained with several repeat experiments (Fig. 1B). Overall, these kinetics studies reveal that the differentiation program of HR⁺ ETPs is fixed to the myeloid lineage, whereas the maturation of HR^−/P ETPs remains flexible, and the progenitors are able to mature along the lymphoid or myeloid lineage pathway.

It has previously been shown that thymic myeloid DCs originate from a DN1c population that unusually expresses intermediate levels of c-Kit (c-Kit^int) and lineage-specific markers (19). It was then suggested that the DN1c population represents thymic seeding precursors for DCs (TSP_DCs) that are devoid of T cell potential (20). Given that DN1c cells have been shown to mature along the T cell lineage pathway (21), it is likely that the DN1c population encompasses precursors with T cell potential (21), as well as TSP_DCs (19, 20). Studies were then performed to determine with which DN1 subset HR⁺ ETPs are associated. The findings demonstrate that HR⁺ ETPs belong only to the DN1c population (Fig. 2A). Given that HR⁺ ETPs are Lin⁻ cells that do not express CD11c or CD8 surface markers, we conclude that they do not represent TSP_DCs (20), especially because the latter are restricted to CD11c CD8α DCs, whereas HR⁺ ETPs give rise to CD11b myeloid cells.

Furthermore, DN1c progenitors from HR^−/− mice express genes essential for T cell lineage commitment (Supplemental Fig. 2) and, upon culture on stromal cells, differentiate to the T cell lineage significantly relative to HR^+/+ DN1c (Fig. 2B), again reconciling the T cell potential previously reported for DN1c cells (21). In contrast, because HR⁺ DN1c progenitors (HR⁺ ETPs) commit to the myeloid lineage (Fig. 1B), but not to the T cell lineage (Fig. 2B), whereas HR^−/− DN1c and HR^−/P ETPs remain flexible and can differentiate to the myeloid (Fig. 1B), as well as the T cell (Fig. 2B) lineage, it is logical to credit the HR with an active role in the commitment of HR⁺ ETPs to the myeloid, but not the T cell, lineage. Perhaps signaling through the HR turns on the myeloid pathway or turns off T cell lineage commitment.

Given that HR⁺ ETPs are fixed toward the myeloid lineage, one would envision that the receptor functions to turn on the myeloid, but not the lymphoid, signaling pathway. Alternatively, signaling through the HR would reinforce blockade of the ETP-differentiation program toward the T cell lineage. Notch1 and IL-7R signaling represent major pathways for commitment of ETPs to the T cell lineage (22–27). We then set up ex vivo experiments to test for expression of Notch1 and IL-7Rα in HR⁺ ETPs in comparison with HR^−/P ETPs. The results show that CD25⁻CD44⁺c-Kit⁺GFP⁺ (HR⁺) ETPs display minimal Notch1 expression in comparison with CD25⁻CD44⁺c-Kit⁺GFP⁻ (HR^−/P) ETPs (Fig. 3A, 3B). Similarly, IL-7Rα expression was significantly lower in HR⁺ ETPs. The reduction in Notch1 and IL-7Rα expression is statistically significant, as indicated by the results compiled from several independent experiments (Fig. 3C). In addition, Notch1 downregulation seems to be critical for inhibition of T cell lineage development, as Flt3L (28) was able to induce IL-7Rα expression in Lin⁻ BM cells but not in HR⁺ ETPs (Fig. 3D). These findings led us to question why HR⁺ ETPs do not express Notch1 and whether it is the cause of inhibition of commitment to the T lineage.

It has previously been shown that, during commitment to the T cell lineage, Notch1 signaling leads to upregulation of Tcf-1 (29) and Hes1 (7) transcription factors. Moreover, Hes1 constrains the expression of C/EBP-α, a transcription factor that promotes myeloid development (7, 30). We then sought to determine whether HR⁺ ETPs would display a reverse expression pattern for these molecules. The findings indicate that Tcf-1 and Hes1 were significantly downregulated at the mRNA and protein levels in HR⁺ ETPs relative to HR^−/P ETPs (Fig. 3E, 3F). In contrast, C/EBP-α is increased in HR⁺ ETPs, which agrees well with the downregulation of its antagonist, Hes1. Also, Notch1 target molecules Ptcrα and Deltex1, as well as Tcf-1–controlled genes Gata3, Lat, and Lck, were downregulated in HR⁺ ETPs (Fig. 3G). Finally, IL-7Rα and CD25, although upregulated in HR^−/P ETPs, remain at minimal levels (Fig. 3G). Altogether, the findings indicate that the signaling pathways and transcription machinery required for commitment to the T cell lineage are shut down in HR⁺ ETPs.

HR^−/− ETPs, unlike HR^+/+ ETPs, are able to commit to the T cell lineage, perhaps because the cells lack HR expression, and Notch1 inhibition is not operative. If this is the case, then restoration of HR expression should nullify maturation toward the T cell lineage. To test this postulate, we set up a series of experiments to assert the involvement of the HR in T cell development. To this end, IL-13Rα1 was cloned into a retroviral vector carrying Thy1.1 marker, the resulting vector (HR-RV) was transduced into HR^−/− ETPs, and the ETPs were assessed for maturation into the T cell, as well as the myeloid, lineage. Initially, we ensured that HR-RV is efficient in transducing ETPs and in driving HR expression. Indeed, HR-RV is able to transduce 46% of HR^−/− ETPs, which is similar to the 58% transduction rate observed with the vector carrying Thy1.1 without IL-13Rα1 (Empty-RV) (Supplemental Fig. 3). More importantly, 96% of the ETPs transduced with HR-RV express HR on the cell surface, as detected by anti–IL-13Rα1 Ab (13). Subsequently, HR-RV–transduced ETPs were cultured on OP9-DL1 or OP9 stromal cells and tested for maturation to lymphoid and myeloid lineages in the presence of IL-4 and IL-13 cytokines. The results show that, in the absence of cytokines, HR-transduced ETPs remain able to commit to the T cell lineage to the same extent as Empty-RV–transduced ETPs (Fig. 4A). This is understandable, because the HR-RV–transduced ETPs, unlike freshly isolated HR⁺ ETPs (from HR^+/+ mice), have never been exposed to IL-4 and IL-13 cytokines. This assumption has been proven correct, because exposure of HR-RV–transduced ETPs to IL-4 or IL-13 reduces maturation to the T cell lineage in a significant manner relative to Empty-RV–transduced ETPs, as determined by cell percentage and number (Fig. 4A). This likely reflects interference with their T cell lineage potential, rather than proliferation or survival, because the total number of live CD45⁺ cells is similar in all OP9-DL1 culture settings (Fig. 4A). Furthermore, the same HR-RV cells cultured on OP9 stromal cells proliferate and mature toward the myeloid lineage, upon addition of IL-4 or IL-13 cytokines, to the same extent as the GM-CSF positive control (Fig. 4B). Note that the cytokines can drive myeloid potential without GM-CSF, perhaps serving as growth factors in the OP9 culture system. Given that HR^−/− ETPs include DN1a, DN1b, and DN1c, all of which can commit to the T cell lineage (Fig. 2B), HR expression and cytokine signaling through the HR restores inhibition of the potential of ETPs to commit to the T cell lineage. The interference of IL-4 and IL-13 cytokine signaling through the HR, with maturation to the T cell lineage, occurs even when HR transduction used HR^−/− DN1a, DN1b ETPs. Indeed, single-cell culture of these ETPs on OP9-DL1 cells shows that T cell lineage potential is inhibited by IL-4 and IL-13 cytokines (Table I). In fact, although 100% of single cells display T cell lineage potential with IL-7+Flt3L, only 13–18% had such a potential when IL-4 and IL-13 were added to the culture. The inhibition of T cell lineage potential by the cytokines is dependent on the HR, because 100% of Empty-RV–transduced HR^−/− DN1a, DN1b ETPs display T cell lineage potential in the presence of either cytokine (Table I). It is possible that the 13–18% of single cells (Table I), as well as the 22–24% of bulk ETPs (Fig. 4A), that displayed T cell lineage in the presence of cytokines is related to lower expression of HR. Altogether, IL-4 and IL-13 signaling through the HR inhibits T cell lineage potential in multipotent ETPs.

Despite the fact that the HR is involved in allergic inflammation (15, 31), differentiation of macrophages (13), death of neonatal Th1 cells (12, 14), and regulation of IL-12 production by DCs (14, 32), our understanding of HR signaling is in its infancy (33), and much less is known about how the receptor affects ETP maturation. By analogy with IL-4R signaling in Th2 cells, we surmised that STAT6 activation may be operative in HR⁺ ETPs. This indeed proved to be correct, because ex vivo phosphorylation of STAT6^Y641 was increased significantly in HR⁺ ETPs compared with HR^−/P ETPs (Fig. 5A). STAT6 activation has been shown to trigger the ERK1/2–Egr1 pathway in other cell types (34); this function would be relevant for HR⁺ ETPs, because Egr1 can upregulate expression of Id3, a factor involved in the repression of Notch1 transcription (35, 36). This indeed was the case, because phosphorylation of ERK1/2^T202/Y204, as well as expression of Egr1 and Id3 transcription factor, was significantly increased in HR⁺ ETPs relative to HR^−/P ETPs (Fig. 5A). Also, the CT value for Notch1 transcription was much higher in HR⁺ ETPs versus HR^−/P ETPs, and the relative mRNA quantity for Notch1 was significantly lower in HR⁺ ETPs relative to HR^−/P ETPs (Fig. 5B). These data suggest that HR-driven STAT6 activation likely triggers downregulation of Notch1 at the transcriptional level.

To ensure that signaling through the HR is responsible for STAT6 activation, downregulation of Notch1 expression, and interference with ETP maturation toward the T cell lineage, HR^−/− ETPs were transduced with HR-RV, treated with cytokine (IL-4 or IL-13), alone or in combination with STAT6 inhibitor, and assessed for phosphorylation of STAT6/ERK1/2 and expression of Id3. The findings indicate that IL-4 or IL-13 alone induces STAT6 and ERK1/2 phosphorylation, as well as Id3 expression and Notch1 downregulation (Fig. 6). However, although the addition of STAT6 inhibitor reduced STAT6 and ERK1/2 phosphorylation, as well as Id3 expression, Notch1 remained optimal (Fig. 6). Compiled results from several experiments indicate that cytokine-induced STAT6 signaling and its consequence on ERK1/2 phosphorylation and Id3 expression are statistically significant (Fig. 6), further confirming that signaling through the HR uses STAT6 activation to downregulate Notch1 expression. In fact, in vitro blockade of STAT6 or ERK1/2 activation in Lin⁻ thymocytes significantly rescues Notch1 surface expression on HR⁺ ETPs (Supplemental Fig. 4). More importantly, inhibition of STAT6 phosphorylation in vivo nullifies activation of its substrate ERK1/2, leading to downregulation of Id3, the master inhibitor of Notch1 transcription (Fig. 7A). Similarly, inhibition of ERK1/2 led to Id3 downregulation without affecting STAT6 phosphorylation, indicating that STAT6-driven Notch1 downregulation operates through the function of ERK1/2 and Id3 transcription factors. In fact, inhibition of STAT6 or ERK1/2 restores Notch1 expression, further confirming the involvement of this pathway in Notch1 downregulation (Fig. 7B). The data are statistically significant, as indicated by the p values obtained from results compiled from several experiments.

Interestingly, in vivo STAT6 inhibition rescues HR⁺ ETP maturation to the T cell lineage, because the percentage of ETP-derived CD25⁺ cells increased from a background level of 4% in the absence of inhibitor to 37% in mice that received STAT6 inhibitor (Fig. 8A). The results are statistically significant, as determined by the cell number compiled from several experiments (Fig. 8A). Because thymocyte progression from ETP to the DN2 (CD25⁺) stage relies on Notch1-mediated induction of its downstream transcriptional target, Tcf-1 transcription factor (29), it is logical to envision that forced expression of Tcf-1 would rescue commitment to the T cell lineage by HR⁺ ETPs. To test this premise, Tcf-1 was cloned into a retroviral vector (Tcf-1–RV) and transduced into HR⁺ ETPs, and the cells were tested for maturation into the T cell lineage upon culture on OP9-DL1 stromal cells. The results show that the percentage of ETP-derived CD25⁺ cells increased from a background level of 4% in the Empty-RV–transduced HR⁺ ETPs to 56% in HR⁺ ETPs recipients of forced Tcf-1 expression (Fig. 8B). This is statistically significant, as determined by the cell number compiled from several experiments (Fig. 8B). Altogether, the HR in ETPs signals Notch1 inhibition and negates commitment to the T cell lineage through activation of STAT6 transcription factor (Fig. 9).

This study demonstrates that environmental IL-4 and IL-13 guide ETP maturation by signaling through their IL-4Rα/IL-13Rα HR. Indeed, ETPs from HR^−/− mice, which arise in an environment in which IL-4 and IL-13 are unable to signal through the HR, commit to the myeloid and lymphoid lineages. In contrast, ETPs from HR^+/+ mice, which arise in an environment in which IL-4 and IL-13 can signal through their HR, commit only to the myeloid lineage. Given that ETPs with a history of IL-7R expression are also restricted to a single lineage choice (8), it is logical to envision that the HR and IL-7R serve as responsive elements to environmental cytokines and influence ETP lineage choice. This study, which demonstrates that signaling through the HR restricts commitment to the T cell lineage, also sheds light on the mechanism by which such a restriction comes about. Indeed, HR⁺ ETPs, which would be physiologically exposed to environmental IL-4/IL-13 in vivo, exhibit an active form of STAT6, whereas their HR⁻ counterparts do not. Interestingly, Notch1, a critical factor for ETP maturation to the T cell lineage (22, 37), was significantly downregulated in HR⁺ ETPs relative to HR⁻ ETPs. These observations pointed to an active blockade of ETP commitment potential to the T cell lineage, which is set off by IL-4/IL-13 interaction with the HR (Fig. 9). As a consequence of STAT6 activation, the ERK1/2-Egr1 pathway is put in motion, leading to induction of Id3, which represses Notch1 transcription, hence inhibition of commitment potential to the T cell lineage (Fig. 9). Also, downregulation of Notch1 frees expression of C/EBPα, a critical factor in ETP commitment to the myeloid lineage (7).

Prior reports indicated that IL-7Rα increases survival and sustains progenitor commitment to lymphoid lineages (8, 38). This function must avoid STAT6 activation to preserve Notch1 expression and elude Id3 transcription. Therefore, it is conceivable that different cytokine receptors use distinct signaling pathways to reinforce lineage fate. This agrees well with an earlier observation indicating that IL-12Rβ sustains commitment of T cell progenitors toward the myeloid lineage (39).

The HR is perhaps tasked to instruct lineage commitment by its ligands, IL-4 and IL-13 cytokines. The puzzle here is where do the cytokines come from and how are they tied to ETP maturation? It has been shown that type 2 innate lymphoid cells (40), which can be generated in the thymus from DN1/DN2 thymocytes (41), produce type II cytokines, including IL-13 (15). Similarly, NKT cells, which also arise in the thymus, can produce IL-4 (42). Thus, it is possible that the type 2 cytokines from these cells signal through the HR to guide ETP commitment. Alternatively, HR signaling could have been initiated during travel of TSPs from the BM to the thymus in conduits where the cytokines would be readily available. Whatever the source of the ligand or the timing of receptor triggering, the findings suggest that environmental, rather than genetic, programs control ETP commitment. Specifically, although clonal distribution of lymphocyte specificity is genetically programmed by random AgR rearrangement, ETP commitment seems to rely on ligand receptor expression and ligand availability. From another perspective, it would be logical to envision that multipotent ETPs (3, 4) would not have been subjected to ligand/receptor interactions, whereas presumed unipotent ETPs (8, 9) have been driven to acquire such an attribute by those environmental factors.

The essence of the findings is that signaling by environmental IL-4/IL-13 through the HR diverts ETP commitment from the T cell lineage to myeloid cells. Because myeloid cells could serve as APCs in T cell selection, the diversion may impact central tolerance and shape the autoimmune repertoire.

The authors have no financial conflicts of interest.