Identification of an Essential Cytoplasmic Region of Interleukin-7 Receptor α Subunit in B-Cell Development

Interleukin-7 (IL-7) is essential for lymphocyte development. To identify the functional subdomains in the cytoplasmic tail of the IL-7 receptor (IL-7R) α chain, here, we constructed a series of IL-7Rα deletion mutants. We found that IL-7Rα-deficient hematopoietic progenitor cells (HPCs) gave rise to B cells both in vitro and in vivo when a wild-type (WT) IL-7Rα chain was introduced; however, no B cells were observed under the same conditions from IL-7Rα-deficient HPCs with introduction of the exogenous IL-7Rα subunit, which lacked the amino acid region at positions 414–441 (d414–441 mutant). Signal transducer and activator of transcription 5 (STAT5) was phosphorylated in cells with the d414–441 mutant, similar to that in WT cells, in response to IL-7 stimulation. In contrast, more truncated STAT5 (tSTAT5) was generated in cells with the d414–441 mutant than in WT cells. Additionally, the introduction of exogenous tSTAT5 blocked B lymphopoiesis but not myeloid cell development from WT HPCs in vivo. These results suggested that amino acids 414–441 in the IL-7Rα chain formed a critical subdomain necessary for the supportive roles of IL-7 in B-cell development.

Activation of signal transduction pathways via IL-7R begins with IL-7 binding to IL-7R complexes, followed by heterodimerization and conformational changes in IL-7Rα and γ c chains [10]. Next, Janus kinase 1 (JAK1) and JAK3 interphosphorylate, inducing kinase activity in both molecules. Activated JAK1 and JAK3 phosphorylate tyrosine residues on their substrates, including other kinases, adaptor molecules, and receptor subunits [11]. Phosphorylated tyrosine residues in the cytoplasmic domains of cytokine receptors can be docking sites for downstream signaling molecules with SH2 domains, such as phosphatidylinositol 3-kinase (PI3K) and signal transducer and activator of transcription 5 (STAT5) [11].
Four tyrosine residues are present in the cytoplasmic domain of mouse IL-7Rα. Among these four tyrosine residues, the third, Y449, is necessary for STAT5 binding to IL-7Rα [12,13]. Because the introduction of a constitutively active form of STAT5 in IL-7Rα −/− hematopoietic stem cells (HSCs) rescues impaired B-cell development [14], STAT5 is considered a critical signaling molecule for IL-7R signaling [12]. Indeed, B-cell development is completely shut down in 7RαYYFY knock-in mice, in which Y449 of IL-7Rα is substituted with phenylalanine (F449). In contrast, the number of early T-cell progenitors in 7RαYYFY knock-in mice is reduced, but the number of mature T cells in the periphery is close to normal. Therefore, other functional subdomains that play a role in the regulation of signal transduction via IL-7R should exist in the cytoplasmic tail of IL-7Rα.
In addition to the tyrosine residues, two motifs, namely Box1 and Box2, which are conserved in a number of cytokine receptors, are also recognized in the IL-7Rα chain [10,15,16]. These two subdomains are known to form binding sites of JAK family kinases. Therefore, if Box1 or Box2 is removed, the mutated cytokine receptors lack the ability to trigger signals upon binding of a cognate cytokine [10,15,16]. The acidic region, which was first identified in IL-2Rβ chain and forms a binding site for Src tyrosine kinases, such as Lck and Fyn, is also found in the IL-7Rα chain [17]. However, no other functional subdomains have been identified in the cytoplasmic tail of the IL-7Rα chain.
Therefore, in the present study, we constructed a series of IL-7Rα-deletion mutants to uncover hidden functional subdomains in the cytoplasmic tail of IL-7Rα chain. Our results provide important insights into B-cell development and IL-7R signal transduction.

Identification of the IL-7Rα Cytoplasmic Regions that Are Necessary for IL-7-Mediated Cell Proliferation
We constructed a series of IL-7Rα deletion mutants to study the roles of different cytoplasmic regions in IL-7R signaling ( Figure 1a). We divided the IL-7Rα cytoplasmic domain into nine segments and made deletion mutants lacking each segment. We also generated one mutant lacking all nine segments (dCyt). Two of the segments were chosen such that we could delete conserved motifs, such as Box1 (d280-307) and Box2 (d308-322) [18]. The d280-307 mutant lacked all amino acid residues between Box1 and Box2. Segments 379-396 and 397-413 were chosen so that only the first (Y390) or second (Y401) tyrosine residue was deleted (d379-396 and d397-413 mutants, respectively). The d414-441 mutant conserved all four tyrosine residues but lacked most of the amino acid residues between the second and third residues. Finally, we deleted the C terminal region of IL-7Rα, which contained the third (Y449) and fourth (Y456) tyrosine residues.
We screened each mutant using two different criteria. The first was the ability to transduce growth signals in an IL-2-dependent mouse T-cell line (CTLL-2 cells). The other was the potential to support B-cell development from hematopoietic progenitor cells (HPCs). In this case, we purified c-Kit + lineage − Sca-1 + (KLS) HPCs from the bone marrow of IL-7Rα −/− mice and introduced wild-type (WT) and mutant IL-7Rα chains using a retroviral system. We used the nonfunctional IL-7Rα YYFY mutant as a negative control and the WT IL-7Rα chain as a positive control. After mutant IL-7Rα chains were introduced into IL-7Rα −/− HPCs, we cultured the IL-7Rα −/− HPCs with exogenous WT or mutant IL-7Rα subunits on OP9 cells in the presence of stem cell factor (SCF), Flt3 ligand, and IL-7 for 6 days. Then, we examined the presence of CD19 + cells after culture.
As was the case with other cytokine receptor subunits, Box1 and Box2 were indispensable for IL-7Rα function based on the IL-7-mediated proliferation of CTLL-2 cells (see d272-279 and d308-322 mutants in Figure 1b). In addition, the d280-307 mutant, which lacked the region between Box1 and Box2, was not functional. Thus, dBox1, d280-307, and dBox2 mutants were not functional, perhaps due to the lack of association with JAK1 [19,20]. The d442-459 mutant was nonfunctional as well, presumably because of the lack of STAT5 binding in the absence of Y449. Although the d323-356 mutant showed lower functionality than the wild-type (WT) IL-7Rα chain, other mutants (i.e., d357-378, d379-396, d397-413, and d414-441) induced cell proliferation at a level comparable to that of the WT IL-7Rα chain (Figure 1b). As was the case with other cytokine receptor subunits, Box1 and Box2 were indispensable for IL-7Rα function based on the IL-7-mediated proliferation of CTLL-2 cells (see d272-279 and d308-322 mutants in Figure 1b). In addition, the d280-307 mutant, which lacked the region between Box1 and Box2, was not functional. Thus, dBox1, d280-307, and dBox2 mutants were not functional, perhaps due to the lack of association with JAK1 [19,20]. The d442-459 mutant was nonfunctional as well, presumably because of the lack of STAT5 binding in the absence of Y449. Although the d323-356 mutant showed lower functionality than the wild-type (WT) IL-7Rα chain, other mutants (i.e., d357-378, d379-396, d397-413, and d414-441) induced cell proliferation at a level comparable to that of the WT IL-7Rα chain (Figure 1b).

The Amino Acids Region from Positions 414 to 441 of the IL-7Rα Subunit Was Necessary for B-Cell Development
There was no discrepancy between the ability to transduce growth signals and the potential to support B-cell development except for with the d414-441 mutant (Figures 1 and 2). Stagnation of Bcell development in IL-7Rα −/− HPCs with the d414-441 mutant occurred at the transition stage from pre-proB (B220 + CD19 − ) to proB (B220 + CD19 + ) cells (Figure 2), where IL-7 stimulation was indispensable for the stage transition [14]. We also examined B-cell differentiation potential in KLS

The Amino Acids Region from Positions 414 to 441 of the IL-7Rα Subunit Was Necessary for B-Cell Development
There was no discrepancy between the ability to transduce growth signals and the potential to support B-cell development except for with the d414-441 mutant (Figures 1 and 2). Stagnation of B-cell development in IL-7Rα −/− HPCs with the d414-441 mutant occurred at the transition stage from pre-proB (B220 + CD19 − ) to proB (B220 + CD19 + ) cells (Figure 2), where IL-7 stimulation was indispensable for the stage transition [14]. We also examined B-cell differentiation potential in KLS cells with the d414-441 mutant in vivo by injecting the cells into 400 rad-irradiated recombinant activating gene 2 (RAG2) −/− mice. We found that no B cells were derived from IL-7Rα −/− HPCs with the d414-441 mutant, as was the case for IL-7Rα −/− HPCs with the nonfunctional IL-7Rα YYFY mutant at 5 weeks after injection ( Figure 3). These results indicated that the amino acid region from positions 414 to 441 of IL-7Rα played a critical role in IL-7-dependent B-lymphocyte development.

The Truncated form of STAT5 (tSTAT5) Was Upregulated in CTLL-2 Cells with the d414-441 Mutant Compared with that in WT Cells after IL-7 Stimulation
The d414-441 mutant could transduce growth signals in CTLL-2 cells as efficiently as WT IL-7Rα ( Figure 1) but did not support B-lymphocyte development (Figures 2 and 3). Because STAT5 plays a critical role in IL-7R signaling, we examined the phosphorylation status of STAT5 in CTLL-2 cells with the d414-441 mutant after IL-7 stimulation. As shown in Figure 4, the level of STAT5 phosphorylation in the d414-441 mutant was comparable to that in WT cells. Moreover, tSTAT5 was

The Truncated form of STAT5 (tSTAT5) Was Upregulated in CTLL-2 Cells with the d414-441 Mutant Compared with that in WT Cells after IL-7 Stimulation
The d414-441 mutant could transduce growth signals in CTLL-2 cells as efficiently as WT IL-7Rα ( Figure 1) but did not support B-lymphocyte development (Figures 2 and 3). Because STAT5 plays a critical role in IL-7R signaling, we examined the phosphorylation status of STAT5 in CTLL-2 cells with the d414-441 mutant after IL-7 stimulation. As shown in Figure 4, the level of STAT5 phosphorylation in the d414-441 mutant was comparable to that in WT cells. Moreover, tSTAT5 was  The d414-441 mutant could transduce growth signals in CTLL-2 cells as efficiently as WT IL-7Rα ( Figure 1) but did not support B-lymphocyte development (Figures 2 and 3). Because STAT5 plays a critical role in IL-7R signaling, we examined the phosphorylation status of STAT5 in CTLL-2 cells with the d414-441 mutant after IL-7 stimulation. As shown in Figure 4, the level of STAT5 phosphorylation in the d414-441 mutant was comparable to that in WT cells. Moreover, tSTAT5 was obviously upregulated in CTLL-2 cells with the d414-441 mutant than in cells expressing WT IL-7Rα ( Figure 4).  tSTAT5 can be generated by partial proteolysis after stimulation with cytokines, including IL-2 and IL-3. We previously demonstrated that tSTAT5 is a dominant-negative form of STAT5 [21]. Therefore, we examined whether B-cell development from HPCs was blocked in the presence of tSTAT5. For this purpose, we purified KLS cells as HPCs from WT mouse bone marrow. After the introduction of tSTAT5 (or control) in WT HPCs using a retroviral system, we injected these cells into RAG2 −/− mice. We then examined spleen cells in RAG2 −/− mice with HPCs with or without tSTAT5 at 5 weeks after injection. We found that B-cell development was severely impaired in the presence of tSTAT5, although the number of Mac-1 + myeloid cells was not changed ( Figure 5). These results suggested that the amino acid region from positions 414 to 441 of the IL-7Rα chain may form a docking site for the molecule, which inhibits the generation of tSTAT5 in cells after IL-7 stimulation.

Discussion
Activation of cytokine receptor signals is triggered by JAKs, resulting in phosphorylation of tyrosine residues of various signal molecules and the cytoplasmic tail of receptor subunits [22]. Therefore, among various protein modifications, phosphorylation is a main driver of signal cascades via cytokine receptors upon cognate ligand binding. However, other protein modifications are necessary for proper cytokine receptor signal transduction. For example, a number of molecules are tSTAT5 can be generated by partial proteolysis after stimulation with cytokines, including IL-2 and IL-3. We previously demonstrated that tSTAT5 is a dominant-negative form of STAT5 [21]. Therefore, we examined whether B-cell development from HPCs was blocked in the presence of tSTAT5. For this purpose, we purified KLS cells as HPCs from WT mouse bone marrow. After the introduction of tSTAT5 (or control) in WT HPCs using a retroviral system, we injected these cells into RAG2 −/− mice. We then examined spleen cells in RAG2 −/− mice with HPCs with or without tSTAT5 at 5 weeks after injection. We found that B-cell development was severely impaired in the presence of tSTAT5, although the number of Mac-1 + myeloid cells was not changed ( Figure 5). These results suggested that the amino acid region from positions 414 to 441 of the IL-7Rα chain may form a docking site for the molecule, which inhibits the generation of tSTAT5 in cells after IL-7 stimulation.  tSTAT5 can be generated by partial proteolysis after stimulation with cytokines, including IL-2 and IL-3. We previously demonstrated that tSTAT5 is a dominant-negative form of STAT5 [21]. Therefore, we examined whether B-cell development from HPCs was blocked in the presence of tSTAT5. For this purpose, we purified KLS cells as HPCs from WT mouse bone marrow. After the introduction of tSTAT5 (or control) in WT HPCs using a retroviral system, we injected these cells into RAG2 −/− mice. We then examined spleen cells in RAG2 −/− mice with HPCs with or without tSTAT5 at 5 weeks after injection. We found that B-cell development was severely impaired in the presence of tSTAT5, although the number of Mac-1 + myeloid cells was not changed ( Figure 5). These results suggested that the amino acid region from positions 414 to 441 of the IL-7Rα chain may form a docking site for the molecule, which inhibits the generation of tSTAT5 in cells after IL-7 stimulation.

Figure 5. Truncated STAT5 (tSTAT5) had inhibitory functions in B-cell development. KLS cells from
WT mice were infected with control or tSTAT5 retroviruses. These cells were injected into 400 radirradiated RAG2 −/− mice. Recipient-derived cells in the spleens of host mice were analyzed at 5 weeks after injection. The numbers in the plots were the means of B and myeloid cell numbers from three mice.

Discussion
Activation of cytokine receptor signals is triggered by JAKs, resulting in phosphorylation of tyrosine residues of various signal molecules and the cytoplasmic tail of receptor subunits [22]. Therefore, among various protein modifications, phosphorylation is a main driver of signal cascades via cytokine receptors upon cognate ligand binding. However, other protein modifications are necessary for proper cytokine receptor signal transduction. For example, a number of molecules are

Discussion
Activation of cytokine receptor signals is triggered by JAKs, resulting in phosphorylation of tyrosine residues of various signal molecules and the cytoplasmic tail of receptor subunits [22]. Therefore, among various protein modifications, phosphorylation is a main driver of signal cascades via cytokine receptors upon cognate ligand binding. However, other protein modifications are necessary for proper cytokine receptor signal transduction. For example, a number of molecules are acetylated in the cytoplasm, playing a role in positive regulation of interferon receptor signals [23]. Acetylated STAT1 and STAT2 exhibit enhanced transcriptional activity upon activation by tyrosine phosphorylation. Recently, we demonstrated that acetylation of JAK1, JAK3, and STAT5 occurs immediately in T cells in an IL-2-dependent manner [21]. This acetylation occurs via CREB binding protein (CBP), which relocates from the nucleus to the cytoplasm upon IL-2 stimulation. Acetylated STAT5 is a target of STAT5 protease, resulting in limited proteolysis and generation of tSTAT5. In this study, we demonstrated that tSTAT5 generation was increased after IL-7 stimulation if the 414-441 region in the IL-7Rα subunit was deleted. Therefore, we propose that the 414-441 region in the IL-7Rα subunit may form a binding site for inhibitors of CBP function. As shown in Figure 6, the 414-441 region contains multiple proline residues, which may serve as a docking site for the SH3 domain, although a conventional PxxP motif is absent. Accordingly, it is difficult to hypothesize which molecules may associate with the 414-441 region just based on the amino acid sequence. acetylated in the cytoplasm, playing a role in positive regulation of interferon receptor signals [23]. Acetylated STAT1 and STAT2 exhibit enhanced transcriptional activity upon activation by tyrosine phosphorylation. Recently, we demonstrated that acetylation of JAK1, JAK3, and STAT5 occurs immediately in T cells in an IL-2-dependent manner [21]. This acetylation occurs via CREB binding protein (CBP), which relocates from the nucleus to the cytoplasm upon IL-2 stimulation. Acetylated STAT5 is a target of STAT5 protease, resulting in limited proteolysis and generation of tSTAT5. In this study, we demonstrated that tSTAT5 generation was increased after IL-7 stimulation if the 414-441 region in the IL-7Rα subunit was deleted. Therefore, we propose that the 414-441 region in the IL-7Rα subunit may form a binding site for inhibitors of CBP function. As shown in Figure 6, the 414-441 region contains multiple proline residues, which may serve as a docking site for the SH3 domain, although a conventional PxxP motif is absent. Accordingly, it is difficult to hypothesize which molecules may associate with the 414-441 region just based on the amino acid sequence. Various deletion mutants and point mutants of cytokine receptor subunits have been generated to identify the functional subdomains in the cytoplasmic tails of receptor subunits [24][25][26]. As a result, multiple signal pathways have been shown to be activated by different regions of the cytoplasmic tails of cytokine receptors. These studies have also shown that proliferation and differentiation signals are independent of one another. Accordingly, the d414-441 IL-7Rα mutant retained the potential to stimulate cell proliferation but lacked the ability to support B-cell development from HPCs. However, it is unclear why there was such a discrepancy despite the observation that STAT5 plays roles in both cell proliferation and support of B-cell development. In B-cell development, the first checkpoint at which IL-7 stimulation is required is the transition from the pre-proB cell population to the proB cell population. Expression of the transcription factor early B-cell factor (EBF) is indispensable for this stage transition [27]. We previously demonstrated that IL-7 stimulation is necessary for upregulation of EBF before entry to the proB cell stage [14]. Moreover, there was a threshold for EBF expression that was sufficient for the transition to the proB cell population from more immature cells. In addition, B-cell progenitors need to be stimulated with IL-7 before the pre-proB cell stage to enable sufficient EBF expression in response to IL-7 for the transition to the proB stage [28]. Therefore, one possible explanation for the discrepancy between proliferation and support of B-cell development by the d414-441 mutant may be related to the sensitivity of the mutant to the strength of STAT5 activity. Transcription levels of EBF could be more sensitive to STAT5 activity than proliferation. Furthermore, IL-7-induced EBF expression is mediated directly by STAT5, whereas cell proliferation is regulated by not only STAT5 but also other signaling components, such as PI3K, which associates with IL-7Rα at the region containing Y449 [29]. Since lack of a hypothetical molecule associated with the 414-441 region of the IL-7Rα chain may diminish IL-7R function in B-cell development, it is possible that the associated molecule plays a role in the regulation of B-cell number in vivo. Further investigations are necessary to determine why IL-7-stimulated cell proliferation and B-cell development have different requirements for the 414-441 region of IL-7Rα.
Notably, tSTAT5 inhibited IL-2-mediated proliferation of CTLL-2 cells, in contrast to IL-7mediated cell growth. In all of our experiments, IL-2 was found to stimulate cell proliferation more strongly than IL-7; IL-7-driven cell proliferation levels which were only approximately 40% of that induced by IL-2. Therefore, cell proliferation in response to IL-2 may be more sensitive to negative * * * * P * * * P * Q * G I L --* P * * Q * I P Q * * S T 414 441 420 430 Various deletion mutants and point mutants of cytokine receptor subunits have been generated to identify the functional subdomains in the cytoplasmic tails of receptor subunits [24][25][26]. As a result, multiple signal pathways have been shown to be activated by different regions of the cytoplasmic tails of cytokine receptors. These studies have also shown that proliferation and differentiation signals are independent of one another. Accordingly, the d414-441 IL-7Rα mutant retained the potential to stimulate cell proliferation but lacked the ability to support B-cell development from HPCs. However, it is unclear why there was such a discrepancy despite the observation that STAT5 plays roles in both cell proliferation and support of B-cell development. In B-cell development, the first checkpoint at which IL-7 stimulation is required is the transition from the pre-proB cell population to the proB cell population. Expression of the transcription factor early B-cell factor (EBF) is indispensable for this stage transition [27]. We previously demonstrated that IL-7 stimulation is necessary for upregulation of EBF before entry to the proB cell stage [14]. Moreover, there was a threshold for EBF expression that was sufficient for the transition to the proB cell population from more immature cells. In addition, B-cell progenitors need to be stimulated with IL-7 before the pre-proB cell stage to enable sufficient EBF expression in response to IL-7 for the transition to the proB stage [28]. Therefore, one possible explanation for the discrepancy between proliferation and support of B-cell development by the d414-441 mutant may be related to the sensitivity of the mutant to the strength of STAT5 activity. Transcription levels of EBF could be more sensitive to STAT5 activity than proliferation. Furthermore, IL-7-induced EBF expression is mediated directly by STAT5, whereas cell proliferation is regulated by not only STAT5 but also other signaling components, such as PI3K, which associates with IL-7Rα at the region containing Y449 [29]. Since lack of a hypothetical molecule associated with the 414-441 region of the IL-7Rα chain may diminish IL-7R function in B-cell development, it is possible that the associated molecule plays a role in the regulation of B-cell number in vivo. Further investigations are necessary to determine why IL-7-stimulated cell proliferation and B-cell development have different requirements for the 414-441 region of IL-7Rα.

Proliferation Assays
CTLL-2 transfectants were maintained in complete medium supplemented with hIL-2. After washing three times with phosphate-buffered saline, 5 × 10 4 cells in complete medium were cultured in each well of a 96-well plate in the presence of IL-7 (10 ng/mL) at 37 • C for 48 h.
[ 3 H]-thymidine (1 µCi) was added to the culture 4 h before harvesting. Cells were harvested with an automatic cell harvester on a glass filter (Harvester 96; TOMTEC, Hamden, CT, USA). Radioactivity was determined USA).
The HPCs used in this paper were KLS cells, in which HSCs were highly enriched [33,34]. For cell surface phenotyping, cells were incubated with normal rat IgG (Sigma, St. Louis, MO, USA) and then fluorescence-or biotin-conjugated antibodies on ice for 20 min. If necessary, cells were further incubated with PE-streptavidin after washing with staining medium (Hanks' Balanced Salt solution (HBSS) with 2% FCS and 0.02% NaN 3 ). FACS analysis was performed using a FACSVantage with the DiVa option equipped with 488-nm argon, 599-nm dye, and 408-nm krypton lasers (BD Bioscience Flow Cytometry Systems) at the FACS facility of Duke University Comprehensive Cancer Center. A FACSAriaIII at the FACS facility of Toho University School of Medicine was also used for cell sorting. In addition, an LSRFortessa X-20 (BD Bioscience) was used for analyses. Data were analyzed with FlowJo software (BD Bioscience). Dead cells were excluded from analyses and sorting as cells showing positive staining with propidium iodide or 7-AAD.

In Vitro and In Vivo Differentiation Assays
In vitro culture of HPCs was performed as described previously [28]. In brief, after retroviral infection, HPCs were cultured on OP9 cells in the presence of IL-7, Flt3 ligand, and SCF for 6 days. In vivo injections were performed as described in Reference [28] as well. In the experiment shown in Figure 3, HPCs were purified from IL-7Rα −/− mice (CD45.2) and were intravenously injected into 400 rad-irradiated RAG2 −/− (CD45.1) mice. For the investigation shown in Figure 5, HPCs were purified from C57Bl/6 mice (CD45.2) and intravenously injected into 400 rad-irradiated RAG2 −/− mice (CD45.1). Funding: This work was supported by the Japan Society for the Promotion of Science Grant-in-Aid for Scientific Research (grant nos. 25460600 and 17K08892) to T.K.; a grant from the TAKEDA Science Foundation (to T.K.); a Strategic Research Foundation Grant-aided Project for Private School at Heisei 26th (S1411015) from the Ministry of Education, Culture, Sports, Science and Technology (to M.K.); and a Grant-in Aid for Private University Research Branding Project from the Ministry of Education, Culture, Sports, Science and Technology (to M.K.).