Interleukin 4 regulates phosphorylation of serine 756 in the transactivation domain of Stat6. Roles for multiple phosphorylation sites and Stat6 function.

Lymphokines interleukin-4 (IL4) and IL13 exert overlapping biological activities via the shared use of the IL4 receptor alpha-chain and signal transducer and activator of transcription 6 (Stat6). Stat6 is critical for T-helper 2 cell differentiation, B-cell Ig class switch, and allergic diseases; thus, understanding its regulation is of central importance. Phosphorylation is crucial for Stat activity. Whereas Stat6 is phosphorylated on Tyr(641), less is known about serine or threonine. We demonstrate in primary human T-cells (>95% CD3+) that IL4 and for the first time IL13 induce Stat6 serine but not threonine phosphorylation that closely paralleled early IL4 receptor alpha-chain activation (10 min). Stat6 uniquely fails to share a positionally conserved Stat serine phosphorylation sequence; however, known phosphoacceptor sites are proline-flanked. Alanine substitutions of these conserved residues revealed that the transactivation domain, which localized Ser(756) but not Ser(827) or Ser(176), is the IL4-regulated site based on phosphoamino acid analysis. Tyr(641) was dispensable for IL4-mediated serine phosphorylation, suggesting that dimerization is not preconditional. Only Stat6 Y641F variant showed a significant effect on IL4-inducible Cepsilon DNA-binding and reporter gene expression. Lastly, recent work has shown that protein phosphatase 2A negatively regulates Stat6. We propose this target residue(s) is distinct from Ser(756) and may be proximal to Tyr(641) at Thr(645), a residue conserved only among Stat6 members. The phosphomimic variants T645E or T645D ablated Stat6 activation, whereas polar uncharged substitutions (Gln or Asn) and additional mutants (Ala, Val, or Phe) showed no effect. These findings suggest that Stat6 has mechanisms of regulation distinct from other Stats.

IL4 1 and IL13 are homologous pleiotropic lymphokines secreted by antigen activated T-cells that act on cells of hematopoietic, endothelial, dendritic, osteoblastic, and fibroblastic origins and are critical for driving T-helper 2 cell differentiation (1,2). T-helper 2 cell overproduction is commonplace during allergic maladies including asthma, allergic rhinitis, and atopic dermatitis via their ability to activate and recruit monocytes, basophils, mast cells, and eosinophils (1,3). The overlapping biological effects of IL4 and IL13 are probably derived from their shared use of the IL4 receptor ␣ (IL4R␣) signaling chain and their ability to activate one or more members of the Janus tyrosine kinases, which subsequently result in the activation of secondary effector molecules, including signal transducer and activator of transcription 6 (Stat) (2,4,5).
Stat6, originally cloned as IL-4-activated transcription factor (4), is responsible for immunoglobulin class switch and positive and negative gene expression in lymphocytes and for promoting the aforementioned pathologies based on studies from mice made Stat6-deficient through homologous recombination (6 -9). Stat6 is one of seven Stat family members that is postulated to be recruited to newly phosphorylated tyrosine residues within activated receptors in the cytoplasm via their Src homology 2 domains, subsequently becoming tyrosine phosphorylated by Janus kinase or Src enzymes on a single and positionally conserved residue that is believed critical for Stat dimerization, nuclear localization, and gene transcription (10,11). However, recent evidence questions this model by proffering that Stats exist as preformed dimers in the absence of tyrosine phosphorylation (12). Several Stats also have been shown to be serine-phosphorylated in response to cytokine stimulation (13). Mapped Stat phosphoserine sites are localized within proline-rich motifs within the transactivation domain. For Stats 1, 3, 4, 5a, and 5b, this site is positionally conserved, whereas substitution/deletion of this residue/region can affect gene transcription (14 -21). Stat6 does not share a positionally conserved phosphoserine acceptor site with other Stat family members; however, it has been shown to be serine-phosphorylated in Ramos Burkitt's lymphoma B-cells (22) and murine splenic B-cells (23) in response to IL4, but the identity of this site(s) has remained elusive.
Various effector molecules also have been identified to neg-atively regulate Stat6 activity including Ship-1, proteasome, and suppressor of cytokine signaling-1 (24 -26). Whether serine phosphorylation of Stat6 positively or negatively regulates its activity is not clear. Earlier studies have reported that the C-terminal truncation of Stat6, which harbors two prolineflanked serine residues, ablates its transactivation potential (27)(28)(29). Recent work by Woetmann et al. (30) also has shown that an unknown phosphoserine or threonine residue can negatively regulate Stat6 activity because calyculin A inhibition of the serine-threonine protein phosphatase 2A (PP2A) disrupts Stat6 function. Coriticosteroids also can negatively affect Stat6 signals that may be mediated in part through tyrosine phosphorylation (31).
To identify these putative phosphoacceptor residues, Stat6 variants were generated against evolutionarily conserved serine residues flanked by prolines and then assessed by metabolic labeling and phosphoamino acid analysis. Herein, we identify the IL4-inducible phosphorylation site and characterize its ability to bind DNA and regulate gene transcription. The first evidence that IL13 modulates Stat6 serine phosphorylation in primary human T-cells also is shown. Lastly, we propose that the calyculin A-sensitive residue is distinct from the IL4inducible serine site and, alternatively, that PP2A may target Thr 645 with its phosphomimic ability to suppress Stat6 function.

Cell Culture and Transfections
Human T-lymphocytes obtained from normal donors were purified as described previously (32) and maintained in RPMI 1640 medium containing 10% fetal calf serum, 2 mM L-glutamine, and penicillin-streptomycin (50 IU/ml and 50 g/ml, respectively) at 37°C with 5% CO 2 . T-lymphocytes were activated for 72 h with phytohemagglutinin (PHA) (1 g/ml) and were subsequently made quiescent by washing and incubating for 24 h in RPMI 1640 medium containing 1% fetal calf serum before exposure to cytokines. Cells were stimulated with recombinant human IL4 or IL13 (catalog number 200-13, Pepro Tech or catalog number 213-IL, R&D system,) at 37°C as indicated below. HEK293 and COS-7 cells obtained from ATCC were grown in the same medium in the absence of PHA. Approximately 1 ϫ 10 8 T-cells or 1 ϫ 10 7 transfected cells were then stimulated with medium or 100 nM IL4 or IL13 at 37°C as indicated in the corresponding figure legends. Cell pellets were frozen at Ϫ70°C until use.

Flow Cytometry
PHA-activated T-lymphocytes were stained with fluorescein isothiocyanate-labeled clone HIT3a (anti-CD3), anti-CD56 for Natural Killer cells, anti-CD14 for monocytes, and phosphatidylethanolamine-labeled anti-CD19 for B-lymphocytes. Antibodies were purchased from BD Biosciences as described previously (32). Cells were analyzed with a FAC-Scan flow cytometer (BD Biosciences).

Solubilization of Membrane Proteins and Immunoprecipitation
Cells were solubilized in 1% Triton X-100 lysis buffer (10 8 cells/ml) and clarified by centrifugation as described previously (33). Supernatants were incubated with 5 l/ml polyclonal rabbit antiserum raised against peptides derived from the extreme C termini of murine forms of Stat6 (catalog number AX56, Advantex Bioreagents, Conroe, TX) or monoclonal anti-His 6 antibody (catalog number MMs-156R, BaBCO). Blots were Western blotted with monoclonal mouse antiphosphotyrosine (catalog number 05-321, UBI, 4G10), anti-Stat6, IL4R␣ (R&D Systems) or anti-His 6 antibody at 1:1000 as indicated previously (32). For serine-phosphatase inhibitor experiments, cells were preincubated with ethanol as mock control or varying concentrations of calyculin A (catalog number C-5552, Sigma) as described in the figure legends. For all of the samples, total protein was determined by BCA method (Pierce).

Stat6 Variants and Transfections
Stat6 human clone and empty vector pcDNA3.1/GS were purchased from Invitrogen. Mutants of Stat6 were prepared using the QuikChange site-directed mutagenesis kit (catalog number 200518, Stratagene, La Jolla, CA) with oligonucleotide primers designed to alter serine residues to alanines, tyrosine to phenylalanine, or threonine to one of several amino acids. The following mutants of human Stat6 were generated: S176A (AGT to GCC); S756A (AGC to GCC); S827A (TCC to GCC); Y641F (TAT to TTT); and Thr 645 (ACC) to Glu (GAA), Ala (GCC), Gln (CAA), Val (GTA), Phe (TTC), Asp (GAC), or Asn (AAT). Before use, the DNA sequence of each mutant was verified. Transfections were performed in either HEK293 or COS-7 cells by FuGENE 6 transfection reagent (catalog number 1-814-443, Roche Applied Science) using 2 g of the Stat6 plasmids/subconfluent COS-7 or HEK293 in 100-mm dishes, and after 48 h, cells were stimulated with 100 nM recombinant human IL4 for 20 min at 37°C and then immunoprecipitated as described above or subjected to luciferase assay.
Luciferase and ␤-Galactosidase Assays-HEK293 cells were transfected with 0.5 g of Stat6 wild type (WT) or variant plasmids, 1 g of a triple repeat of the C⑀ gene promoter linked to the pGL3 luciferase reporter vector (Promega, Madison, WI), and 0.1 g of the pCH110 plasmid containing the ␤-galactosidase gene as described previously (18). After 32 h, cells were stimulated with 1 nM recombinant human IL4 for 16 h at 37°C. Luciferase and ␤-galactosidase activities were determined using the Dual-Light kit according to the manufacturer's instruction (catalog number BD100LP, Tropix). To correct for differences in transfection efficiencies, luciferase activities were normalized to the ␤-galactosidase values in each individual sample. The results presented are representative data minimally from three independent experiments performed in triplicate.

Electrophoretic Mobility Shift Assay
Transfected cells expressing Stat6 WT or variants were subsequently treated without or with 100 nM human IL4 for 15 min and pelleted by centrifugation, and nuclear extracts were isolated and stored at Ϫ70°C as described previously (34). Nuclear extracts (5 g) were reacted with a Stat6 DNA binding element C⑀ that had been labeled with [␣-32 P]ATP (4). For supershift assays, nuclear extracts were preincubated with 1 g of mouse isotype control or His 6 antisera at 4°C for 1 h. Samples were then incubated with probe for an additional 15 min at room temperature. The DNA-protein complexes were resolved on a 5% polyacrylamide gel containing 0.25ϫ Tris borate EDTA that was prerun in 0.25ϫ Tris borate EDTA buffer for 1 h at 100 V. After the loading of samples, gels were run at room temperature for 2 h at 150 V. Gels were then dried by heating under vacuum and exposed to film at Ϫ70°C.

[ 32 P]Orthophosphate Labeling and Phosphoamino Acid Analysis
PHA-activated T-cells or transfected cells were metabolically labeled with 1 mCi/ml [ 32 P]orthophosphate (ICN Radiochemicals) for 2 h at 37°C, stimulated with 100 nM human IL4 or IL13 for 20 min, and then lysed and immunoprecipitated as described previously (35). Labeled proteins were visualized by autoradiography, and bands corresponding to Stat6 proteins were excised and exposed to limited hydrolysis in 6 N HCl at 110°C for 60 min. Samples were then dried, resuspended in water with phosphoamino acid standards, and spotted onto a thin layer cellulose acetate gel. One-dimensional thin layer electrophoresis was performed at 1500 V for 40 min in buffer containing pyridine:acetic acid:water at a 10:100:1890 ratio. Standards were visualized with ninhydrin, and samples were analyzed by autoradiography as described previously (35).

Mass Spectrometry
A HEK293 stable cell line expressing the WT Stat6 plasmid was stimulated with 200 nM CA for 40 min at 37°C, and Stat6 was immunocaptured and separated on SDS-PAGE as described above. The Coomassie Blue-stained protein was extracted and digested with trypsin, and peptides were isolated as described previously (36). The identification of phosphorylated peptides was performed next on a PE Sciex (Foster City, CA) API 3000 tandem quadrupole mass spectrometer equipped with a Protana (Odense, Denmark) nanoelectrospray source. The samples were dissolved in an aqueous solution of 50% methanol and 1% formic acid, and 2-3 l were deposited in the gold/palladiumcoated glass nanoelectrospray capillaries. The samples were analyzed for phosphorylated peptide identification using negative ion precursor ion scanning for m/z 79.1 with nitrogen as the collision gas and collision energy of 100 eV. Positive ion full-scanning spectra were then recorded utilizing the first quadrupole as the mass filter and a cone potential of 70 V. Product ion spectra of the appropriate positively charged precursor ions then were recorded to identify the site of phosphorylation with nitrogen as the collision gas and collision energies of 20 -40 eV (36).

IL4 and IL13 Induce Tyrosine and Serine Phosphorylation of Stat6 in PHA-activated Human T-cells-To
investigate the biological regulation of Stat6 driven by IL4 or IL13 in T-lymphocytes, primary human T-cells were isolated and activated for 72 h with PHA, stained, and subjected to fluorescence-activated cell sorter analysis for contaminating Natural Killer cells (CD56), B-cells (CD19), and monocytes (CD14). T-cells (CD3ϩ) represented 95% of this cell population (Fig. 1A). Because Stat6 may be recruited to the IL4R␣ via three phosphorylated tyrosine residues (37-39), we first examined the receptor tyrosine phosphorylation driven by IL4 and IL13 from 0 to 60 min to determine the time point for maximal phosphorylation. The immunoprecipitated receptor was subsequently immunoblotted with antiphosphotyrosine antibodies. The IL4R␣ attained maximal tyrosine phosphorylation levels by either cytokine that peaked at 5-10 min (Fig. 1B, lanes b and c and lanes  h and i). In comparison, Stat6 recruitment and tyrosine phosphorylation of Tyr 641 showed a slightly protracted phosphorylation profile with detectable phosphorylation observed within 2.5 min and measured over the entire 40-min time course. (Fig.  1C) (27,28). IL4 was more robust in its ability to induce tyrosine phosphorylation of both effector molecules against equally loaded samples (Fig. 1, B and C, lower panels).
Previous studies have shown that IL4 induced Stat6 serine phosphorylation in Ramos Burkitt's lymphoma B cells (22) and murine splenic B-cells (23). To investigate whether IL4 or IL13 is competent to induce a Stat6 serine phosphorylation in primary human T-cells, cells were radiolabeled with [ 32 P]orthophosphate. Given the above activation kinetics of the IL4R␣ and Stat6, cells were stimulated with either IL4 or IL13 for 20 min. Cell lysates were immunoprecipitated with Stat6 antibodies and subjected to protein separation. A single Coomassie Blue-stainable protein band was observed in each lane with an apparent molecular mass of 113 kDa, which corresponded with the autoradiograph (Fig. 1D, lower panel). Each band was excised and subjected to phosphoamino acid analysis. Autoradiography revealed that the Stat6 protein showed radiolabeled phosphate incorporation into tyrosine and serine, but not threonine residues, in response to IL4 or IL13 stimulation of primary T-cells (Fig. 1D, lanes b and c).
Effect of Alanine Substitution of Serines 176, 756, and 827 or Tyrosine 641 of Stat6 as Judged by Phosphoamino Acid Analysis-All of the growth factor-mediated Stat serine phosphorylation sites mapped to date are flanked by one or more proline residues (11). To investigate the location of the Stat6 serine phosphorylation site(s), we identified only three evolutionarily conserved residues that met the criteria. This included Ser 176 found in the helical coiled-coiled domain, which promotes protein-protein interaction, and two that localized to the transactivation domain (Ser 756 and Ser 827 ) as depicted in Fig. 2A. To determine whether these sites were the putative IL4-IL13inducible serine sites, HEK293 or COS-7 cells that express functional IL4 receptors but low to undetectable levels of Stat6 were transiently transfected with expression plasmids for WT or mutant forms of Stat6. Cells were metabolically labeled with [ 32 P]orthophosphate and stimulated with or without 100 nM IL4 and immunopurified Stat6 via epitope capture, separated by SDS-PAGE, and visualized by autoradiography (Fig. 2, B  and C, lower panels). The radiolabeled proteins were then excised and subjected to phosphoamino acid analysis after limited acid hydrolysis (Fig. 2, B and C, upper panels, lanes a-j). inhibition of serine phosphorylation that was reduced to background levels in both cell types while tyrosine phosphorylation was not affected. It is concluded from these studies that Ser 756 is the IL4-regulated Stat6 serine phosphorylation site. Stat6 did show some variability in COS-7 versus HEK293 cells since IL4-inducible threonine phosphorylation was occasionally but not reproducibly detected (panel C, lanes a and b). Lastly, the mutation of the conserved Y641F retained the ability to become serine-phosphorylated in both cell lines, suggesting that tyrosine and serine phosphorylation events can be mutually exclusive.
Analysis of DNA Binding Activities of Stat6 Mutants-The ability of wild type protein and Stat6 mutants to bind to an oligonucleotide probe was evaluated (Fig. 3). Protein nuclear extracts isolated from IL4-stimulated HEK293-Stat6-transfected plasmids were found to be fully capable of forming DNA complexes with a 32 P-labeled probe with the exception of the Y641F variant (lane l). These results suggest that unlike Tyr 641 (27,28), Ser 756 phosphorylation is not essential for nuclear localization or Stat6 DNA binding (lane h). Recent work by Braunstein et al. (12) has shown that Stat1 and Stat3 exist as preformed dimers in the absence of tyrosine phosphorylation yet cannot bind DNA (12). Here we show that, although IL4 can induce Stat6 serine phosphorylation in the absence of tyrosine phosphorylation, phosphorylation of Ser 756 alone is not sufficient to promote DNA binding (Fig. 3, lanes k and l). Whether serine phosphorylation may affect the oligomeric state of Stat6 remains to be determined.
Effect of Mutating Ser 176 , Ser 756 , Ser 827 , and Tyr 641 on IL4induced Transcription-Earlier work (27) shows that the truncated Stat6⌬677 variant, void of two of three proline-flanked serines, is critical for gene expression. To analyze the role of all of the conserved proline-juxtaposed serines on Stat6 transactivation potential including Ser 756 , luciferase reporter assays were performed utilizing the C⑀ gene promoter. A triple repeat of this Stat6 binding element was transfected into HEK293 and COS-7 cells along with the expression plasmids encoding the Stat6 variants. A constitutively expressed ␤-galactosidase gene was also included to compensate for differences in transfection efficiencies. Luciferase activity was measured in extracts of cells that had been incubated in the absence or presence of IL4 for 16 h (Fig. 4, A and B). Wild-type Stat6 consistently mediated a 2-3-fold induction of reporter gene expression in either cell type in response to IL4 stimulation. Mutant Stat6 S756A mediated a response comparable to that of Stat6-WT, whereas Stat6 Y641F was reduced to background levels. Within these cell types, we conclude that serine phosphorylation of Stat6 is not a requirement for C⑀ gene transcription and that IL4inducible Ser 756 phosphorylation in the absence of Tyr 641 phosphorylation is not sufficient to rescue its transcriptional activity. These results are in contrast to earlier studies with Stat5, its closest relative in which the deletion of Stat5a-regulated serine site elevated DNA binding activity, suggesting that this site might function in the capacity of a negative regulator (40,41). These findings imply different roles for phosphoserine in gene transcription by Stat5 -d) or serine to alanine mutants S176A (lanes e and f), S756A (lanes g and h) lanes a and b), S176A (lanes c and d), S756A (lanes  e and f), S827A (g and h), and Y641F (lane i and j). Panel C, at 48 h posttransfection of HEK293 cells expressing Stat6 WT or T645E variant, cells were pretreated with 200 nM CA as described above followed by stimulation with 100 nM IL4 for 10 min and cell lysates were harvested and immunoprecipitated with antibodies to His 6 and subjected to phosphotyrosine Western blot (upper panel) and His 6 reblot (lower panel). Panel D, nuclear extracts from cells were treated and harvested as described in panel C, and cells were subjected to electrophoretic mobility shift assay analysis using the radiolabeled Stat6 DNA binding probe (C⑀). Arrow denotes Stat6 binding for WT untreated (lanes a and b), T645E (lanes c and  d) To determine the identity of this regulated site, the stable WT Stat6-expressing HEK293 cell line was pretreated with 200 nM calyculin A without cytokine and Stat6 column affinity was purified, separated on SDS-PAGE, and stained with Coomassie Blue. Stat6 samples that showed retarded gel mobility compared with untreated control were excised and subjected to trypsin digestion, peptide isolation, and identification of charged species via mass spectrometry as described under "Experimental Procedures." A scan of phosphate precursor ions at m/z 79.1 yielded a peak at 926.3 was singly charged and localized to the tryptic fragment spanning Gly 640 -Lys 647 . Tandem mass spectrum of the positively charged m/z 928.4 was acquired, suggesting that the peaks were consistent with phosphorylation on a threonine residue that could reside at Thr 645 (data not shown).
To determine whether this residue may be a putative site for PP2A causing a loss of Stat6 function (30) This condition was also reflected in the loss of transcriptional activity for T645E variant compared with WT Stat6 (panel E) in which reporter activity of the C⑀-luciferase was reduced to unstimulated background levels. Taken together, these findings suggest that Thr 645 may act as a site of negative regulation for Stat6 by inhibiting tyrosine phosphorylation and DNA binding.
To determine whether the loss of IL4-mediated tyrosine phosphorylation of T645E was attributed to polar or steric considerations, several additional mutations were generated. As shown in Fig

DISCUSSION
Stat6 plays a key role in a variety immune-based diseases such as allergy but is also frequently found constitutively activated in Hodgkin and Reed-Sternberg cells of Hodgkin lymphoma (42). IL4 and IL13 exert their biological effects through Stat6 to drive cell growth, survival, or differentiation in a cell-dependent manner (38,39,43,44). This study indicated that, in PHA-activated primary human T-cells, IL4 or IL13 is competent to induce Stat6 serine phosphorylation while reconstitution studies mapped this regulated site to a single serine residue (human Ser 756 ). Phosphorylation of Ser 756 was not dependent on tyrosine phosphorylation of Tyr 641 . In HEK293 and COS-7 cells, Ser 756 was not critical for positively or negatively regulating DNA binding or transactivation potential. Lastly, evidence was provided that demonstrated that PP2A may exert a negative regulatory role on Stat6 via Thr 645 and be distinct from Ser 176 , Ser 756 , and Ser 827 . The Stat6 Thr 645 phos-phomimic variants were not readily tyrosine-phosphorylated, able to bind DNA, or regulate transcriptional activity in a cytokine-inducible manner.
Stat phosphoacceptor sites are represented by a positionally conserved tyrosine residue and a serine that is typically located within one or more proline-rich motifs of the transactivation domain (11). For Stats 1, 3, 4, this site is localized within a Pro-X-Ser-Pro motif, whereas Stat5a and Stat5b maps to a Pro-Ser-Pro site. Stat5a has an additional site at Ser 779 that is flanked by a proline residue (41). Other than being confined within the transactivation domain, Stat6 Ser 756 is not positionally conserved among other Stats and does not share significant homology through this region yet does retain a juxtaposed proline residue (Fig. 6). These findings suggest that Stat6 is differentially regulated by a serine kinase as compared with other Stats.
Current data support the notion that serine phosphorylation of Stats can positively and negatively regulate DNA binding and gene transcription. Stat6 transcriptional activity maps to a core region of residues spanning 677-791, whereas the introduction of a stop codon to generate Stat6⌬677, which harbors Ser 756 , retains the conserved phosphotyrosine site, Src homology 2, and DNA binding domains but is transcriptional inactive. However, we found that the mutation of S756A failed to show a significant and reproducible change in DNA binding, although a minor increase in DNA binding was occasionally observed for S756A (Fig. 3, lane h). To further clarify this issue, the Stat6 S756E variant to mimic phosphorylated Stat6 was generated; however, DNA binding and reporter activity were equivalent to wild type Stat6 (data not shown).
Stat6 contains a modular proline-rich transactivation domain, which exists in some transcription factors including BSAP (45) and EKLF (28,46) that may act to recruit proteins involved in chromatin reorganization. Additionally, Stat6 has been shown to interact with other regulators including NF-B (47,48), CREB-binding protein, NcoA-1, a CREB-binding protein-associated member of the p160/steroid receptor coactivator family (49) and the glucocorticoid receptor. Whether these proteins may interact with phosphorylated Ser 756 remains to be determined. Additionally, the identity of the Stat6 serine kinase is not immediately known. Several Stat serine kinases have been proposed to phosphorylate the PMSP motif, a consensus mitogen-activated protein kinase phosphorylation sequence (50). Supportive evidence have shown that Stats coprecipitates with ERK1/2 (16,52) and can be inhibited by MEK1/2 poisons (14,51,53) as well as inhibitors to phosphatidylinositol 3-kinase (55) and mammalian target of rapamycin (56) but not p38 and JNK (57)(58)(59). We performed similar studies using WT Stat6-transfected HEK293 cells and failed to see any changes in Stat6 serine phosphorylation in response to IL4 stimulation, results similar to two earlier studies (22,23). A recent study (54) finds that the Stat5a Ser 779 residue found within a SP site can directly interact and be phosphorylated by Pak1 and stimulate ␤-casein promoter activity. Pak1 is not a proline-directed serine-threonine kinase, and any role in phosphorylating Stat6 in T-cells is not presently clear. However, IL4 stimulation of FIG. 6. Sequence alignment of Stat SP-motif phosphoacceptor sites. Depicted are sequence alignments of mapped cytokine/growth factor inducible serine phosphorylation sites in human Stats. Note that each serine phosphoacceptor site is localized within a SP-motif.
HEK293 or COS-7 cells failed to show activation of Pak1 as measured by phosphoantibodies directed to the catalytic autophosphorylation site (Thr 423 , data not shown), whereas other family members were not tested.
Recent work has demonstrated that Stat6 cycles between active and inactive forms of the protein that is independent of new protein synthesis (51). One enzyme that can inactivate Stat6 is the serine-threonine phosphatase PP2A, which is inhibited by CA (30). While Woetmann et al. (30) reported that phosphoamino acid analysis of CA primarily protected Stat6 serine phosphorylation and to a lesser extent threonine, we found predominantly the phosphorylation of Thr 645 in our cell line, HEK293. Interestingly, Thr 645 is four amino acids Cterminal to the Tyr 641 , which is mandatory for DNA binding and transcriptional activation (27,28). Treatment of WT Stat6 with CA, the mutation of T645E or Asp to mimic a phosphorylated residue, showed significantly reduced tyrosine phosphorylation, DNA binding activity, and transcriptional activity following IL4 pretreatment (Fig. 5). Polar uncharged residues substituted for Thr 645 such as Gln or Asn showed no affect, whereas non-polar and sterically similar Val and the much bulkier hydrophobic Phe did not disrupt Stat6 tyrosine phosphorylation in response to IL4. It is tempting to speculate that the phosphorylation of Thr 645 electrostatically uncouples the tyrosine kinase responsible for activating Tyr 641 , which in turn blocks Stat oligomerization. Additionally, this phosphothreonine may oppose energetically low levels of Stat6-phosphorylated Tyr 641 to bind to its reciprocal Stat Src homology 2 domain-interacting partner, resulting in the loss of gene transcription. Interestingly, Thr 645 is conserved in mouse, rat, and human Stat6 but not other Stats. Lastly, unlike Stat6, we failed to detect altered gel mobility of Stat3, Stat5, or Stat5b following CA treatment in various cell types including T-cells (data not shown), further highlighting differences in these Stats.
In conclusion, IL4 and IL13 promote Stat6 serine phosphorylation in PHA-activated human T-cells, and Ser 756 represents the primary cytokine-inducible serine phosphorylation that is localized within a Stat SP-motif. This residue is not positionally conserved among other Stat family members and does not appear to be critical for DNA binding or transcriptional activity in non-T-cell lines. Ser 756 probably is not a target of PP2A, in contrast to Thr 645 , which is unique to Stat6. These findings suggest that Stat6 has shared and distinct mechanisms of regulation mediated by phosphorylation as compared with other Stat family members.