Intracellular Interaction of Interleukin (IL)-32α with Protein Kinase Cϵ (PKCϵ) and STAT3 Protein Augments IL-6 Production in THP-1 Promonocytic Cells*

Background: IL-32α is known to interact with FAK1, and IL-32α overexpression in chronic myeloid leukemia cells increases natural killer cell-mediated killing. Results: IL-32α interacted with PKCϵ and STAT3, mediated STAT3 phosphorylation, and thereby augmented IL-6 production. Conclusion: IL-32α elevated IL-6 production through interaction with PKCϵ and STAT3. Significance: The interaction of IL-32α with PKCϵ and STAT3 reveals a new intracellular mediatory role of IL-32α. IL-32α is known as a proinflammatory cytokine. However, several evidences implying its action in cells have been recently reported. In this study, we present for the first time that IL-32α plays an intracellular mediatory role in IL-6 production using constitutive expression systems for IL-32α in THP-1 cells. We show that phorbol 12-myristate 13-acetate (PMA)-induced increase in IL-6 production by IL-32α-expressing cells was higher than that by empty vector-expressing cells and that this increase occurred in a time- and dose-dependent manner. Treatment with MAPK inhibitors did not diminish this effect of IL-32α, and NF-κB signaling activity was similar in the two cell lines. Because the augmenting effect of IL-32α was dependent on the PKC activator PMA, we tested various PKC inhibitors. The pan-PKC inhibitor Gö6850 and the PKCϵ inhibitor Ro-31-8220 abrogated the augmenting effect of IL-32α on IL-6 production, whereas the classical PKC inhibitor Gö6976 and the PKCδ inhibitor rottlerin did not. In addition, IL-32α was co-immunoprecipitated with PMA-activated PKCϵ, and this interaction was totally inhibited by the PKCϵ inhibitor Ro-31-8220. PMA-induced enhancement of STAT3 phosphorylation was observed only in IL-32α-expressing cells, and this enhancement was inhibited by Ro-31-8220, but not by Gö6976. We demonstrate that IL-32α mediated STAT3 phosphorylation by forming a trimeric complex with PKCϵ and enhanced STAT3 localization onto the IL-6 promoter and thereby increased IL-6 expression. Thus, our data indicate that the intracellular interaction of IL-32α with PKCϵ and STAT3 promotes STAT3 binding to the IL-6 promoter by enforcing STAT3 phosphorylation, which results in increased production of IL-6.

IL-32␣ is known as a proinflammatory cytokine. However, several evidences implying its action in cells have been recently reported. In this study, we present for the first time that IL-32␣ plays an intracellular mediatory role in IL-6 production using constitutive expression systems for IL-32␣ in THP-1 cells. We show that phorbol 12-myristate 13-acetate (PMA)-induced increase in IL-6 production by IL-32␣-expressing cells was higher than that by empty vector-expressing cells and that this increase occurred in a time-and dose-dependent manner. Treatment with MAPK inhibitors did not diminish this effect of IL-32␣, and NF-B signaling activity was similar in the two cell lines. Because the augmenting effect of IL-32␣ was dependent on the PKC activator PMA, we tested various PKC inhibitors. The pan-PKC inhibitor Gö6850 and the PKC⑀ inhibitor Ro-31-8220 abrogated the augmenting effect of IL-32␣ on IL-6 production, whereas the classical PKC inhibitor Gö6976 and the PKC␦ inhibitor rottlerin did not. In addition, IL-32␣ was co-immunoprecipitated with PMA-activated PKC⑀, and this interaction was totally inhibited by the PKC⑀ inhibitor Ro-31-8220. PMAinduced enhancement of STAT3 phosphorylation was observed only in IL-32␣-expressing cells, and this enhancement was inhibited by Ro-31-8220, but not by Gö6976. We demonstrate that IL-32␣ mediated STAT3 phosphorylation by forming a trimeric complex with PKC⑀ and enhanced STAT3 localization onto the IL-6 promoter and thereby increased IL-6 expression. Thus, our data indicate that the intracellular interaction of IL-32␣ with PKC⑀ and STAT3 promotes STAT3 binding to the IL-6 promoter by enforcing STAT3 phosphorylation, which results in increased production of IL-6.
IL-32 is known as a multifunctional proinflammatory cytokine produced by various types of cells, including T cells, natural killer cells, monocytes, epithelial cells, and vascular endothelial cells. Elevated IL-32 levels have been associated with several inflammatory diseases, such as chronic obstructive pulmonary disease (1), rheumatoid arthritis (2), and Crohn disease (3).
Interestingly, despite that IL-32 has been studied as a secretable factor for its proinflammatory function, its cognate receptor has not yet been identified. Some reports have shown that IL-32 is detected mostly in cell lysates rather than in culture supernatants (3)(4)(5)(6)(7)(8). Some reports have also indicated that IL-32␤ is the most abundant isoform among six splice variants (IL-32␣, IL-32␤, IL-32␥, IL-32␦, IL-32⑀, and IL-32) and that it is involved in the activation-induced cell death of T cells (4,5). IL-32␤ and IL-32␥ are also known to induce the anti-inflammatory cytokine IL-10 (9, 10). It was recently reported that IL-32␣ overexpression in chronic myeloid cells increases natural killer cell-mediated killing (11). IL-32␣ is known to be strongly expressed in pancreatic cancer cells (12). In addition, the interaction of IL-32␣ with integrin as well as paxillin through its ␣-helix bundle structure was reported (13). These data suggest the existence of isoform-specific or cell type-specific functional differences among IL-32 isoforms; however, it is unclear whether IL-32 has intracellular functions.
PKC is a family of serine/threonine kinases that are known to be involved in cell growth, migration, and inflammation (14). Specific PKC isoforms are crucial to the regulation of myeloid, erythroid, and megakaryocytic development (15)(16)(17)(18). A variety of tissues, such as those of the nervous, cardiac, and immune systems, express PKC⑀, and the role of PKC⑀ is important for their proper function (19 -22). PKC⑀ may be a useful therapeutic target to treat disease conditions, such as inflammation (19), ischemia (23), addiction (24), pain (25), anxiety (26,27), and cancer (28,29).
IL-6 expression has been shown to be regulated by several PKC isoforms (30,31). IL-6 production is induced by PKC via various signaling pathways, but it is only slightly induced by the PKC activator phorbol 12-myristate 13-acetate (PMA) 2 alone (32). STAT3 (signal transducer and activator of transcription 3) is also known to induce IL-6 production in starved cancer cells, whereas JAK1/STAT3 signaling is known to mediate IL-6 signaling (33)(34)(35). Several studies have shown that PKC⑀ interacts with STAT3 to induce its constitutive activation in prostate cancer and skin cancer (36,37). Although previous studies have provided evidence for its action as a soluble inducer of inflammation, our study shows an unexpected action of IL-32, namely its interaction with PKC⑀ and STAT3 in elevating IL-6 production as an intracellular mediator.
Construction of Expression Vectors and IL-6 Reporter Plasmid-The pcDNA3.1 ϩ -6ϫMyc vector was generated by inserting the 6ϫMyc tag from the pCS3MT vector and then subcloning IL-32␣ cDNA into this vector using EcoRI and XhoI. STAT3 cDNA was PCR-amplified from the human spleen cDNA library (Clontech, Palo Alto, CA) and subcloned into pCS3MT-6ϫMyc. A 5ϫFLAG tag was generated by ligating hybridized 5Ј-phosphorylated sense and antisense DNA strands of the FLAG tag sequence (Xenotech, Daejeon, Korea), and the tag was then inserted into pcDNA3.1 ϩ . cDNAs for PKC␣, PKC␦, PKC⑀, PKC, and retinoid X receptor were synthesized by RT-PCR of total RNA collected from THP-1 cells; they were then PCR-amplified and subcloned into the pcDNA3.1 ϩ -5ϫFLAG vector using EcoRI and XhoI. The IL-6 promoter region (Ϫ1145 to ϩ19) was PCR-amplified from THP-1 genomic DNA. The primer set was 5Ј-GGTACCATC-CTGAGGGGAAGAGGG-3Ј (sense) and 5Ј-GCTCCTGGA-GGGGAGATAGAGCTT-3Ј (antisense). The PCR product was digested with NheI and XhoI restriction enzymes. The digested fragment (Ϫ226 to ϩ14) was ligated into the pGL3-Basic vector.
ChIP Assay-For this experiment, we used a commercially available ChIP assay kit (Millipore) according to the manufacturer's instructions. Briefly, THP-1-IL-32␣ cells were treated with PKC inhibitors for 1 h, and then both THP-1-EV and THP-1-IL-32␣ cells were treated with PMA for 3 h, including inhibitor-treated samples. The cells were fixed with 1% formaldehyde, lysed with kit lysis buffer, and sonicated with five pulses for 5 s each. After centrifugation at 13,000 rpm for 20 min, the supernatants were mixed with anti-Myc tag antibody or normal mouse IgG and maintained overnight. Protein A-agarose/ salmon sperm DNA (60 l, 50% slurry) was added to each sample, and the pulled down DNA fragments were eluted. PCR amplification using the eluted DNA as the template was performed for 35 cycles at an annealing temperature of 59°C. The primers used for PCR amplification of the IL-6 promoter were 5Ј-GTCACATTGCACAATCTTAAT-3Ј (sense, Ϫ162 to Ϫ142) and 5Ј-GAGCCTCAGACATCTCCAGTC-3Ј (antisense, Ϫ21 to Ϫ1).
Statistical Analysis-Statistical significance was analyzed by Student's unpaired two-tailed t test.

IL-32␣ Up-regulates IL-6 Production upon PMA Stimulation-
We generated a stable expression system for IL-32␣ by transfecting THP-1 promonocytic cells with 6ϫMyc-tagged IL-32␣ because the endogenous IL-32␣ protein is hardly detected in contrast with its transcript (Fig. 1A). Western blotting did not reveal any IL-32 isoform with THP-1 cells. Only the IL-32␤ transcript was identified by RT-PCR and sequencing (data not shown). We observed that the IL-6 transcript was weakly expressed by IL-32␣ overexpression, but upon treatment with 10 nM PMA, the levels of IL-6 transcripts in THP-1-IL-32␣ cells were markedly higher than those in THP-1-EV cells. In THP-1-EV cells, PMA stimulation resulted only in minimal expression of IL-6 mRNA (Fig. 1B). We examined the effect of other stimulants (LPS and poly(I:C)) on IL-6 induction in both cell lines. These stimulants did not induce IL-6 even though they slightly enhanced IL-6 production upon co-treatment with PMA (Fig. 1C). We further confirmed that the effect of PMA on IL-6 production was time-and dose-dependent. In THP-1-IL-32␣ cells, the IL-6 protein was detected in the culture medium even at 3 h after PMA stimulation, and the protein levels increased steeply until 48 h. However, in THP-1-EV cells, the amount of IL-6 secretion was less than half that in THP-1-IL-32␣ cells at 48 h ( Fig. 2A). In the dose-related experiment, IL-6 production in THP-1-IL-32␣ cells was more than three times that in empty vector cells (Fig. 2B). The levels of IL-1␤ and TNF␣ were below the detection limit in both cell lines (data not shown). IL-8 levels were also increased by PMA treatment, but the expression levels were similar in THP-1-EV and THP-1-IL-32␣ cells (Fig. 2C).
MAPK Does Not Contribute to IL-32␣-induced Up-regulation of IL-6 Production-We examined whether MAPK signaling mediates the effect of IL-32␣ on IL-6 production because IL-32 is known to activate p38 signaling pathways (39). When THP-1-EV and THP-1-IL-32␣ cells were treated with PD98059 for ERK inhibition, SB203580 for p38 inhibition, and SP600125 for JNK inhibition, the up-regulated production of IL-6 at both the mRNA and protein levels was sustained for longer periods of time in THP-1-IL-32␣ cells than in THP-1-EV cells (Fig. 3, A  and B), although p38 or ERK inhibition affected IL-6 production compared with PMA-alone treatment in both cell lines (Fig. 3B). This may be because p38 or ERK is also involved in IL-6 induction. JNK did not seem to be involved in IL-6 production in this system as described elsewhere (40,41). Our data imply that the augmented production of IL-6 by IL-32␣ was not mediated by MAPK signaling pathways and that other signal molecules may be involved in the augmenting effect of IL-32␣ on IL-6 production by THP-1 cells.
PKC⑀ Is Involved in Enhanced Production of IL-6 by IL-32␣-We expected the involvement of a certain type of PKC because the increase in IL-6 production by IL-32␣ was PMA-dependent. As shown in Fig. 4 (A and B), the increase in IL-6 production by IL-32␣ was totally abrogated by the pan-PKC inhibitor Gö6850, but not by the classical PKC inhibitor Gö6976. This implies that PKCs other than the classical PKCs may be involved in the IL-32␣ effect. Nonetheless, it appears that classical PKCs were involved in IL-6 production by a mechanism not involving IL-32␣ because IL-6 production was decreased by Gö6976 treatment in THP-1-EV cells (Fig. 4, A and B). We treated the cells with rottlerin, a PKC␦ inhibitor, or Ro-31-8220, which inhibits classical PKCs and PKC⑀ among novel PKCs. As shown in Fig. 4 (C and D), Ro-31-8220 treatment totally abrogated the augmenting effect of IL-32␣ on IL-6 production. Rottlerin also decreased IL-6 production in both THP-1-EV and THP-1-IL-32␣ cells, but IL-6 production was still maintained at higher levels in THP-1-IL-32␣ cells than in THP-1-EV cells; this implies that PKC␦ has a minor effect on the augmented production of IL-6 by IL-32␣.

JOURNAL OF BIOLOGICAL CHEMISTRY 35559
PMA-activated PKC⑀ Interacts with IL-32␣-To investigate how PKCs are involved in IL-32␣-induced augmentation of IL-6 production, we performed immunoprecipitation experiments after cotransfecting HEK293 cells with 6ϫMyc-tagged IL-32␣ and each PKC isoform (␣, ␦, ⑀, and ) tagged with the 5ϫFLAG. Consistent with the results shown in Fig. 4, IL-32␣ was found to interact with PKC⑀ and PKC␦. The extent of interaction of IL-32␣ with PKC␦ was weaker than that with PKC⑀ (Fig. 5A), as expected from the results of Fig. 4 (C and D). IL-32␣ interacted with PMA-activated PKC⑀ (Fig. 5, B and C). The interaction between IL-32␣ and PKC⑀ was further verified by immunoprecipitation after cotransfection of HEK293 cells with both expression vectors or in THP-1-IL-32␣ cells. As shown in Fig. 5D, IL-32␣ was immunoprecipitated with PKC⑀ upon PMA treatment. This interaction was suppressed by the PKC⑀-specific inhibitor Ro-31-8220. These data indicate that PMA-activated PKC⑀ binds to IL-32␣. Fig. 5E shows IL-32␣ co-immunoprecipitated with endogenous PKC⑀ upon PMA stimulation in THP-1-IL-32␣ cells, and this interaction was blocked by Ro-31-8220 treatment. Although the classical PKC inhibitor Gö6976 affected this interaction, the association was still maintained (Fig. 5D). These data imply that PKC⑀ is the major factor involved in the effect of IL-32␣ on IL-6 production.
IL-32␣ Does Not Modulate NF-B Signaling but Reinforces STAT3 Phosphorylation through Forming Trimeric Complex with PKC⑀-Various cytokines, including IL-6, are known to be induced by NF-B signaling. In fact, the NF-B consensus sequence is found on the IL-6 promoter (42,43). Interestingly, the NF-B-binding sequence partially overlaps with the STAT3 consensus sequence (see Fig. 7C), which suggests that STAT3 also contributes to IL-6 induction. STAT3 and NF-B have been reported to cooperatively induce IL-6 (35). On the basis of these facts, we examined NF-B signaling and STAT3 phosphorylation status after PMA treatment of THP-1-EV and THP-1-IL-32␣ cells. As shown in Fig. 6A, phospho-IB␣ was increased by PMA treatment, whereas IB␣ was gradually

IL-32␣ Interacts with PKC⑀ and STAT3
degraded in both cell lines. This implies that NF-B signaling was not changed by IL-32␣. However, PMA treatment induced greater STAT3 (Ser-727) phosphorylation in THP-1-IL-32␣ cells than in THP-1-EV cells (Fig. 6B). The phosphorylation of STAT3 was suppressed by Ro-31-8220 treatment, but not by Gö6976 treatment (Fig. 6C). Thus, these data indicate that IL-32␣ modulates STAT3 signaling via PKC⑀, but not NF-B signaling. We further delineated the relationship between PKC⑀, IL-32␣, and STAT3. The immunoprecipitation assay revealed that the interaction of IL-32␣ with STAT3 and PKC⑀ resulted in the formation of a trimeric complex. As shown in Fig. 7A, IL-32␣ co-immunoprecipitated with STAT3 as well as PKC⑀. Trimeric complex formation was dependent on PKC⑀ activation because Ro-31-8220 inhibited complex formation. These data imply that PKC⑀ phosphorylates STAT3, which is mediated by IL-32␣.

IL-32␣ Interacts with PKC⑀ and STAT3
OCTOBER 12, 2012 • VOLUME 287 • NUMBER 42 IL-32␣ Reinforces STAT3 Phosphorylation by PKC⑀ and Augments IL-6 Gene Expression by Promoting STAT3 Localization onto IL-6 Promoter-Next, we investigated whether IL-32␣ mediates the phosphorylation of STAT3 by PKC⑀. After cotransfecting Myc-tagged STAT3 with or without IL-32␣ into HEK293 cells, Myc-tagged STAT3 was immunoprecipitated and then analyzed for phospho-STAT3 using phospho-STAT3 (Ser-727) antibody (Fig. 7B). STAT3 Ser-727 was phosphorylated in the presence of IL-32␣ upon PMA stimulation, but in the absence of IL-32␣, its phosphorylation was significantly decreased. It is obvious that PKC⑀ phosphorylates STAT3 Ser-727 because STAT3 Ser-727 phosphorylation was inhibited by Ro-31-8220, but not by Gö6976. PKC⑀ co-immunoprecipitated with IL-32␣ as well as STAT3. On the other hand, we observed STAT3 phosphorylation despite no input of PKC⑀ (Fig. 7B,  fourth lane). This effect may be attributed to endogenous PKC⑀ because HEK293 cells are known to express all types of PKC isoforms (44). Using ChIP, we next demonstrated that the enhanced phosphorylation of STAT3 by IL-32␣ induced a greater amount of STAT3 to present on the IL-6 promoter (Fig.  7C). STAT3 localization onto the IL-6 promoter was severely suppressed by the PKC⑀ inhibitor Ro-31-8220, but not by the classical PKC inhibitor Gö6976. The effect of IL-32␣ on IL-6 promoter activity was analyzed by cotransfecting HEK293 cells with STAT3 and PKC⑀ expression vectors. As shown in Fig. 7D, in the absence of IL-32␣, IL-6 reporter activity was decreased to almost half that in the presence of IL-32␣. The reporter activity was suppressed to the basal level by Ro-31-8220, but was not FIGURE 7. IL-32␣ interacts simultaneously with STAT3 and PKC⑀, mediates STAT3 phosphorylation by PKC⑀, and increases IL-6 gene expression. A, HEK293 cells were cotransfected with 6ϫMyc-tagged IL-32␣ and STAT3 and 5ϫFLAG-tagged PKC⑀. Cells were treated with 20 nM PMA for 90 min. For inhibitor-treated samples, cells were treated with 10 M Ro-31-8220 (Ro31) or 10 M Gö6976 (6976) 1 h before PMA treatment. Immunoprecipitations (IP) with 5 g of anti-IL-32 antibody (KU32-52) and normal mouse IgG were conducted. The levels of IgG light chain bands show constant loadings. B, HEK293 cell lysates were prepared in the same way and then immunoprecipitated with 1.5 g of anti-Myc antibody and analyzed for phospho-STAT3 (Ser-727) and pulled down PKC⑀. The arrow indicates an unidentified band. C, the expression levels of the transfected genes were determined by Western blotting with 20 g of whole cell lysates (WCL). ChIP was performed with 3 g of anti-STAT3 antibody and normal rabbit IgG, and then the immunoprecipitated IL-6 promoter with STAT3 was PCR-amplified. The schematic diagram shows the ChIP region (160 bp) of the IL-6 promoter. D, IL-6 promoter-firefly luciferase reporter plasmid (0.5 g), STAT3 (1 g), and PKC⑀ (1 g) were cotransfected into HEK293 cells with or without IL-32␣ expression vector (1 g). After overnight incubation, cells were treated with 20 nM PMA for an additional 24 h. Cells were treated with 10 M Ro-31-8220 or 10 M Gö6976 1h before PMA treatment of inhibitor-treated samples. All values are means Ϯ S.E. *, p Ͻ 0.0001 (presence versus absence of IL-32␣).

DISCUSSION
IL-32 is known to be a proinflammatory cytokine and probably exerts its effects by binding to its cell surface receptor, although the receptor has not yet been identified. Although various cell types, including T cells, natural killer cells, monocytes, macrophages, epithelial cells, and endothelial cells, are known to express IL-32, not many cell types have been reported to secrete this molecule. IL-32 has even been reported to be a membrane-associated protein that is released via a non-classical secretory pathway (45). IL-32 seems to be multifunctional because it has been shown to induce proinflammatory cytokines (8,39,46), apoptosis (4,47), and cell differentiation (48,49).
A recent report indicated that IL-32␣ and IL-32␤ interact with integrin and that IL-32␣ binds to paxillin and FAK1 (focal adhesion kinase 1), which implies that IL-32 may be involved in the formation of the focal adhesion protein complex (13). In this study, we found for the first time that IL-32␣ interacts with PKC⑀ and STAT3 upon PMA stimulation and thereby up-regulates IL-6 production. Many reports have indicated that IL-32 induces IL-6, but the precise mechanism remains elusive. Our data suggest that IL-32␣ functions intracellularly through interaction with PKC⑀ and STAT3. We also found that IL-32␣ interacts with PKC␦ (Fig. 4A). These results imply that IL-32␣ may be an adaptor protein for PKC, which is known to be a receptor for activated C kinase (RACK). RACK1 is an anchoring protein for activated PKC␤II that mediates the binding of Src tyrosine kinase, integrin, and phosphodiesterase. PKC⑀-specific RACK2 is a coated vesicle protein that is involved in vesicular release and cell-to-cell communication (50). RACK1 and RACK2 interact with their specific partners, PKC␤II and PKC⑀, respectively. Ten isotypes of PKCs have been identified, and it is thought that every PKC may have a specific RACK. PKC⑀ induces prostate cancer or skin cancer by phosphorylating STAT3. PKC␦ has also been known to interact with and phosphorylate STAT3 Ser-727 (51). Although a previous report indicated that the PKC␦ inhibitor rottlerin inhibited IL-6 production in a PKC␦-independent manner (52), our data show that IL-6 production was only slightly inhibited by rottlerin in IL-32␣-expressing cells, which suggests that PKC␦ may be implicated in IL-32␣-mediated IL-6 up-regulation to some extent. In the conventional pathway, STAT3 is activated by IL-6 signaling. However, in this study, we showed that the interaction of IL-32␣ with PKC⑀ and STAT3 induces STAT3 Ser-727 phosphorylation and enhances IL-6 production.