| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
-Induced Production of CXC Ligand 9 (CXCL9), CXCL10, and CXCL11 in Human KeratinocytesDepartment of Dermatology, Teikyo University School of Medicine, Tokyo 173-8605, Japan
Address all correspondence and requests for reprints to: Dr. Naoko Kanda, Department of Dermatology, Teikyo University School of Medicine, 11-1, Kaga-2, Itabashi-Ku, Tokyo 173-8605, Japan. E-mail: nmk{at}med.teikyo-u.ac.jp.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B (NF-B) and signal transducer and activator of transcription (STAT)1 and that of CXCL11 depends on interferon (IFN)-regulatory factor (IRF)-1. We examined in vitro effects of prolactin on CXCL9, CXCL10, and CXCL11 production in human keratinocytes. Although prolactin alone was ineffective, it enhanced IFN-
-induced secretion and mRNA expression of CXCL9, CXCL10, and CXCL11 in parallel to the activation of STAT1, NF-B, and IRF-1. Inhibitors of Janus kinase (JAK), p38 MAPK, and MAPK/ERK kinase (MEK) suppressed prolactin- plus IFN--induced CXCL9, CXCL10, and CXCL11 production and NF-B, STAT1, and IRF-1 activities. Prolactin induced phosphorylation of JAK2 and ERK, whereas IFN- induced phosphorylation of JAK1, JAK2, and p38 MAPK. Prolactin modestly or IFN- greatly induced tyrosine phosphorylation of STAT1, and both were suppressed by JAK inhibitor. Prolactin modestly or IFN- greatly induced serine phosphorylation of STAT1, which was suppressed by MEK or p38 MAPK inhibitor, respectively. Prolactin induced phosphorylation of inhibitory B
and NF-B p65, which was suppressed by MEK inhibitor. These results suggest that prolactin may enhance IFN--induced CXCL9, CXCL10, and CXCL11 production in keratinocytes via activation of STAT1, NF-B, and IRF-1 through JAK2 and MEK/ERK pathways. Prolactin may promote type 1 T cell infiltration into psoriatic lesions via these chemokines. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
(IFN-)-producing type 1 T cells play important roles in the pathogenesis of psoriasis (1, 2); the presentation of bacterial, viral, or autoantigen by antigen-presenting cells (Langerhans cells or dendritic cells) to T cells activates and induces them to produce inflammatory cytokines such as IFN- or TNF- that stimulate the expression of adhesion molecules on keratinocytes or endothelial cells or to produce cytokines such as IL-6 that stimulate proliferation of keratinocytes (2). Keratinocytes, in turn, are stimulated to secrete cytokines, such as TGF- or vascular endothelial growth factor, in an autocrine or paracrine manner to promote the growth of themselves or endothelial cells (2). In addition, the expression of adhesion molecules by endothelial cells allows extravasation of T cells into the skin. The trafficking of type 1 T cells into psoriatic skin lesions also requires specific chemokines and chemokine receptors (2). IFN--induced chemokines CXC ligand 9 (CXCL9), CXCL10, and CXCL11 preferentially recruit type 1 T cells by binding cell surface CXCR3 (3). The expression of CXCL9, CXCL10, or CXCL11 by epidermal keratinocytes is enhanced in psoriatic lesions (1), which may induce the migration of type 1 T cells from dermis to epidermis and trigger skin inflammation. We have recently found that in vitro CXCL9, CXCL10, and CXCL11 production by human keratinocytes depends on the activities of transcription factors, nuclear factor-B (NF-B) and signal transducer and activator of transcription (STAT)1 and that of CXCL11 also depends on IFN-regulatory factor (IRF)-1 (4).
Prolactin (PRL) is a hormone mainly produced in anterior pituitary gland, although it can be produced in extrapituitary sites (5). The PRL released plays an important role in the regulation of a variety of physiological cellular functions, such as proliferation, differentiation, or survival, by endocrine and paracrine/autocrine mechanisms (5). PRL is a type I cytokine superfamily member such as IL-2 to IL-7 (5) and shows a variety of immunoregulatory effects; PRL enhances proliferation or IFN- production in T cells or natural killer cells alone or together with IL-2 or IL-12 (6). PRL induces the expression of /
T cell receptors on rat pre-T lymphocyte Nb2 cells (7). PRL stimulates antigen-presenting cells by increasing the expression of MHC class II or costimulatory molecules, CD40, B7-1, or B7-2 (8). It has been hypothesized that PRL may modulate the skin immune system and may be involved in the pathogenesis of psoriasis; psoriatic patients are often associated with hyperprolactinemia, and bromocriptine, which inhibits PRL release, is therapeutically effective for psoriatic patients (8, 9). PRL detected in the skin is derived from the circulation, PRL-producing migratory lymphocytes, and local synthesis by keratinocytes or fibroblasts (5). Functional PRL receptors (PRLRs) are detected on epidermal keratinocytes, and PRL effectively increased the in vitro growth of keratinocytes (10), indicating that PRL may be involved in the hyperproliferation of keratinocytes in psoriasis. It is thus plausible that PRL may also be related to the enhanced production of CXCL9, CXCL10, or CXCL11 by keratinocytes in psoriatic lesions. PRL enhances IL-8 production in monocytic THP-1 cells in vitro (11) and enhances monocyte chemoattractant protein-1 production in vivo in rat corpus luteum (12). However, it has not been examined whether PRL may induce chemokine production in epidermal keratinocytes.
In this study, we investigated in vitro effects of PRL on CXCL9, CXCL10, and CXCL11 production in human keratinocytes. We found that PRL up-regulated IFN--induced production of these chemokines by activating transcription factors STAT1, NF-B, and IRF-1.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, monoclonal antihuman PRLR antibody, and control murine IgG1 were from R&D Systems (Minneapolis, MN).
Culture of keratinocytes
Human neonatal foreskin keratinocytes were cultured in serum-free keratinocyte growth medium (KGM) (Clonetics, Walkersville, MD) consisting of keratinocyte basal medium (KBM) supplemented with 0.5 µg/ml hydrocortisone, 5 ng/ml epidermal growth factor, 5 µg/ml insulin, and 0.5% bovine pituitary extract. Cells in the third passage were used.
Chemokine secretion
The keratinocytes (5 x 104/well) were seeded in triplicate into 24-well plates in 0.4 ml KGM, adhered overnight, washed, and incubated with phenol red-free KBM for 24 h. The cells were washed and incubated with the indicated concentrations of IFN- and/or PRL in KBM for 48 h. The culture supernatants were assayed for CXCL9, CXCL10, or CXCL11 by ELISA (R&D). Because PRL increased the number of keratinocytes 1.3- to 1.5-fold compared with controls, chemokine secretion was normalized to the cell number after culture. In some experiments, the keratinocytes were preincubated with anti-PRLR antibody, control murine IgG1, or signal inhibitors for 30 min before the addition of cytokines.
RT-PCR
The keratinocytes were incubated as above for 12 or 2 h to analyze the mRNA levels of chemokines or IRF-1, respectively. Total cellular RNA was extracted, adjusted to the cell number after culture, and reverse-transcribed to produce cDNA. The cDNA was thermocycled for PCR using specific primers as described (13). PCR products were analyzed by electrophoresis, and densitometric analysis was performed by ATTO lane analyzer version 3 (ATTO Corp., Osaka, Japan). mRNA levels of chemokines or IRF-1 were normalized to that of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and are shown as a fold induction.
Plasmid and transfection
pNF-B-luc, pGAS-luc, or pISRE-luc were purchased from Clontech (San Jose, CA), containing four copies of NF-B elements, two copies of STAT1-binding sequences, or five copies of IRF-1-binding sequences, respectively, in front of the TATA box upstream of the firefly luciferase reporter. Transient transfection was performed with Fugene 6 (Roche, Indianapolis, IN) as described (14). The keratinocytes were plated in 24-well plates and grown to about 60% confluence. pNF-B-luc, pGAS-luc, or pISRE-luc and herpes simplex virus thymidine kinase promoter-linked renilla luciferase vector (pRL-tk) mixed with Fugene 6 were added to the keratinocytes. After 6 h, the cells were washed and incubated in phenol red-free KBM for 24 h and then treated with the indicated concentrations of IFN- and/or PRL. After 18 h, the firefly and renilla luciferase activities of the cell extracts were quantified by the dual luciferase assay system (Promega, Madison, WI). The transcriptional activities of NF-B, STAT1, or IRF-1 were expressed as ratios of firefly/renilla luciferase activity.
Cell-based ELISA
The phosphorylation levels of inhibitory B (IB), NF-B p65, STAT1, JAK1, JAK2, p38 MAPK, or ERK were analyzed by cell-based ELISA as described (15). We used monoclonal antibodies against phosphorylated or pan-IB, p65, p38 MAPK (SuperArray Bioscience Corp., Frederick, MD), serine- or tyrosine-phosphorylated or pan-STAT1 or ERK (Raybiotech, Inc., Norcross, GA), tyrosine-phosphorylated JAK1 or JAK2 (Stressgen Bioreagent, Ann Arbor, MI) or pan-JAK1 or -JAK2 (Abgent, San Diego, CA). The keratinocytes (1 x 104/well) were seeded in triplicate into 96-well plates in 0.1 ml KGM, adhered overnight, washed, and incubated with phenol red-free KBM for 24 h. The cells were washed and preincubated with signal inhibitors for 30 min and then incubated with 10 ng/ml PRL and/or 1 ng/ml IFN- for 5 min to analyze IB phosphorylation or for 15 min to analyze the phosphorylation of the other proteins, respectively. The cells were fixed, quenched, blocked, and incubated with primary antibodies and then incubated with horseradish-peroxidase-conjugated secondary antibodies, washed, and developed. Phosphorylation status was shown as an absorbance ratio of phosphorylated protein/total protein.
Western blotting
The expression of PRLR was analyzed by Western blotting as described (16), using whole-cell lysates from keratinocytes and breast cancer MCF7 cells (American Type Culture Collection, Manassas, VA), antihuman PRLR antibody or anti-GAPDH antibody (Santa Cruz Biotechnology, Santa Cruz, CA), and secondary antibodies.
Measurement of PRL release
Keratinocytes were cultured with KBM alone or IFN- (1 ng/ml) for 48 h. The supernatant PRL level was measured by ELISA as described (17) using antihuman PRL antibody and biotinylated antihuman PRL antibody (R&D). The sensitivity of this assay was 70 pg/ml.
Statistical analyses
One-way ANOVA with Dunnet’s multiple comparison test was used in Figs. 2
, A–C, and 3. One-way ANOVA with Tukey-Kramer multiple comparison test was used in Table 1 and Figs. 4, 5, 6, A–D, and 7. A P value of less than 0.05 was considered significant.
|
|
|
|
|
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
).
|
-induced CXCL9, CXCL10, and CXCL11 production-induced secretion of these chemokines (Fig. 2
, A–C). PRL (10 ng/ml) increased IFN--induced (1 ng/ml) CXCL9, CXCL10, and CXCL11 secretion by 2.55-, 2.76-, and 3.81-fold, respectively. In parallel with protein secretion, PRL enhanced IFN--induced CXCL9, CXCL10, and CXCL11 mRNA expression (Fig. 2D).
CXCL9, CXCL10, or CXCL11 promoters contain STAT1, NF-B, or IRF-1-binding sequences, and these transcription factors may mediate the transcription of these genes (18, 19, 20). We have recently found that in vitro CXCL9, CXCL10, and CXCL11 production by human keratinocytes depend on the activities of NF-B and STAT1 and that of CXCL11 also depends on IRF-1 (4). We thus examined whether PRL may increase the activities of these transcription factors in parallel to the enhancement of CXCL9, CXCL10, or CXCL11 production. IFN- (1 ng/ml) alone increased the transcriptional activities of STAT1, NF-B, and IRF-1 9.08-, 1.79-, and 2.30-fold compared with controls, respectively (Fig. 3, A–C). PRL (10 ng/ml) increased the IFN--induced STAT1, NF-B, and IRF-1 activities 2.17-, 1.94-, and 2.03-fold, respectively, although it did not increase the activities of these transcription factors in the absence of IFN-. These results indicate that PRL may enhance IFN--induced CXCL9, CXCL10, or CXCL11 production by up-regulating STAT1, NF-B, or IRF-1 activities.
|
(1 or 10 ng/ml) was less than detectable limit (<70 pg/ml). Anti-PRLR antibody did not reduce the IFN--induced CXCL9, CXCL10, or CXCL11 secretion or STAT1, NF-B, or IRF-1 activities (Table 1), indicating that endogenous PRL from keratinocytes may not promote the effects of IFN- in an autocrine manner. However, this antibody decreased the IFN- plus PRL-induced chemokine secretion or transcription factor activities to the levels with IFN- alone, indicating that the synergistic effects of exogenous PRL may be mediated by PRLR on keratinocytes.
Involvement of MEK/ERK and p38 MAPK in IFN- and/or PRL-induced production of CXCL9, CXCL10, or CXCL11
It is reported that PRL activates STAT1 via stimulation of receptor-associated JAK2 or activates MEK/ERK, p38 MAPK, or JNK signaling pathways dependent on cell types (21). It is also reported that IFN- activates STAT1 via stimulation of receptor-associated JAK1 and JAK2 or activates MEK/ERK or p38 MAPK dependent on cell types (22, 23, 24). We thus examined the involvement of these signals in PRL-induced stimulation of chemokine production in synergy with IFN-. CXCL9, CXCL10, or CXCL11 secretion induced by IFN- alone was completely suppressed by JAK inhibitor I, and partially suppressed by p38 MAPK inhibitor SB203580, but not by U0126 or SP600125, inhibitors of MEK or JNK, respectively (Fig. 4, A–C), indicating the involvement of JAKs and p38 MAPK in the chemokine production by IFN- alone. On the other hand, PRL plus IFN--induced CXCL9, CXCL10, or CXCL11 secretion was completely suppressed by JAK inhibitor I, and partially suppressed by U0126 and less potently by SB203580, but not by SP600125, indicating the involvement of MEK/ERK in addition to JAKs and p38 MAPK in PRL plus IFN--induced production of these chemokines. Similar results were obtained for IFN-- or PRL plus IFN--induced CXCL9, CXCL10, or CXCL11 mRNA expression (Fig. 4D). The addition of these kinase inhibitors did not reduce the viability of keratinocytes (>95% viable).
Involvement of JAKs, MEK/ERK, and p38 MAPK in PRL plus IFN--induced activation of STAT1, NF-B, or IRF-1
We then examined whether JAKs, MEK/ERK, or p38 MAPK may be involved in IFN-- or PRL plus IFN--induced stimulation of STAT1, NF-B, or IRF-1 activities. IFN--induced stimulation of STAT1, NF-B, or IRF-1 activities was suppressed by JAK inhibitor I and SB203580 but not by U0126 or SP600125 (Fig. 5, A–C). IFN- plus PRL-induced stimulation of STAT1 or IRF-1 activities was mostly suppressed by JAK inhibitor I and partially by U0126 and SB203580 (Fig. 5, A and C), whereas IFN- plus PRL-induced NF-B activity was partially suppressed by JAK inhibitor I, U0126, or SB203580 (Fig. 5B). Thus, JAKs and p38 MAPK may be responsible for IFN--induced activation of STAT1, NF-B, or IRF-1, whereas MEK/ERK, in addition to JAKs and p38 MAPK, may be involved in PRL plus IFN--induced activation of these transcription factors. Although MEK/ERK inhibitor U0126 did not reduce STAT1, NF-B, or IRF-1 activities induced by IFN- alone, it suppressed those induced by PRL plus IFN-, indicating that MEK/ERK may be selectively required for the synergistic effects of PRL. The inhibitory effects of these kinase inhibitors on NF-B, STAT1, or IRF-1 activities (Fig. 5) paralleled those on chemokine production (Fig. 4); IFN--induced CXCL9, CXCL10, and CXCL11 production was suppressed by JAK inhibitor I and SB203580, whereas that induced by IFN- plus PRL was suppressed by JAK inhibitor I, U0126, and SB203580. This indicates that the target signals such as JAKs, p38 MAPK, or MEK/ERK may promote chemokine production via activating STAT1, NF-B, or IRF-1 in IFN- and/or PRL-stimulated keratinocytes.
The transcriptional activity of STAT1 is regulated by phosphorylation; phosphorylation of STAT1 at Tyr701 induces its dimerization, nuclear translocation, and binding to DNA (25). Phosphorylation of STAT1 at Ser727 is essential to maximize its transactivation potential (25). We thus examined the involvement of signaling pathways in Tyr701 or Ser727 phosphorylation of STAT1. As shown in Fig. 6A, IFN- alone remarkably or PRL alone modestly induced Tyr701 phosphorylation of STAT1 (the increase of phospho/total STAT1 ratio was 0.44 or 0.20, respectively), and the addition of both cytokines gave additive effects on the tyrosine phosphorylation (the increase was 0.64). The STAT1 tyrosine phosphorylation by IFN- and/or PRL was suppressed by JAK inhibitor I (Fig. 6A). IFN- alone induced Ser727 phosphorylation of STAT1 (Fig. 6B), which was suppressed by SB203580 and JAK inhibitor I, indicating the requirement for p38 MAPK and JAKs for the Ser727 phosphorylation by IFN-. PRL alone modestly induced Ser727 phosphorylation of STAT1, which was suppressed by U0126 and JAK inhibitor I (Fig. 6B), indicating the requirement for MEK/ERK and JAKs for the Ser727 phosphorylation by PRL. The addition of IFN- and PRL gave additive effects on Ser727 phosphorylation of STAT1, which was suppressed by JAK inhibitor I, U0216, or SB203580 (Fig. 6B).
NF-B p50/p65 is normally sequestered in cytoplasm by its interaction with IB, and upon stimulation, IB is phosphorylated, ubiquitinated, and degraded, leading to the release of active NF-B (26). Phosphorylation of NF-B p65 at Ser536 is reported to enhance its transactivation capacity (27). We thus examined the involvement of JAKs, MEK/ERK, or p38 MAPK in phosphorylation or degradation of IB or phosphorylation of p65 in IFN- and/or PRL-stimulated keratinocytes. IFN- alone induced Ser32 phosphorylation of IB (Fig. 6C) and Ser536 phosphorylation of p65 (Fig. 6D), both of which were suppressed by JAK inhibitor I but not by U0126 or SB203580, indicating the involvement of JAKs. PRL alone modestly induced phosphorylation of IB (Fig. 6C) and p65 (Fig. 6D), both of which were suppressed by U0126 and JAK inhibitor I. PRL plus IFN-, compared with either cytokine alone, provided additive effects on phosphorylation of IB (Fig. 6C) or p65 (Fig. 6D), which was suppressed by JAK inhibitor I or U0126.
The transcriptional activity of IRF-1 depends on its expression level (18). IFN- alone increased the IRF-1 mRNA level, which was suppressed by JAK inhibitor I or SB203580 (Fig. 6E). Although PRL alone did not increase the IRF-1 mRNA level, PRL enhanced IFN--induced IRF-1 mRNA expression, which was suppressed by U0126 in addition to JAK inhibitor I or SB203580, indicating the involvement of MEK/ERK in the synergistic effect of PRL (Fig. 6E).
These results totally suggest that p38 MAPK is involved in STAT1 Ser727 phosphorylation and IRF-1 expression induced by IFN-. On the other hand, MEK/ERK may be involved in Ser727 phosphorylation of STAT1, phosphorylation of IB and p65, and IRF-1 expression induced by PRL alone or together with IFN-. JAKs may be involved in all of these effects and Tyr701 phosphorylation of STAT1 by IFN- and/or PRL.
IFN- activates JAK1/JAK2 and p38 MAPK, whereas PRL activates JAK2 and MEK/ERK
We examined whether PRL and/or IFN- may activate JAK1, JAK2, p38 MAPK, or MEK/ERK in keratinocytes. The activation of JAK1 or JAK2 induces their autophosphorylation at tyrosine residues (23), whereas p38 MAPK and ERK are activated by dual phosphorylation at threonine and tyrosine residues (28). IFN- alone induced autophosphorylation of both JAK1 (Fig. 7A) and JAK2 (Fig. 7B), which was suppressed by JAK inhibitor I. PRL alone did not promote tyrosine phosphorylation of JAK1 either in the presence or absence of IFN- (Fig. 7A) but did, however, modestly induce that of JAK2 (Fig. 7B), which was suppressed by JAK inhibitor I. The addition of IFN- and PRL gave additive effects on JAK2 tyrosine phosphorylation (Fig. 7B). PRL alone enhanced phosphorylation of ERK (Fig. 7C), whereas IFN- did not alter its phosphorylation either in the presence or absence of PRL. IFN- alone enhanced phosphorylation of p38 MAPK, whereas PRL did not alter its phosphorylation either in the presence or absence of IFN- (Fig. 7D). The PRL-induced phosphorylation of ERK- or IFN--induced phosphorylation of p38 MAPK was completely suppressed by U0216 or SB203580, respectively, and both were partially suppressed by JAK inhibitor I (Fig. 7, C and D). These results suggest that IFN- activates JAK1, JAK2, and p38 MAPK, dependent on JAK1/2, whereas PRL activates JAK2 and MEK/ERK, dependent on JAK2. These signals may cooperatively promote the transcriptional activities of STAT1, NF-B, and IRF-1, leading to the promotion of CXCL9, CXCL10, and CXCL11 production.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-induced CXCL9, CXCL10, and CXCL11 production in human keratinocytes. IFN- activated JAK1/JAK2 and p38 MAPK, whereas PRL activated JAK2 and MEK/ERK pathways. The PRL-induced signals did not per se induce the chemokine expression but interacted with the IFN--induced signals at several points to activate STAT1, NF-B, or IRF-1 and synergistically promoted CXCL9, CXCL10, and CXCL11 expression dependent on these transcription factors.
IFN- induced IRF-1 expression, which was further enhanced by PRL (Fig. 6E). The increase in the IRF-1 level led to the increase of its transcriptional activity. IRF-1 promoter contains the NF-B element and -activated sequence that binds STAT1 (29). Thus, NF-B and STAT1 may cooperatively drive IRF-1 transcription and mediate synergy between IFN- and PRL. CXCL11 promoter contains elements for binding NF-B, STAT1, and IRF-1 (19), and all these elements may functionally mediate CXCL11 transcription by PRL in synergy with IFN-. On the other hand, CXCL9 and CXCL10 promoters contain elements for NF-B and STAT1 but not for IRF-1 (20, 30). Thus, NF-B and STAT1, but not IRF-1, may cooperate on CXCL9 or CXCL10 promoters and synergistically enhance the transcription in IFN- plus PRL-stimulated keratinocytes.
PRL selectively activates JAK2 in keratinocytes (Fig. 7B). It is reported that PRLR interacts with JAK2 via a proline-rich motif proximal to its transmembrane domain (21). Upon PRL binding, PRLR may dimerize and associate with and activate JAK2, and several tyrosine residues of PRLR may be phosphorylated by the activated JAK2. The phosphorylated receptor tyrosine residues may provide binding sites for the src homology domain 2 (SH2) of STAT1. Thus, STAT1 may be recruited to PRLR and phosphorylated at Tyr701 by the PRLR-associated JAK2 (21). On the other hand, IFN- activates both JAK1 and JAK2 in keratinocytes (Fig. 7, A and B). It is reported that JAK2 or JAK1 is associated with IFN- R2 or R1, respectively (22). Upon IFN- binding to the receptor and receptor dimerization, JAK2 phosphorylates itself and JAK1 and increases both kinase activities. JAK1 then phosphorylates IFN- R1, creating a docking site for STAT1, and recruits STAT1, whereas JAK2 phosphorylates STAT1 (22). Such interdependent effects of JAK1 and JAK2 may induce highly effective tyrosine phosphorylation of STAT1. The STAT1 tyrosine-phosphorylating effect by PRL was moderate and may not per se increase its transcriptional activity but may sustain or supplement the effect by IFN- and result in the synergistic enhancement of its transcriptional activity.
PRL activated the MEK/ERK pathway in keratinocytes, and the activation was dependent on JAK2 (Fig. 7C). This is possibly because JAK2 may recruit and activate signaling molecules leading to ERK by tyrosine phosphorylating itself or PRLR. In PRL-stimulated mammary epithelial cells, tyrosine-phosphorylated JAK2 recruits an adaptor protein src homology collagen protein (SHC) containing SH2 (31). The recruited SHC is then tyrosine phosphorylated by JAK2 and acts as a linker for other SH2-containing signaling molecules. The tyrosine-phosphorylated SHC further recruits Grb2/Sos complex, a guanine nucleotide exchange factor activating p21 ras, which stimulates the raf-1/MEK/ERK pathway in these cells (31). A similar mechanism may act in PRL-stimulated keratinocytes.
The PRL-activated ERK or its downstream kinase appeared to phosphorylate STAT1 at Ser727 (Fig. 6B). On the other hand, serine phosphorylation of STAT1 by IFN- was mediated by p38 MAPK or its downstream kinase (Fig. 6B). Kinases phosphorylating STAT1 at Ser727 may thus vary with the stimuli or cell types (32, 33). The activation of p38 MAPK by IFN- may involve Ca2+-dependent proline-rich tyrosine kinase 2 (Pyk2) because Pyk2 preferentially activates p38 MAPK (34). IFN- stimulates Pyk2, which further activates the signaling cascade of MEK kinase kinase 4/MAPK kinase 6, an upstream pathway of p38 MAPK in immortalized keratinocyte HaCaT cells (24). PRL may at least additively augment the STAT1 serine phosphorylation by IFN- and, in concert with its tyrosine-phosphorylating effect, synergistically enhance the transcriptional activity of STAT1.
Although p38 MAPK was required for IFN--induced activation of NF-B (Fig. 5B), it is not responsible for IFN--induced phosphorylation of IB or Ser536 phophorylation of p65 (Fig. 6, C and D). Such effects may possibly be mediated by the other IFN--induced and JAK-dependent signals such as protein kinase C or double-stranded RNA-dependent protein kinase (35, 36). p38 MAPK may activate NF-B in different ways; mitogen- and stress-activated protein kinase-1, downstream of p38 MAPK, may phosphorylate NF-B p65 at Ser276 and increase its transactivation capacity (37). Alternatively, p38 MAPK may mediate phosphorylation of TATA-binding protein, a general transcriptional apparatus required for NF-B-mediated transcription (38). On the other hand, MEK/ERK activated by PRL induced phosphorylation of IB or Ser536 phosphorylation of p65 (Fig. 6, C and D). It is reported that ERK phosphorylates and stabilizes IB kinase-ß, which phosphorylates IB or p65 (27) or activates 90-kDa ribosomal S6 kinase, which phosphorylates IB (39). Similar effects of ERK may work in PRL-stimulated keratinocytes. The phosphorylation of IB and p65 by PRL may occur in parallel with that by IFN- and consequently converge on the activation of NF-B and the NF-B-dependent transcription of IRF-1 and chemokines.
In addition to the present results in neonatal keratinocytes, we have obtained the preliminary results that PRL potentiates the IFN--induced production of CXCL9, CXCL10, and CXCL11 in normal adult human keratinocytes (our unpublished observation). Our results in vitro suggest that PRL may in vivo promote CXCL9, CXCL10, and CXCL11 production by keratinocytes in psoriatic lesions where PRL and IFN- may coexist. We should further examine whether PRL may reproduce such effects in vitro on keratinocytes derived from psoriatic lesions. CXCL9, CXCL10, and CXCL11 released from epidermal keratinocytes bind CXCR3 on activated type 1 T cells in the dermis, inducing their migration toward the epidermis along a gradient of chemokines (3). The production of these chemokines by keratinocytes may thus generate abundant infiltrates of type 1 T cells. Because T cells can secrete PRL (5) and PRL also enhances IFN- production in T cells (5), IFN- and PRL, derived from the infiltrated T cells, may act on keratinocytes and cooperatively enhance their CXCL9, CXCL10, or CXCL11 production, indicating positive feedback control. Thus, PRL may potentiate the cross-talk between keratinocytes and type 1 T cells in cooperation with IFN-. Our results also support that PRL can be a candidate therapeutic target for psoriasis. One antipsoriatic drug, cyclosporin A, is known to suppress PRL binding to the PRLR (40). This agent may thus suppress PRL-induced CXCL9, CXCL10, or CXCL11 production in keratinocytes, indicating one possible mechanism of its therapeutic efficacy for psoriasis. Additional studies should also aim to develop more specific therapy using antibodies, small interfering RNA, or antisense oligonucleotides against PRL.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Disclosure Statement: The authors have nothing to disclose.
First Published Online January 25, 2007
Abbreviations: CXCL, CXC ligand; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; IB, inhibitory B; IFN-, interferon-; IRF, IFN regulatory factor; JAK, Janus kinase; JNK, c-Jun N-terminal kinase; KBM, keratinocyte basal medium; KGM, keratinocyte growth medium; MEK, MAPK/ERK kinase; NF-B, nuclear factor-B; PRL, prolactin; PRLR, PRL receptor; pRL-tk, herpes simplex virus thymidine kinase promoter-linked renilla luciferase vector; Pyk2, proline-rich tyrosine kinase 2; SH2, src homology domain 2; SHC, src homology collagen protein; STAT, signal transducer and activator of transcription.
Received December 6, 2006.
Accepted for publication January 18, 2007.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-induced production of CXCL9, CXCL10, and CXCL11 in human keratinocytes. Eur J Immunol 37:338-350 by natural killer and T cells. Ann NY Acad Sci 917:505–513 to in rat Nb2 pre-T lymphoma cells. Biochem Biophys Res Commun 220:958–962[CrossRef][Medline] (IFN-)-inducible chemokines IFN-inducible T-cell chemoattractant and monokine induced by IFN- and of their receptor CXC receptor 3 in human renal cell carcinoma. Cancer 103:258–267[CrossRef][Medline]B. Cell Mol Life Sci 61:1167–1175[CrossRef][Medline]B activity is required to elicit interferon--induced expression of chemokine CXC ligand 9 (CXCL9) and CXCL10 in human tumour cell lines. Biochem J 376:393–402[CrossRef][Medline]B activation mediates the cross-talk between extracellular matrix and interferon- (IFN-) leading to enhanced monokine induced by IFN- (MIG) expression in macrophages. J Biol Chem 277:43757–43762-inducible gene expression but not an antiviral state. EMBO J 15:799–809[Medline], but not IFN-, signaling. EMBO J 18:2480–2488[CrossRef][Medline]-dependent tyrosine phosphorylation of MEKK4 via Pyk2 is regulated by annexin II and SHP2 in keratinocytes. Biochem J 388:17–28[CrossRef][Medline]B is a target for the mitogen-activated 90 kDa ribosomal S6 kinase. EMBO J 16:3133–3144[CrossRef][Medline]B activation and transformation in resistant JB6 cells. Carcinogenesis 25:1991–2003-containing complexes play a predominant role in induction of IFN--inducible protein, 10 kDa (IP-10) by IFN- alone or in synergy with TNF-. J Immunol 161:4736–4744- and interferon ß-stimulated gene expression through STAT proteins. Science 269:1721–1723 and IFN- on phosphorylation of p65 and p50 (rel) in the K562 cell line: implications for altered interaction with RXRß. Cytokine 8:357–364[CrossRef][Medline]B by tumor necrosis factor- and -interferon in preneuronal cells. J Biol Chem 274:4801–4806B p65 subunit by mitogen- and stress-activated protein kinase-1 (MSK1). EMBO J 22:1313–1324[CrossRef][Medline]B-dependent gene expression. The role of TATA-binding protein (TBP). J Biol Chem 274:30858–30863B pathway mediates phorbol 12-myristate 13-acetate-induced megakaryocytic differentiation of K562 cells. J Biol Chem 276:13186–13191This article has been cited by other articles:
 |
|
 |
|
  K. Kurozumi, J. Hardcastle, R. Thakur, M. Yang, G. Christoforidis, G. Fulci, F. H. Hochberg, R. Weissleder, W. Carson, E. A. Chiocca, et al. Effect of Tumor Microenvironment Modulation on the Efficacy of Oncolytic Virus Therapy J Natl Cancer Inst, December 5, 2007; 99(23): 1768 - 1781. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
, P < 0.05 vs. IFN-
, P < 0.05 vs. IFN-
, P < 0.05 vs. IFN-
