BIOLOGY OF REPRODUCTION 74, 699–705 (2006) Published online before print 21 December 2005. DOI 10.1095/biolreprod.105.045450
Regulation of the MAFF Transcription Factor by Proinflammatory Cytokines in Myometrial Cells1 Wael Massrieh,3,4 Anna Derjuga,3 Florence Doualla-Bell,3,4 Chun-Ying Ku,5,6 Barbara M. Sanborn,5,6 and Volker Blank2,3,4 Lady Davis Institute for Medical Research3 and Department of Medicine,4 McGill University, Montreal, Quebec, Canada H3T 1E2 Departments of Biochemistry and Molecular Biology,5 University of Texas Medical School, Houston, Texas 77030 Biomedical Sciences,6 Colorado State University, Fort Collins, Colorado 80523 transforming agent of an avian retrovirus [1–3]; v- maf and its human cellular counterpart MAF (previously known as c-Maf) have been implicated in tumorigenesis [4]. There are two subgroups of MAF proteins, the large and the small MAF. The large MAF, containing an activation domain, include MAF, MAFA, MAFB, and NRL proteins [5–9]. Large MAF play important roles in the differentiation and development of a variety of hematopoietic and neural tissues [1, 3]. The small MAF, devoid of an obvious transactivation domain, include MAFF, MAFG, and MAFK factors [10, 11]. A unique domain located adjacent to the basic domain, called the extended homology region, is common to all MAF transcription factors and has been shown to be important for their DNA binding specificity [12]. Small MAF proteins can form heterodimers with the CNC (cap’n’ collar) proteins, which also belong to the bZIP family of transcription factors. The resulting CNC/small MAF complexes have been shown to recognize NFE2 (nuclear factor erythroid derived 2)-, MARE (MAF recognition element)-, and ARE (antioxidant response element)-type of DNA binding motifs [1, 3]. The small MAF are widely expressed, but transcript levels vary considerably in different tissues [1]. Small MAF homodimers have been reported to function as transcriptional repressors [13–15]. In contrast, it was reported that CNC/small MAF heterodimers serve as transcriptional activators of gene expression [1, 16–18]. Small MAF play a role in the cellular response to hydrogen peroxide, heavy metals, and electrophiles [19–21]. Data from several studies suggest that CNC/small MAF heterodimers are involved in the regulation of detoxifying enzyme genes and in the response to oxidative stress [14, 15, 17, 18, 22, 23]. Targeted disruption of the MafK and MafF loci yielded null mice that were indistinguishable from their wild-type littermates, indicating that these small MAF factors are dispensable in vivo and are replaced by other small MAF [24, 25]. MafG null mice are also viable but suffer from a mild thrombocytopenia [26]. Thus, functional redundancy has made it difficult to assign specific in vivo roles to each of the small MAF transcription factors. The human MAFF protein has previously been identified in a one-hybrid assay as a factor binding to regulatory sequences of the oxytocin receptor gene [27]. MAFF transcripts are highly expressed in term myometrium but are not present in early gestation (14 wk) and nonpregnant myometrial tissue [27]. It was hypothesized that the MAFF transcription factor may be involved in uterine gene regulation at the onset of parturition. Recent evidence suggests that cytokines may play an important role in the events that lead to preterm labor and delivery, causing morbidity and mortality of the fetus and serious health complications for the mother [28, 29]. Increased cytokine levels due to uterine infections, such as in chorioamnionitis,
ABSTRACT The MAF (proto-)oncogene family of basic-leucine zipper transcription factors plays crucial roles in the control of mammalian gene expression and development. Here we analyzed the regulation of the human MAFF gene, coding for a small MAF transcription factor, in uterine smooth muscle cells. We found that MAFF transcript levels are induced by proinflammatory cytokines in PHM1–31 myometrial cells. We observed an important induction by interleukin 1 beta (IL1B) and a weaker upregulation by tumor necrosis factor (TNF), whereas interleukin 6 (IL6) treatment had no effect. Time course experiments revealed a rapid induction of MAFF transcripts within 30 min following IL1B treatment. The presence of actinomycin D inhibited the upregulation, suggesting that regulation of MAFF mRNA levels occurs at the transcriptional level. We generated a MAFF-specific antiserum and determined that MAFF protein was also induced by TNF and IL1B in PHM1– 31 cells. In contrast, it was particularly interesting that the transcript and protein levels of the highly homologous MAFG and MAFK genes are not modulated by these cytokines. Our results suggest a possible specific role for MAFF in proinflammatory cytokine-mediated control of myometrial gene expression and provide the first link between a small MAF transcription factor and the inflammatory response. CNC, cytokines, gene regulation, IL1B, MAF, parturition, signal transduction, TNF, transcription factor, uterus
INTRODUCTION MAF (v-maf musculoaponeurotic fibrosarcoma oncogene homolog [avian]) basic-leucine zipper (bZIP) transcription factors play crucial roles in gene regulation, differentiation, oncogenesis, and development in many organisms. The founding member, the v-maf oncogene, was isolated as the 1
Supported by McGill Cancer Consortium and J.W. McConnell McGill Major Research studentships to W.M. Also supported by NIH award (HD09618) to B.M.S. V.B. supported by a Chercheur Boursier Fonds de la Recherche en Sante´ du Quebec and a McGill University Charles O. Monat award. Also supported by grants from the Hospital for Sick Children Foundation/Institute for Human Development, Child and Youth Health-CIHR and from the Cancer Research Society Inc. to V.B. 2 Correspondence: Volker Blank, Lady Davis Institute for Medical Research, Department of Medicine, McGill University, 3755 Cote Sainte-Catherine Road, Montreal, Quebec, Canada H3T 1E2. FAX: 514 340 7573; e-mail:
[email protected] Received: 15 July 2005. First decision: 19 August 2005. Accepted: 12 December 2005. Ó 2006 by the Society for the Study of Reproduction, Inc. ISSN: 0006-3363. http://www.biolreprod.org
699
700
MASSRIEH ET AL.
FIG. 1. PHM1–31 cells. Phase contrast image of the immortalized pregnant human myometrial cell line PHM31–1 in culture at passage 19. Original magnification 3200.
may cause the early onset of contractions [29]. The proinflammatory cytokines interleukin 1 beta (IL1B), interleukin 6 (IL6), and tumor necrosis factor (TNF) and the chemotactic factor IL8 have been implicated as mediators in both preterm and term labor, particularly in association with intrauterine infection. Elevated levels of these inflammatory cytokines have been observed in the amniotic fluid following infections ascending from the vagina and cervix [28]. Furthermore, it has been reported that expression of the oxytocin receptor is regulated by IL1B, IL6, and TNF in uterine smooth muscle cells [29–31]. Hence, there appears to be a link between the presence of inflammatory cytokines in the uterus or amnion and the induction of labor. Based on these previous results, we investigated the regulation of the MAFF transcription factor by proinflammatory cytokines. We found that both IL1B and TNF can rapidly induce MAFF transcript and protein levels in myometrial cells, whereas the expression of the highly homologous MAFG and MAFK genes is not modulated. Inhibitor studies suggest that regulation of the MAFF gene takes place at the transcriptional level. Since proinflammatory cytokines have been implicated as mediators in both preterm and term labor, we speculate that MAFF might participate in the control of these processes. Our results suggest that transcriptional complexes comprising MAFF may play a role in the regulation of uterine gene expression, linking for the first time small MAF transcription factor-mediated gene regulation to cytokine signaling and inflammation. MATERIALS AND METHODS Cell Culture Human embryonic kidney 293T (HEK293T, ATCC) cells were cultured in minimal essential medium alpha, 10% bovine serum, plus penicillin and streptomycin antibiotics.
Derivation of PHM1–31 Cells PHM1–31 myometrial cells were derived from the immortalization mixture used to derive PHM1–41 cells [32]. Briefly, myometrium was removed from the upper edge of the uterine segment at the time of caesarian section in a term pregnant patient who was not in labor, using a protocol that had received prior institutional approval (University of Texas Health Sciences Center, Houston). Myometrial cells were isolated by enzymatic digestion, infected at passage 2 with replication-defective adenovirus vector pLXSN16E6E7 (J.K. McDougal, University of Washington, Seattle) expressing the E6/E7 proteins of human
FIG. 2. Small MAF expression in the PHM1–31 myometrial cell line. Northern analysis of total RNA from PHM1–31 cells induced for 3 h with IL1B (10 ng/ml), IL6 (10 ng/ml), and TNF (10 ng/ml) or combinations of these cytokines. Ten micrograms of total RNA per lane were loaded. MAFF (A), MAFG (B), and MAFK (C) and ACTB transcripts are indicated. Quantification of at least three independent experiments is shown. Statistically significant differences are indicated by one (P , 0.05) or two (P , 0.01) asterisks. papilloma virus 16 and the neomycin resistance gene, and selected with the neomycin analog Geneticin (Invitrogen) [32]. PHM1–31 cells were recovered by subcloning foci and were maintained at 378C in high-glucose DMEM containing 0.1 mg/ml Geneticin, 10% fetal bovine serum, 2 mM L-glutamine, 50 U/ml penicillin, and 50 lg/ml streptomycin. The cells reach confluency 3–4 days after being cultured at 2 3 105 cells/100-mm dish and have been used in this study up to passage 25. The cells maintain smooth muscle a-actin expression at passage 27 and retain the ability to respond to oxytocin and thapsigargin with an increase in intracellular calcium.
Cytokine and Actinomycin D Treatment Cells reaching 80%–90% confluency were FBS starved for 16 to 24 h. They were then treated with TNF (100 ng/ml; Medicorp), IL1B (0.005–100 ng/ml; Research Diagnostics), or IL6 (100ng/ml; RDI) or combinations of these cytokines. Induction time was as indicated in the figures. For the inhibition experiments, the cells were starved overnight, and then actinomycin D (5 lg/ml) dissolved in dimethyl sulfoxide (DMSO) (0.5%) or DMSO alone (0.5%) was added to the cells 1 h before beginning the 3-h treatment with IL1B (1 ng/ml).
Cloning and Northern Analysis The human MAFF and MAFK cDNAs, comprising the entire coding region, were recovered by RT-PCR using the following primer pairs: MAFF-F2 (5 0 GGGCACCTTCTGCAAACATGT-3 0 ) and MAFF-R2 (5 0 -GAGGCGGCGCTCAGGCACTTT-3 0 ) and hMAFK-F1 (5 0 -CGTGCCCGGGTTATGACGACT3 0 ) and hMAFK-R1 (5 0 -GGCCGGCACTAGGATGCAGC-3 0 ). The resulting PCR fragments were cloned into the pCR-BluntII-TOPO vector (Invitrogen). To obtain MAFF- and MAFK-specific probes for Northern analysis, we performed an EcoRI digestion of the pCR-BluntII-MAFF and pCR-BluntIIMAFK plasmids, respectively. For MAFG, we amplified a 407-base-pair cDNA fragment with the primers MAFG611F (5 0 -CCGATCGTAGGGACGCGCGT3 0 ) and MAFG1018R (5 0 -CCACTCGGGAGTGGAGGGAA-3 0 ) using the pMT2 MAFG vector as template [33] and cloned it into the pCR-BluntII-TOPO vector. The resulting plasmid was digested with EcoRI to obtain a MAFG specific probe. For Northern analysis, total RNA was isolated using the protocol described for Trizol (Invitrogen). After subjecting RNA to gel electrophoresis, the gel was transferred overnight using standard techniques. The membrane was hybridized as described [34].
Transfection For transfection studies, the MAFF and MAFK cDNA fragments, obtained by EcoRI digestion of the pCR-BluntII-MAFF and pCR-BluntII-MAFK plasmids, were cloned into the EcoRI site of the pMT2 expression vector. The pMT2 MAFG plasmid has been described previously [33]. Transient transfections of HEK293T cells were performed using the standard calciumphosphate coprecipitation procedure. We used 10 lg of the pMT2 MAFF,
701
MAFF REGULATION BY CYTOKINES
FIG. 3. Dose response and time course of MAFF induction by cytokines. A) Dose response of MAFF transcript levels to IL1B (0.005–100 ng/ml) in PHM1–31 cells. Ten micrograms of total RNA per lane was loaded. MAFF and ACTB (control) transcripts are indicated. Quantification of four independent experiments is shown using nonlinear regression analysis. B) Short time course analysis of MAFF mRNA induction by Il1B (1 ng/ml) in PHM1–31 cells. Ten micrograms of total RNA per lane was loaded. MAFF and ACTB (control) transcripts are indicated. Quantification of four independent experiments is shown using nonlinear regression analysis. C) Extended time course analysis of MAFF mRNA induction by Il1B (100 ng/ml) in PHM1–31 cells. Ten micrograms of total RNA per lane were loaded. MAFF and ACTB (control) transcripts are indicated. Figure is representative of three independent experiments.
pMT2 MAFG, or pMT2 MAFK expression vectors to transfect a 100-mm dish of 30%–50% confluent HEK293T cells.
Generation of MAFF Antiserum To construct the GST-MAFF fusion protein vector, we generated a human MAFF cDNA using the primer pairs Bam-MAFF (5 0 -GTCGGATCCATGTCTGTGGATCCCCTATCC-3 0 ) and MAFF-Eco (5 0 -GCAGAATTCTAGGAGCAGGAGGCCGGGCC-3 0 ) and digested the amplicon with EcoRI and BamHI. Subsequently, the product was cloned into the EcoRI and BamHI sites of the pGEX-2TK fusion vector (Amersham). Rabbits at Pocono Rabbit Farm & Laboratory, Inc., were immunized with the purified GST fusion protein comprising the entire 164-amino-acid coding region of human MAFF. Preimmune serum was used to confirm the specificity of the generated antiserum to MAFF.
Immunoblot Analysis To prepare whole-cell extracts, the cells were scraped using 13 PBS and centrifuged, and the pellets were then resuspended in NB buffer (250 mM sucrose, 420 mM NaCl, 10 mM Tris/HCl, 2 mM MgCl2, 1 mM CaCl2, 1% Triton X-100) and protease inhibitors. After incubation on ice for 10 min,
samples were briefly centrifuged, and the supernatant was collected. Cytoplasmic extracts were generated using the same procedure as for wholecell extracts except that NaCl in buffer was omitted. Nuclear extracts were prepared by resuspending the pellet from the cytoplasmic extract preparation in NB buffer with 420 mM NaCl, centrifuging, and collecting the supernatant. The protein concentrations were determined using a protein assay kit (Bio Rad, Hercules, CA); 20–30 lg of the lysates were electrophoresed in 4%–12% NuPage Novex Bis-Tris (Invitrogen) (Fig. 6) or 12% SDS-polyacrylamide gels (Fig. 7). Resolved proteins were transferred electrophoretically to PVDF membranes (Millipore). After transfer, the membrane was blocked for 1 h at room temperature and was subsequently incubated overnight with polyclonal sera specific for either ACTB (actin, beta) (1:10 000) (Sigma #A-5441), human MAFF (1:200), human MAFG (1:200) [33], or human MAFK (1 lg/ml) (Santa Cruz). Secondary HRP-conjugated antibodies (Pierce) were applied for 1 h, and subsequently the proteins were visualized using chemiluminescence (Pierce).
Quantification and Statistical Analysis Quantification of Northern experiments was performed using a phosphoimager (Molecular Dynamics) and Image Quant software (version 5.2) or densitometry using NIH image software (version 1.63). All data are expressed
702
MASSRIEH ET AL.
FIG. 4. Inhibition of MAFF induction by actinomycin D. MAFF transcript levels in control and IL1B (1 ng/ml) induced PHM1–31 cells in the presence or absence of actinomycin D. The presence of dimethyl sulfoxide (DMSO), the vehicle for actinomycin D, is indicated. Three independent experiments were performed with similar results.
levels were induced by IL1B and TNF (data not shown). To minimize the need for primary myometrial tissue obtained from patients undergoing cesarian section, we used myometrial PHM1–31 cells in our subsequent studies. These cells were derived from primary myometrial cells by immortalization through overexpression of the human papilloma virus E6/E7 proteins [32]. The appearance of PHM1–31 cells was determined by phase contrast light microscopy (Fig. 1). The morphology of the cells is similar to that of regular proliferating smooth muscle cells. They appear long and spindle shaped with a central nucleus and a sheet-like growth pattern at confluency. We examined MAFF mRNA expression in PHM1–31 cells treated with proinflammatory cytokines (Fig. 2). We found a significant upregulation of MAFF mRNA levels by TNF and an even stronger induction by IL1B. In contrast, treatment with IL6 did not result in a significant change of MAFF mRNA levels (Fig. 2A). Combination of TNF and IL1B did not increase MAFF mRNA levels further than IL1B treatment alone (Fig. 2A). This result suggests that PHM1–31 cells, although immortalized, have conserved their myometrial phenotype with respect to cytokine responses. In contrast to MAFF, the transcript level of the highly homologous MAFG and MAFK genes were not induced by
as the mean 6 SEM and were analyzed using one-way analysis of variance or, where stated, nonlinear regression analysis. Values of P , 0.05 (*) or P , 0.01 (**) were used as criteria for declaring significance.
RESULTS Induction of MAFF Transcripts by Cytokines in PHM1–31 Myometrial Cells Proinflammatory cytokines such as IL1B, TNF, and IL6 are likely to play a role in regular and preterm labor [28, 29] as well as in the regulation of oxytocin receptor expression in myometrial cells [29, 30]. We hypothesized that MAFF expression may be modulated by cytokines. Our preliminary results in primary myometrial cells showed that MAFF mRNA
FIG. 5. Specificity of MAFF antiserum. Immunoblots of whole-cell extracts from HEK293T cells that are not transfected (mock) or transfected with constructs coding either for human MAFF, MAFG, or MAFK. The blots were incubated with antisera specific for each of the small MAF. Twenty micrograms of whole-cell extract was loaded per well. Small MAF are indicated by an arrow.
FIG. 6. Cellular localization of small MAF protein in myometrial cells. Immunoblot of whole-cell, cytoplasmic, or nuclear extracts prepared from PHM1–31 cells incubated with antisera specific for MAFF (A), MAFG (B), or MAFK (C) and ACTB. Twenty to 30 lg of protein were loaded per well.
MAFF REGULATION BY CYTOKINES
703
analyzing whole-cell extracts, nuclear and cytoplasmic extracts prepared from PHM1–31 cells (Fig. 6). In accordance with its function as a transcription factor, MAFF is found in the nucleus of PHM1–31 cells. As expected, MAFG and MAFK are equally found in the nucleus (Fig. 6, B and C). In addition, we examined whether the upregulation of MAFF transcripts also leads to an increase in MAFF protein levels. Analysis of nuclear extracts of cytokine-induced PHM1–31 cells showed that both TNF and IL1B induce MAFF protein expression (Fig. 7A). In contrast, expression of MAFG and MAFK protein was not affected by cytokine treatment (Fig. 7, B and C). Hence, cytokine regulation of MAFF mRNA and protein is coordinate (Fig. 2). DISCUSSION
FIG. 7. Small MAF protein levels following cytokine treatment. Immunoblot of nuclear extracts from PHM1–31 cells induced with cytokines for 3 h. The blots were incubated with antisera specific for either MAFF (A), MAFG (B), or MAFK (C) and ACTB. Twenty to 30 lg of protein were loaded per well.
IL1B or TNF, suggesting that stimulation by these cytokines is specific for the MAFF gene (Fig. 2, B and C). Regulation of MAFF Expression at the Transcriptional Level Dose-response experiments revealed that 0.005 ng/ml of IL1B upregulates MAFF gene expression, and maximal induction is reached with 0.5 ng/ml of IL1B (Fig. 3A). Experiments using PHM1–31 cells revealed induction of MAFF gene transcription by 30 min, reaching maximal levels by 1 h. Elevated MAFF transcript levels were sustained over a 24-h period (Fig. 3, B and C). In addition, actinomycin D inhibited induction of MAFF mRNA by IL1B (Fig. 4), suggesting regulation at the transcriptional level. Analysis of MAFF Protein Levels in Myometrial Cells To characterize MAFF biochemically, we generated an antiserum specific for this protein. We showed that the antiserum specifically recognizes MAFF, an 18-kDa protein, in cell extracts from human embryonic kidney 293T (HEK293T) cells transiently transfected with an expression vector coding for human MAFF, using immunoblot analysis (Fig. 5). The MAFF-specific antiserum does not cross-react with the homologous small MAF transcription factors MAFG and MAFK in immunoblot assays (Fig. 5). We investigated whether MAFF protein is expressed in myometrial cells
Members of the MAF and CNC families of bZIP transcription factors play diverse roles in mammalian gene regulation, differentiation, oncogenesis, and development [1, 3]. In the present study, we found an intriguing new link between proinflammatory cytokine signaling and MAF transcription factor gene regulation in uterine smooth muscle cells. In our studies, we examined MAFF expression in the myometrial cell line PHM1–31. We found that PHM1–31 cells mimic primary myometrial cells with respect to morphology and upregulation of MAFF in response to cytokines (Figs. 2, 3, and 7; data not shown) [35, 36]. We conclude that PHM1–31 cells provide a valuable model to study myometrial gene expression and function. Most important, our novel data show that MAFF mRNA is rapidly induced by IL1B and TNF in primary myometrial and PHM1– 31 cells (Figs. 2 and 3; data not shown). The presence of actinomycin D inhibits this induction (Fig. 4), suggesting regulation at the transcriptional level. The induction of MAFF also occurs at the protein level in the PHM1–31 cell line (Fig. 7). It is of interest to note that IL6 does not modulate MAFF transcript and protein levels, suggesting that the observed regulation is specific to a subset of cytokines. Moreover, the expression of the highly homologous MAFG and MAFK transcripts and proteins is not modulated by these cytokines (Figs. 2 and 7), suggesting a specific function of MAFF in the inflammatory response. This is of interest, as overlapping functions and functional redundancy in studies using in vitro, transgenic, and gene-targeting approaches have made it difficult to assign specific roles to each of the small MAF [13, 24–26, 33, 37–39]. Examining MAFF as well as double small MAF-deficient mice, using, for example, bacterial and viral induced inflammation models, may thus reveal novel functions for these transcription factors. The identification of novel targets of cytokine signaling in myometrial cells is important, as elevated levels of proinflammatory cytokines such as IL1B and TNF in the uterus or amnion are likely to function as mediators in both preterm and normal term labor, in particular in association with cervical and intrauterine infection in the mother [29, 40]. Elevated levels of IL1B are observed in the amniotic fluid only late in pregnancy, during the third trimester. With infection, the levels of IL1B, IL6, TNF, and IL8 rise even further, and this increase is linked to the early onset of labor [29, 41]. About one-third of preterm births are associated with infective processes, accounting for a high number of neonatal deaths and a higher incidence of brain, lung, and metabolic disorders [42]. Thus, a better understanding of proinflammatory cytokine signaling in gestational tissues is crucial. Based on these previous studies implicating cytokines in normal and preterm labor, we hypothesize that MAFF may participate in these processes. This is further
704
MASSRIEH ET AL.
supported by our finding that prostaglandin-endoperoxide synthase 2 (PTGS2, previously known as cyclooxygenase-2) mRNA, proposed to be a major regulator of gestation and parturition [36], is stimulated by IL1B and TNF but not IL6 in the PHM1–31 cell line (data not shown). Earlier reports have shown that PTGS2 expression is upregulated by IL1B in primary myometrial cells [43, 44]. The stimulation of MAFF expression by IL1B and TNF may also be important with respect to the regulation of the oxytocin receptor gene, a putative MAFF target [27], by cytokines. The oxytocin-signaling pathway may play an important role in the control of parturition, and its abnormal regulation may be implicated in preterm labor [45]. There have been conflicting reports as to whether oxytocin receptor gene expression is increased, decreased, or not changed following treatment of myometrial cells with IL1B and IL6 [30, 31, 46, 47]. Thus, it is not clear how MAFF may be implicated in the control of the oxytocin pathway by cytokines. Transfection experiments using MAFF cDNA alone failed to modulate oxytocin receptor gene transcription [27]. This result may be explained by the absence of appropriate CNC factors, which function as binding partners of MAFF. Further studies using a dominant-negative MAFF factor or cotransfection experiments, expressing both MAFF and CNC proteins, should clarify whether the oxytocin receptor gene is regulated by MAFF. Previous reports showed that MAFF is able to interact with members of the CNC transcription factor family, including NFE2, NFE2-like 1, and NFE2-like 2 [13, 23]. We propose that CNC/MAFF heterodimers may be involved in the regulation of myometrial gene expression by proinflammatory cytokines. CNC/small MAF heterodimers have been shown to bind a variety of AP1-like binding motifs, including the NFE2, MARE, and ARE recognition sites [1, 48]. It has been proposed earlier that small MAF factors, which do not contain an activation domain, increase the binding site specificity of the complex, whereas the large CNC subunit confers the transcriptional activation properties to the heterodimer [11]. Thus, cytokine stimulation leading to increased MAFF levels may result in gene expression changes of a specific set of genes in myometrial cells. In conclusion, our studies suggest that the MAFF transcription factor may play a role in inflammatory responses in uterine smooth muscle cells. This is important because of the possible role that proinflammatory cytokines may play in preterm and normal term labor [29]. In future studies, it will be critical to identify the MAFF target genes in myometrial cells. This research should eventually lead to a better understanding of the molecular mechanisms governing small MAF factormediated regulation of gene expression and help to uncover the possible functions of MAFF in the inflammatory response in uterine smooth muscle cells. ACKNOWLEDGMENTS
5.
6.
7.
8. 9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21. 22.
23.
We would like to thank Benoıˆt Che´nais for helpful discussions and critical reading of the manuscript and Zaynab Nouhi for assistance with revision experiments. 24.
REFERENCES 1. Blank V, Andrews NC. The Maf transcription factors: regulators of differentiation. Trends Biochem Sci 1997; 22:437–441. 2. Motohashi H, Shavit JA, Igarashi K, Yamamoto M, Engel JD. The world according to Maf. Nucleic Acids Res 1997; 25:2953–2959. 3. Motohashi H, O’Connor T, Katsuoka F, Engel J, Yamamoto M. Integration and diversity of the regulatory network composed of Maf and CNC families of transcription factors. Gene 2002; 294:1. 4. Chesi M, Bergsagel PL, Shonukan OO, Martelli ML, Brents LA, Chen T,
25.
26.
27.
Schrock E, Ried T, Kuehl WM. Frequent dysregulation of the c-maf protooncogene at 16q23 by translocation to an Ig locus in multiple myeloma. Blood 1998; 91:4457–4463. Pouponnot C, Nishizawa M, Calothy G, Pierani A. Transcriptional stimulation of the retina-specific QR1 gene upon growth arrest involves a Maf-related protein. Mol Cell Biol 1995; 15:5563–5575. Kataoka K, Fujiwara KT, Noda M, Nishizawa M. MafB, a new Maf family transcription activator that can associate with Maf and Fos but not with Jun. Mol Cell Biol 1994; 14:7581–7591. Cordes SP, Barsh GS. The mouse segmentation gene kr encodes a novel basic domain-leucine zipper transcription factor. Cell 1994; 79:1025– 1034. Ogino H, Yasuda K. Induction of lens differentiation by activation of a bZIP transcription factor, L-Maf. Science 1998; 280:115–118. Swaroop A, Xu JZ, Pawar H, Jackson A, Skolnick C, Agarwal N. A conserved retina-specific gene encodes a basic motif/leucine zipper domain. Proc Natl Acad Sci U S A 1992; 89:266–270. Fujiwara KT, Kataoka K, Nishizawa M. Two new members of the maf oncogene family, mafK and mafF, encode nuclear b-Zip proteins lacking putative trans-activator domain. Oncogene 1993; 8:2371–2380. Andrews NC, Kotkow KJ, Ney PA, Erdjument-Bromage H, Tempst P, Orkin SH. The ubiquitous subunit of erythroid transcription factor NF-E2 is a small basic-leucine zipper protein related to the v-maf oncogene. Proc Natl Acad Sci U S A 1993; 90:11488–11492. Kusunoki H, Motohashi H, Katsuoka F, Morohashi A, Yamamoto M, Tanaka T. Solution structure of the DNA-binding domain of MafG. Nat Struct Biol 2002; 9:252–256. Igarashi K, Kataoka K, Itoh K, Hayashi N, Nishizawa M, Yamamoto M. Regulation of transcription by dimerization of erythroid factor NF-E2 p45 with small Maf proteins [see comments]. Nature 1994; 367:568–572. Dhakshinamoorthy S, Jaiswal AK. Small maf (MafG and MafK) proteins negatively regulate antioxidant response element-mediated expression and antioxidant induction of the NAD(P)H:Quinone oxidoreductase1 gene. J Biol Chem 2000; 275:40134–40141. Nguyen T, Huang HC, Pickett CB. Transcriptional regulation of the antioxidant response element: activation by Nrf2 and repression by MafK. J Biol Chem 2000; 275:15466–15473. Kotkow KJ, Orkin SH. Dependence of globin gene expression in mouse erythroleukemia cells on the NF-E2 heterodimer. Mol Cell Biol 1995; 15: 4640–4647. Wild AC, Moinova HR, Mulcahy RT. Regulation of gamma-glutamylcysteine synthetase subunit gene expression by the transcription factor Nrf2. J Biol Chem 1999; 274:33627–33636. Gong P, Hu B, Stewart D, Ellerbe M, Figueroa YG, Blank V, Beckman BS, Alam J. Cobalt induces heme oxygenase-1 expression by a hypoxiainducible factor-independent mechanism in Chinese hamster ovary cells: regulation by Nrf2 and MafG transcription factors. J Biol Chem 2001; 276:27018–27025. Crawford DR, Leahy KP, Wang Y, Schools GP, Kochheiser JC, Davies KJ. Oxidative stress induces the levels of a MafG homolog in hamster HA1 cells. Free Radic Biol Med 1996; 21:521–525. Suzuki T, Blank V, Sesay JS, Crawford DR. Maf genes are involved in multiple stress response in human. Biochem Biophys Res Commun 2001; 280:4–8. Moran JA, Dahl EL, Mulcahy RT. Differential induction of mafF, mafG and mafK expression by electrophile-response-element activators. Biochem J 2002; 361:371–377. Itoh K, Chiba T, Takahashi S, Ishii T, Igarashi K, Katoh Y, Oyake T, Hayashi N, Satoh K, Hatayama I, Yamamoto M, Nabeshima Y. An Nrf2/ small Maf heterodimer mediates the induction of phase II detoxifying enzyme genes through antioxidant response elements. Biochem Biophys Res Commun 1997; 236:313–322. Marini MG, Asunis I, Chan K, Chan JY, Kan YW, Porcu L, Cao A, Moi P. Cloning MafF by recognition site screening with the NF-E2 tandem repeat of HS2: analysis of its role in globin and GCSI genes regulation. Blood Cells Mol Dis 2002; 29:145–148. Kotkow KJ, Orkin SH. Complexity of the erythroid transcription factor NF-E2 as revealed by gene targeting of the mouse p18 NF-E2 locus. Proc Natl Acad Sci U S A 1996; 93:3514–3518. Onodera K, Shavit JA, Motohashi H, Katsuoka F, Akasaka J, Engel JD, Yamamoto M. Characterization of the murine mafF gene [in process citation]. J Biol Chem 1999; 274:21162–21169. Shavit JA, Motohashi H, Onodera K, Akasaka J, Yamamoto M, Engel JD. Impaired megakaryopoiesis and behavioral defects in mafG-null mutant mice. Genes Dev 1998; 12:2164–2174. Kimura T, Ivell R, Rust W, Mizumoto Y, Ogita K, Kusui C, Matsumura Y, Azuma C, Murata Y. Molecular cloning of a human MafF homologue,
MAFF REGULATION BY CYTOKINES
28. 29.
30.
31.
32.
33.
34. 35.
36. 37.
38.
which specifically binds to the oxytocin receptor gene in term myometrium. Biochem Biophys Res Commun 1999; 264:86–92. Guzick DS, Winn K. The association of chorioamnionitis with preterm delivery. Obstet Gynecol 1985; 65:11–16. Saji F, Samejima Y, Kamiura S, Sawai K, Shimoya K, Kimura T. Cytokine production in chorioamnionitis. J Reprod Immunol 2000; 47: 185–196. Schmid B, Wong S, Mitchell BF. Transcriptional regulation of oxytocin receptor by interleukin-1beta and interleukin-6. Endocrinology 2001; 142: 1380–1385. Rauk PN, Friebe-Hoffmann U, Winebrenner LD, Chiao JP. Interleukin-6 up-regulates the oxytocin receptor in cultured uterine smooth muscle cells. Am J Reprod Immunol (Copenhagen) 2001; 45:148–153. Monga M, Ku CY, Dodge K, Sanborn BM. Oxytocin-stimulated responses in a pregnant human immortalized myometrial cell line. Biol Reprod 1996; 55:427–432. Blank V, Kim MJ, Andrews NC. Human MafG is a functional partner for p45 NF-E2 in activating globin gene expression. Blood 1997; 89:3925– 3935. Church GM, Gilbert W. Genomic sequencing. Proc Natl Acad Sci U S A 1984; 81:1991–1995. Dong YL, Gangula PR, Fang L, Yallampalli C. Differential expression of cyclooxygenase-1 and -2 proteins in rat uterus and cervix during the estrous cycle, pregnancy, labor and in myometrial cells. Prostaglandins 1996; 52:13–34. Kniss DA. Cyclooxygenases in reproductive medicine and biology. J Soc Gynecol Investig 1999; 6:285–292. Motohashi H, Katsuoka F, Shavit JA, Engel JD, Yamamoto M. Positive or negative MARE-dependent transcriptional regulation is determined by the abundance of small Maf proteins. Cell 2000; 103:865–875. Onodera K, Shavit JA, Motohashi H, Yamamoto M, Engel JD. Perinatal
39.
40. 41. 42. 43. 44.
45. 46.
47. 48.
705
synthetic lethality and hematopoietic defects in compound mafG::mafK mutant mice. EMBO J 2000; 19:1335–1345. Motohashi H, Katsuoka F, Engel JD, Yamamoto M. Small Maf proteins serve as transcriptional cofactors for keratinocyte differentiation in the Keap1-Nrf2 regulatory pathway. Proc Natl Acad Sci U S A 2004; 101: 6379–6384. Gomez R, Romero R, Edwin SS, David C. Pathogenesis of preterm labor and preterm premature rupture of membranes associated with intraamniotic infection. Infect Dis Clin North Am 1997; 11:135–176. Keelan JA, Blumenstein M, Helliwell RJA, Sato TA, Marvin KW, Mitchell MD. Cytokines, prostaglanding and parturition—a review. Placenta 2003; 17:S33–S46. Challis JRG. Mechanism of parturition and preterm labor. Obstet Gynecol Surv 2000; 55:650–660. Rauk PN, Chiao JP. Interleukin-1 stimulates human uterine prostaglandin production through induction of cyclooxygenase-2 expression. Am J Reprod Immunol (Copenhagen) 2000; 43:152–159. Hertelendy F, Molnar M, Romero R. Interferon gamma antagonizes interleukin-1beta-induced cyclooxygenase-2 expression and prostaglandin E(2) production in human myometrial cells. J Soc Gynecol Investig 2002; 9:215–219. Zingg HH. Oxytocin and uterine activity. Front Horm Res 2001; 27:57– 65. Helmer H, Tretzmuller U, Brunbauer M, Kaider A, Husslein P, Knofler M. Production of oxytocin receptor and cytokines in primary uterine smooth muscle cells cultivated under inflammatory conditions. J Soc Gynecol Investig 2002; 9:15–21. Doualla-Bell F. Roˆle des cyclooxyge´nases placentaires et ute´terines. Reproduction Humaine et Hormones 2003; 16:39–45. Motohashi H, Yamamoto M. Nrf2-Keap1 defines a physiologically important stress response mechanism. Trends Mol Med 2004; 10:549– 557.
