- Email: [email protected]
Hypoxic induction of human visfatin gene is directly mediated by hypoxia-inducible factor-1
FEBS Letters 580 (2006) 4105–4113
Hypoxic induction of human visfatin gene is directly mediated by hypoxia-inducible factor-1 Soo-Kyung Baeb,e,1, Su-Ryun Kima,d,1, Jong Gab Kimb,e, Jee Yeon Kimb, Tae Hyeon Kooc, Hye-Ock Janga, Il Yuna, Mi-Ae Yood, Moon-Kyoung Baea,* a
College of Dentistry and Research Institute for Oral Biotechnology, Pusan National University, Pusan 602-739, South Korea b College of Medicine, Pusan National University, Pusan 602-739, South Korea c Department of Biology, Korea Science Academy, Pusan 614-822, South Korea d Department of Molecular Biology, Pusan National University, Pusan 609-735, South Korea e Medical Research Center for Ischemic Tissue Regeneration, Pusan National University, Pusan 602-739, South Korea Received 4 May 2006; revised 19 June 2006; accepted 23 June 2006 Available online 30 June 2006 Edited by Robert Barouki
Abstract Visfatin has been originally identified as a growth factor for early stage B cells and recently known as an adipokine. Here, we report that hypoxia induces the visfatin mRNA and protein levels in MCF7 breast cancer cells. We also demonstrate that induction of visfatin gene is regulated by hypoxia-inducible factor-1a (HIF-1a). Moreover, 5 0 -flanking promoter region of human visfatin gene contains two functional HIF responsive elements (HREs), activating the expression of visfatin. Mutation of these HREs in the visfatin promoter abrogates activation of a luciferase reporter gene driven by visfatin promoter under hypoxia. Taken together, our results demonstrate that visfatin is a new hypoxia-inducible gene of which expression is stimulated through the interaction of HIF-1 with HRE sites in its promoter region. 2006 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. Keywords: Visfatin; Pre-B-cell colony-enhancing factor; Hypoxia; Hypoxia-inducible factor-1a; Breast cancer cells
1. Introduction Adipokines are biologically active polypeptides growth factors and cytokines that appear to be preferentially produced by adipose tissue [1,2]. They exert their effects on the normal and neoplastic breast tissues by endocrine and paracrine mechanisms; in most cases, they are also secreted by cancerous epithelial cells, providing for autocrine regulation [3]. They can stimulate cancer epithelial cell proliferation, invasion, and angiogenesis which is essential for breast cancer development and progression [3]. Visfatin (pre-B-cell colony-enhancing factor) is firstly isolated from peripheral blood lymphocyte and recently characterized as a novel adipokine [4,5]. In addition, visfatin is
*
Corresponding author. Fax. +82 51 245 8388. E-mail address: [email protected] (M.-K. Bae). 1
Denotes co-first authors.
expressed on the various normal, inflamed and tumor tissues [5–9]. More recently, acute mechanical and multiple inflammatory stimuli increased visfatin expression [6,7], but precise mechanism(s) underlying this visfatin induction remains poorly resolved. Furthermore, molecular mechanism of visfatin gene regulation in tumor progression has not been elucidated yet. Hypoxia detected in the central regions of solid tumors can be a leading cause of angiogenesis, a fundamental determinant of malignant tumor progression [10,11]. A key factor in this process is hypoxia-inducible factor (HIF), which regulates transcription of hypoxia-activated genes and consists of HIF1a and HIF-1b heterodimer. The a subunit of HIF-1 is rapidly degraded under normoxic condition and stabilized under hypoxia, while HIF-1b is expressed constitutively [12]. Since the production of many adipokines including leptin, IL-6, VEGF, and heparin-binding epidermal growth factor-like growth factor are tightly regulated by hypoxia [13–17], we hypothesized that hypoxic microenvironment within tumor mass activates the expression of visfatin. In this present study, we investigated whether the expression of visfatin is upregulated in response to hypoxia and further examined whether the HIF-1 is involved in hypoxia-induced activation of human visfatin gene in breast cancer MCF7 cell line. Herein, our data provide evidence that hypoxic induction of visfatin gene is mediated by transactivation of HIF-1 through HIF-1 consensus binding site (HRE) in the proximal promoter of visfatin gene.
2. Materials and methods 2.1. Reagents and antibodies Cobalt chloride and L -mimosine were from Sigma. Mouse monoclonal anti-HIF-1a and a-tubulin antibodies were purchased from BD transduction laboratories and Biogenex, respectively. Rabbit and mouse monoclonal anti-visfatin antibodies were from AdipoGen. Alexa 594-conjugated goat anti-mouse IgG was purchased from Molecular Probes. 2.2. Cell culture MCF7 cells, Hepa-1c4 and Hepa-1c1c7 cells were maintained in DMEM supplemented with 10% FBS (Invitrogen) and 1% antibiotics. For hypoxic condition, cells were incubated at 5% CO2 level with 1% O2 balanced with N2 in hypoxic chamber (Forma Scientific).
0014-5793/$32.00 2006 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.febslet.2006.06.052
4106 2.3. Plasmids and constructs A genomic DNA fragment of the human visfatin gene [6] containing 2 kb of 5 0 -flanking region was prepared by PCR amplification of human genomic DNA. A 1949 bp PCR product was obtained and subcloned into a pCR2.1/TA vector (Invitrogen). From this construct, a full-length (p(1949)luc) and a deletion mutant of visfatin gene (p(708)luc) were amplified by PCR and subcloned into pGL3 luciferase reporter vector (Promega), respectively. All constructs were confirmed by automatic DNA sequencing analysis. 2.4. Real-time PCR Real-time PCR quantification was performed using a SYBR Green approach (Light Cycler; Roche Applied Science, Penzberg, Germany). Cycling parameters consisted of 1 cycle of 95 C for 10 min, followed by amplification for 30 cycles of 95 C for 10 s, 57 C for 5 s, and 72 C for 7 s. Subsequently, a melting curve program was applied with continuous fluorescence measurement. The entire cycling process including data analysis took less than 1 h and was monitored using the LightCycler software program (version 4.0). 2.5. Immunocytochemistry Cells cultured on a coverglass were fixed in 4% paraformaldehyde for 10 min, blocked with 0.5% Triton X-100/PBS for 5 min, and then labeled with appropriate primary antibodies and Alexa 594-conjugated secondary antibody. Coverslips were mounted in Vectastain containing DAPI (Vector Laboratories). Cells were analyzed using fluorescent microscopy (Nikon). 2.6. RNA interference experiment MCF7 cells were transfected with siRNA specific for HIF-1a (antisense: 5 0 -UCAAACACACUGUGUCCAG-3 0 ; sense: 5 0 -CUGGACACGUGUGUUUGA-3 0 ) at a concentration of 100 nM using the Lipofectamine 2000 (Invitrogen). After transfection, the cells were exposed to a normoxic or hypoxic atmosphere for the indicated times. 2.7. Transient transfection and reporter gene analysis Cells were plated on 24-well plates and transfected with the luciferase constructs and pCMV-b-gal using Lipofectamine 2000 (Invitrogen). On the following day, the cells were incubated with hypoxic condition for 20 h and then solubilized. Cell lysates were analyzed for b-galactosidase enzyme assay and for luciferase activity using assay kit (Promega) and luminometer (Turner Designs). Each extract was assayed at least three times, and relative luciferase activity was calculated as RLU/b-galactosidase. 2.8. Site-directed mutagenesis Six putative HIF-1 binding sites (HRE: 5 0 -RCGTG-3 0 ) within p(708)luc were targeted for mutagenesis. The mutations were made with a QuickChange site-directed mutagenesis kit (Stratagene) and checked by automatic DNA sequencing. 2.9. Chromatin immunoprecipitation assay Chromatin immunoprecipitation (ChIP) analysis was performed with the ChIP kit (Upstate Biotechnology) according to the manufacturer’s instructions with minor modifications and anti-HIF-1a monoclonal antibody (Novus). Chromatins were immunoprecipitated from 1 · 106 cells incubated under normoxia and hypoxia for 16 h, respectively. Sequences of promoter-specific primers included the visfatin promoter region 383 to 258 as follows: sense, 5 0 -AGAGCTGGCGTCTGGGAG-3 0 ; and anti-sense, 5 0 -GCCTTCACCCCGTCACCC-3 0 . 2.10. Immunohistochemistry Human breast cancer tissue specimens were prepared and thin sections (4 lm) from selected areas were used. In brief, after deparaffinization and blocking of endogenous peroxidase activity, antigen retrieval was routinely performed using PBS containing 0.04% alkaline protease solution (Promega). The primary monoclonal antibodies used in this study were mouse anti-pan-visfatin and rabbit anti-HIF-1a. Tissues were stained according to the standard avidin–biotin method using ABC reagent (Vectastain Elite kit, Vector Labs.) and counterstained with Harris hematoxylin.
S.-K. Bae et al. / FEBS Letters 580 (2006) 4105–4113
3. Results 3.1. Hypoxia increases human visfatin mRNA and protein levels To examine the effects of hypoxia on visfatin gene expression, MCF7 cells were incubated under hypoxic condition for periods of 16 h and 24 h. As shown in Fig. 1A, hypoxic treatment produced significant increases in visfatin mRNA levels. Using real-time RT-PCR, we also quantified the enhanced expression level of visfatin mRNA under hypoxia (Fig. 1B). Next, to analyze whether hypoxic condition could affect visfatin protein levels, Western immunoblot analysis was performed. As shown in Fig. 1C, 1% O2 and cobalt chloride, hypoxic mimic agent, increased visfatin protein levels in a time-dependent manner. The increased level of visfatin protein by hypoxia was confirmed by immunocytochemistry. As shown in Fig. 1D, a stronger staining of visfatin protein was observed in hypoxic MCF7 cells than in normoxic cells. These results indicated that hypoxia significantly upregulates visfatin mRNA and protein levels. 3.2. Hypoxic induction of visfatin mRNA and protein is dependent on HIF-1 HIF-1 is a key transcription factor of the transcriptional responses to hypoxia. To determine the crucial role of HIF-1a in hypoxia-mediated upregulation of visfatine gene, we transfected HIF-1a expression vector into normoxic MCF7 cells. As shown in Fig. 2A, the overexpressed HIF-1a increased the visfatin protein level even under normoxic condition, which shows similar increase in the expression of visfatin under hypoxia. The stability of HIF-1a protein is known to be markedly downregulated by prolyl hydroxylase domain protein (PHD) that hydroxylates HIF-1a under normoxia, followed by its polyubiquitination and proteasomal degradation. We found that treatment of normoxic MCF7 cells with L -mimosine, a well-recognized PHD inhibitor [18], induced visfatin mRNA levels (Fig. 2B). To confirm whether endogenous visfatin gene is one of direct HIF-1 target genes, we also used specific siRNA against HIF-1a. Transfection of MCF7 cells with siRNA against HIF-1a markedly reversed hypoxia-induced visfatin expression (Fig. 2C). Thus, these data demonstrated that HIF-1a is directly involved in the induction of the visfatin gene expression. 3.3. Transcriptional activation of the visfatin gene promoter by hypoxia We next focused on detailed analysis of human visfatin promoter to find out functional HIF-1 binding sites. Therefore, we searched putative HIF-1 binding sequences in 5 0 -flanking region of the human visfatin gene [6]. An extensive analysis of 5 0 visfatin promoter sequence (between positions 1949 and +1), using the computer database TRANSFAC, revealed the presence of several putative HIF-1 consensus-binding sites. The relative position and orientation of these elements are indicated at the top of Fig. 3. Based on this information, we cloned two fragments containing up to 1949 bp and up to 708 bp of the 5 0 -flanking region of the human visfatin gene, and then constructed luciferase reporter vectors, named as p(1949)luc and p(708)luc, respectively. These constructs were then introduced into MCF7 cells and incubated with hypoxia. As shown in Fig. 3A, reporter activities of both constructs under hypoxia were significantly increased compared to that of pGL3 alone. Because the transcriptional activity of p(708)luc
S.-K. Bae et al. / FEBS Letters 580 (2006) 4105–4113
4107
Fig. 1. Effect of hypoxia on the expression of visfatin gene. (A) MCF7 cells were incubated under 21% O2 or 1% O2 for the indicated times. RT-PCR analysis was performed using specific primers for human visfatin. b-actin served as an internal control. (B) MCF7 cells were incubated under normoxic or hypoxic conditions for 16 h or 24 h. Total RNAs were isolated and then analyzed by real-time PCR using visfatin primers. The expression level under normoxia (0 h hypoxia) was set to 100%, and the values are normalized to actin mRNA levels. (C) MCF7 cells were incubated for the indicated times under 1% O2 or treated with 100 lM cobalt chloride. Visfatin protein levels were examined by Western blotting using antivisfatin. a-tubulin served as loading control. (D) Photographs showed immunocytochemically staining cells with anti-pan visfatin antibody. The signals were detected using Alexa 594-conjugated secondary antibody (red).
under hypoxia is very similar to that of p(1949)luc, we expected that potential functional HREs exist on fragment from 708 to +1, and then we used this fragment (from 708 to +1) in the following experiments. To test whether exogenous HIF1a overexpression can cause the same stimulatory effects on visfatin promoter, HIF-1a expression vector was added with p(708)luc. As expected, transfection of HIF-1a expression vector stimulated reporter activity even in normoxic condition: this was further enhanced by the hypoxia (Fig. 3B). To determine whether the functional HIF-1 complex is required for transcriptional activation of visfatin by hypoxia, we took advantage of the mouse hepatoma Hepa-1c4 cells. This cell line is derived from parental Hepa-1c1c7 cells and deficient in HIF1b subunit, which fails to form functional HIF-1 heterodimer [19]. The p(708)luc reporter construct was transiently transfec-
ted into Hepa-1c4 cells and Hepa-1c1c7 cells under normoxic or hypoxic conditions. The luciferase activities indicated a substantial induction in hypoxic Hepa-1c1c7 cells, whereas no significant increase was observed in Hepa-1c4 cells in response to hypoxia (Fig. 3C). HIF-1a and HIF-2a are two highly related pHLH/PAS homologous transcription factors that undergo increased function at low oxygen levels [20]. Transcriptionally active heterodimer composed of HIF-2a and ARNT, like HIF-1a and ARNT, binds to HRE and activates hypoxia-induced genes [21]. To check whether HIF-2a might be involved in hypoxia-induced visfatin promoter activation, we performed a luciferase assay using an p(708)luc reporter and HIF-2a. The p(708)luc activity in the presence of HIF-1a was significantly increased, whereas no increase was observed in the presence of HIF-2a expression vector (Fig. 3D). Thus, functional
4108
S.-K. Bae et al. / FEBS Letters 580 (2006) 4105–4113
Fig. 2. The role of HIF-1a in the regulation of visfatin gene expression. (A) MCF7 cells were transfected with HIF-1a expression vector and left untreated or exposed to 1% O2 for 24 h. Anti-visfatin or HIF-1a Western blotting from protein extracts was performed. a-tubulin served as a loading control (left panel). The right panel showed densitometric analysis assessing relative visfatin protein expression level in at least three times independent experiments. The density of the control bands (21% O2 for 24 h) was set to 100%. (B) Cell lysates were prepared from MCF7 cells treated with 400 or 800 lM L -mimosine for 16 h. RT-PCR analysis was performed using specific primers for visfatin or b-actin. (C) MCF7 cells were transfected with siRNAs specific for HIF-1a or non-specific siRNA for negative control. After transfection, the cells were incubated under normoxic (N) or hypoxic (H) conditions for 24 h and subjected to Western blot analysis against visfatin or HIF-1a (left). The relative visfatin protein expression level was measured in at least three times independents followed by densitometry. The expression under normoxia transfected with control siRNA was set to 100% (right).
HIF-1 is necessary to activate the transcription of the visfatin gene in response to hypoxia. 3.4. Identification of functional HRE in human visfatin gene promoter To determine which elements within the region from 708 to +1 of visfatin promoter are actually involved in hypoxiamediated visfatin gene induction, specific mutations were individually introduced into the core sequence of the 8 putative HRE motifs within p(708)luc construct. As shown in Fig. 4A, although wild-type p(708)luc construct supported the hypoxic induction of luciferase activity by 6-fold, this activation was decreased when the HRE-2, -4 or -5 motifs were mutated. Moreover, the enhanced luciferase activity under hypoxic condition was abrogated markedly by 30% when the mutated HRE-4 (Mut4), or -5 (Mut5) constructs were transfected. Interestingly, HRE sites of mutated HRE-4 and -5 motifs shared complete conservation with the mouse and rat visfatin promoter (data not shown). We further demonstrated that double mutations
of both HRE 4 and 5 located between 358 and 346 sites completely abolished the hypoxic induction of p(708)luc construct (Fig. 3B), indicating that these two HRE motifs are essential and sufficient for the hypoxic visfatin response. Next, to obtain direct evidence for the interaction between HIF-1a and the visfatin promoter, we used ChIP assay to measure the HIF-1a recruitment to the visfatin promoter. As shown in Fig. 4C, HIF-1a bound effectively to the visfatin promoter under hypoxic treatment. In contrast, non-specific binding was unaffected by hypoxia, as demonstrated by results using control IgG as the primary antibody for immunoprecipitation. 3.5. Expression of visfatin and HIF-1a proteins in human breast carcinoma We next examined if there is a correlation between visfatin and HIF-1a protein expression in human neoplastic breast tissues. We carried out immunostaining of 22 human breast cancer tissues, applying anti-visfatin and anti-HIF-1a to sections of human breast carcinoma. As shown in Fig. 5, visfatin (21
S.-K. Bae et al. / FEBS Letters 580 (2006) 4105–4113
4109
Fig. 3. Visfatin promoter sequences mediate transcriptional responses to hypoxia and HIF-1. (A) The location and orientation of putative HRE sites within 5 0 -flanking region of the human visfatin gene defined by the core sequence 5 0 -RCGTG-3 0 were indicated at the top by the arrows. 1949-kb and 708-kb fragment of the 5 0 -flanking region of the human visfatin gene were subcloned into upstream of a luciferase reporter genes, respectively. MCF7 cells were transiently transfected with luciferase reporter vectors as indicated and subsequently incubated under hypoxia for 24 h. (B) MCF7 cells were transfected with p(708)luc, pBOS-hHIF-1a, pBOS-hARNT, or pEF-BOS as indicated and then incubated under hypoxia for 20 h. (C) Hepa1c1c7 (wild-type) and the HIF-1b/ARNT-deficient variant Hepa-1c4 cells were transfected with p(708)luc. Subsequently, the cells were incubated at 21% O2 (N: normoxia) or 1% O2 (H: hypoxia) for 20 h. (D) MCF7 cells were transfected with p(708)luc and HIF-1a or HIF-2a expression vector as indicated and then incubated for 20 h. All data (A–D) expressed as percentage ± S.E. from at least three separate occasions with triplicate. RLA means a relative luciferase activity.
of 22 cases; 95.4%) and HIF-1a (15 of 22 cases; 68.2%) were highly expressed in a malignant epithelium of nearly all of the cases of human breast neoplasia. Interestingly, the expres-
sion of HIF-1a was positively associated with visfatin, suggesting the correlation between visfatin and HIF-1a in actual breast tumors.
4110
S.-K. Bae et al. / FEBS Letters 580 (2006) 4105–4113
Fig. 4. Specificity of HIF-1 dependent activation and binding to the visfatin promoter sequences. (A) The regions of the mutated HRE-1–6 (Mut1–6) are indicated by triangle at the top. The mutations in two base pairs in each of the HIF-1 binding sites are indicated in bold. MCF7 cells were transfected with wild-type and mutated promotor constructs. After transfection, the cells were cultured at 1% oxygen for 20 h before the assessment of luciferase activity. (B) MCF7 cells were transfected with wild-type and mutated HRE-4, -5, -4/5 (double mutation in both HRE-4 and -5 sites), and then the cells were incubated at 1% O2 for additional 20 h. Each experiment was performed in triplicate on at least three separate occasions. (C) MCF7 cells were cultured under normoxia (N) or hypoxia (H) conditions for 16 h as described and subjected to ChIP assay. Anti-HIF-1 a polyclonal antibody (a-HIF-1a) or mouse IgG (a-IgG) was used to precipitate sonicated chromatin. Sonicated cell lysate (2 ll) was used as an input control. The gel pictures are representative of three independent experiments. IP, immunoprecipitation.
4. Discussion Visfatin is a secreted protein that is preferentially produced in visceral adipose tissue and known as a novel adipokine
exerting insulin-mimetic effects on various insulin-sensitive tissues [5]. The cDNA sequence of visfatin gene is identical to that of pre-B cell colony-enhancing factor (PBEF) which was originally cloned as a cytokine-like growth factor, expressed
S.-K. Bae et al. / FEBS Letters 580 (2006) 4105–4113
4111
Fig. 5. Immunostaining for visfatin and HIF-1a in human breast tumor. Sections of human breast cancer tissues were stained with antibodies that recognize the visfatin and HIF-1a (bottom, 200·). Sections were stained without primary anti-visfatin and anti-HIF-1a antibodies as controls (top, 200·). Samples were counterstained with hematoxylin to reveal morphology.
by lymphocyte, enhancing the effects of IL-7 and stem cell factor on early stage B cells [4]. In addition, visfatin has been reported to be expressed not only in the various normal organ tissues, but also in inflamed tissue under inflammatory stimuli and tumor tissues such as doxorubicin-responsive breast cancer and colorectal cancer [4,6–9,22]. Although it was reported using computer-based analysis of nucleotide sequences that the proximal promoter region of the human visfatin contains a number of putative binding motifs for transcription factors [6], the molecular basis and detailed regulatory mechanism of visfatin gene expression have not been elucidated yet. In this report, we showed for the first time that the human visfatin gene is transcriptionally activated in response to hypoxia through a mechanism that involves binding of the HIF-1 to functional HRE sites located within the proximal promoter region. In support of this conclusion, we found that: (i) low oxygen tension and CoCl2 increase visfatin mRNA and protein levels in breast cancer cell line, (ii) exogenous overexpression of HIF-1a markedly induces visfatin expression, whereas knockdown of HIF-1a using siRNA suppressed the hypoxic induction of visfatin gene, (iii) the 5 0 -flanking region of the human visfatin gene contains HREs that are necessary and sufficient for hypoxia-induced transcriptional activation of visfatin promoter-luciferase reporter gene, and (iv) the immunopositive region for visfatin is positively associated with HIF-1a in human breast tumor tissue. Obesity is a risk factor for breast cancer in post-menopausal women [23]. Circulating levels of adipokines are high in obesity and stimulate the growth of breast cancer cells; some of these are potentially involved in the regulation of angiogenesis which process is important for progression of breast cancer [24]. For example, leptin, one of typical adipokines, induces the proliferation of endothelial cells and elicits a marked enhancement of
angiogenesis [25]. In addition, the leptin is necessary to regulate adipose tissue expansion, which is one of the few sites of active angiogenesis in adult [26,27]. Based on these reports, we raised the possibility that visfatin may also regulate angiogenesis. As expected, we found that visfatin protein accelerated the formation of new blood vessel in chick chorioallantoic membranes (unpublished data), suggesting that visfatin-mediated angiogenesis may be involved not only in enlargement of tumor mass but also in excessive growth of adipose tissue for the settlement of obesity. We are now investigating about molecular mechanism(s) for the angiogenic activities of visfatin and elucidating the relationship between hypoxia and visfatin gene induction in normal cell types including adipose cells and normal breast cell lines. Recently, it is reported that visfatin promotes the efficient acquisition of a stable and mature smooth muscle cell phenotype that is essential for developing and remodeling of blood vessel [28]. Therefore, visfatin may act as a regulator for the maintenance of vascular homeostasis in vascular endothelial and smooth muscle cells. Our observations add the human visfatin gene to a list of genes regulated by hypoxia via the HIF-1-dependent pathway. Hypoxia, a common consequence of solid tumor growth in breast cancer and other cancers, serves to propagate a cascade of molecular pathways which include angiogenesis, glycolysis, and alterations in microenvironmental pH [29]. Consistent with a stimulatory effect of a local hypoxic environment on the expression of target genes in tumors, we showed marked expression of visfatin gene in both hypoxia-treated breast cancer cells and human breast tumor tissues. Besides the role as a tumor-promoting factor, hypoxia is one of key factors in pathogenesis of acute lung injury [30]. Recently, it was reported that pre-B-cell colony-enhancing factor (also known as visfatin) is a potential novel biomaker in acute lung injury [31].
4112
Taking into account these reports, we consider that visfatin may participate in response to hypoxia in several pathological situations. We showed that HIF-1 and HIF-2 are not equally effective on the p(708)luc visfatin promoter, as HIF-1a protein was able to induce the visfatin promoter at higher levels than HIF-2a (Fig. 3D). The activation control reporter plasmid pSV40pro-EpoHRE-Luc, containing the four HRE core sequence of erythropoietin gene, was also tested in reporter assay. However, either exogenous HIF-2a or HIF-1a induced transcription of the control HRE reporter plasmid at similar levels (data not shown), indicating the specificity of HIF-1 for the activation of visfatin promoter. In addition, several reports showed that basal HIF-2a mRNA and protein level under hypoxia were low or absent in the various breast cancer cell lines, whereas these cells dramatically expressed the high-level HIF-1a protein in response to hypoxia [32,33]. Recently, it has been reported that the expression of hypoxia-induced genes was critically dependent on HIF-1a but not HIF-2a in breast cancer cells [34]. Based on these observations, we strongly suggest that HIF-1a protein, rather than HIF-2a, may be predominantly expressed in human breast cancer cells and preferentially bind to HREs of the visfatin promoter, leading to the activation of the human visfatin gene. Hypoxia is not the only condition that stabilizes HIF-1a and activates HIF-1 transcriptional activity. Several hormones, growth factors and viral proteins, including insulin and insulin-like growth factor [35,36], angiotensin II, thrombin and platelet-derived growth factor [37], endothelin-1 [38], and Hepatitis B virus X protein [39], have been shown to increase the level of HIF-1a in various cell types. In addition, inflammatory cytokines, such as interleukin 1b and tumor necrosis factor a, also induce HIF-1 activity in normoxic cells [40]. Considering that multiple inflammatory stimuli (including lipopolysaccharide, interleukins, and tumor necrosis factor a) significantly increased visfatin expression [6], it would be of interest to investigate whether these stimuli could regulate visfatin gene expression in some types of cells via hypoxia-independent HIF-1 pathways. In Fig. 5, the expressional correlation between visfatin and HIF-1a genes was observed in neoplastic human breast tissues, supporting our result that visfatin is highly expressed under the regulation of HIF-1a protein. However, since expressional ratio of visfatin (21 of 22 cases; 95.4%) was higher than that of HIF-1a (15 of 22 cases; 68.2%), we cannot exclude possibility that HIF-1-independent pathway may be also involved in the regulation of visfatin gene expression. In conclusion, the data presented here are consistent with the visfatin gene being a hypoxia-inducible gene. Moreover, we demonstrated that hypoxia mediates the upregulation of visfatin gene via HIF-1a and HIF-1-dependent transcriptional activity, as described for several other genes regulated by low oxygen availability. Our data may provide a molecular explanation for patho-physiological regulation of visfatin in breast tumorigenesis and a novel approach as a target for the development of new therapeutic strategies. Acknowledgements: This work was supported by the research grant from a Vascular System Research Center grant from KOSEF, Korea Research Foundation Grant funded by Korea Government (MOEHRD, basic Research Promotion Fund) (KRF-2005-204C00051) (to M.-K. Bae) and MRC program of MOST/KOSEF (R13-2005-009) (to S.-K. Bae).
S.-K. Bae et al. / FEBS Letters 580 (2006) 4105–4113
References [1] Koerner, A., Kratzsch, J. and Kiess, W. (2005) Adipocytokines: leptin – the classical, resistin – the controversical, adiponectin – the promising, and more to come. Best Pract. Res. Clin. Endocrinol. Metab. 4, 525–546. [2] Trayhurn, P. and Wood, I.S. (2004) Adipokines: inflammation and the pleiotropic role of white adipose tissue. Br. J. Nutr. 3, 347–355. [3] Rose, D.P., Komninou, D. and Stephenson, G.D. (2004) Obesity, adipocytokines, and insulin resistance in breast cancer. Obes. Rev. 3, 153–165. [4] Samal, B., Sun, Y., Stearns, G., Xie, C., Suggs, S. and McNiece, I. (1994) Cloning and characterization of the cDNA encoding a novel human pre-B-cell colony-enhancing factor. Mol. Cell. Biol. 2, 1431–1437. [5] Fukuhara, A., Matsuda, M., Nishizawa, M., Segawa, K., Tanaka, M., Kishimoto, K., Matsuki, Y., Murakami, M., Ichisaka, T. and Murakami, H., et al. (2005) Visfatin: a protein secreted by visceral fat that mimics the effects of insulin. Science 5708, 426–430. [6] Ognjanovic, S., Bao, S., Yamamoto, S.Y., Garibay-Tupas, J., Samal, B. and Bryant-Greenwood, G.D. (2001) Genomic organization of the gene coding for human pre-B-cell colony enhancing factor and expression in human fetal membranes. J. Mol. Endocrinol. 2, 107–117. [7] Ye, S.Q., Zhang, L.Q., Adyshev, D., Usatyuk, P.V., Garcia, A.N., Lavoie, T.L., Verin, A.D., Natarajan, V. and Garcia, J.G. (2005) Pre-B-cell-colony-enhancing factor is critically involved in thrombin-induced lung endothelial cell barrier dysregulation. Microvasc. Res. 3, 142–151. [8] Hufton, S.E., Moerkerk, P.T., Brandwijk, R., de Bruine, A.P., Arends, J.W. and Hoogenboom, H.R. (1999) A profile of differentially expressed genes in primary colorectal cancer using suppression subtractive hybridization. FEBS Lett. 1–2, 77–82. [9] Folgueira, M.A., Carraro, D.M., Brentani, H., Patrao, D.F., Barbosa, E.M., Netto, M.M., Caldeira, J.R., Katayama, M.L., Soares, F.A. and Oliveira, C.T., et al. (2005) Gene expression profile associated with response to doxorubicin-based therapy in breast cancer. Clin. Cancer Res. 20, 7434–7443. [10] Bergers, G. and Benjamin, L.E. (2003) Tumorigenesis and the angiogenic switch. Nat. Rev. Cancer 6, 401–410. [11] Vaupel, P. (2004) The role of hypoxia-induced factors in tumor progression. Oncologist, 10–17. [12] Kallio, P.J., Wilson, W.J., O’Brien, S., Makino, Y. and Poellinger, L. (1999) Regulation of the hypoxia-inducible transcription factor 1alpha by the ubiquitin-proteasome pathway. J. Biol. Chem. 10, 6519–6525. [13] Grosfeld, A., Andre, J., Hauguel-De Mouzon, S., Berra, E., Pouyssegur, J. and Guerre-Millo, M. (2002) Hypoxia-inducible factor 1 transactivates the human leptin gene promoter. J. Biol. Chem. 45, 42953–42957. [14] Ambrosini, G., Nath, A.K., Sierra-Honigmann, M.R. and FloresRiveros, J. (2002) Transcriptional activation of the human leptin gene in response to hypoxia. Involvement of hypoxia-inducible factor 1. J. Biol. Chem. 37, 34601–34609. [15] Jeong, H.J., Chung, H.S., Lee, B.R., Kim, S.J., Yoo, S.J., Hong, S.H. and Kim, H.M. (2003) Expression of proinflammatory cytokines via HIF-1alpha and NF-kappaB activation on desferrioxamine-stimulated HMC-1 cells. Biochem. Biophys. Res. Commun. 4, 805–811. [16] Forsythe, J.A., Jiang, B.H., Iyer, N.V., Agani, F., Leung, S.W., Koos, R.D. and Semenza, G.L. (1996) Activation of vascular endothelial growth factor gene transcription by hypoxia-inducible factor 1. Mol. Cell. Biol. 9, 4604–4613. [17] Jin, K., Mao, X.O., Sun, Y., Xie, L., Jin, L., Nishi, E., Klagsbrun, M. and Greenberg, D.A. (2002) Heparin-binding epidermal growth factor-like growth factor: hypoxia-inducible expression in vitro and stimulation of neurogenesis in vitro and in vivo. J. Neurosci. 13, 5365–5373. [18] Warnecke, C., Griethe, W., Weidemann, A., Jurgensen, J.S., Willam, C., Bachmann, S., Ivashchenko, Y., Wagner, I., Frei, U., Wiesener, M. and Eckardt, K.U. (2003) Activation of the hypoxia-inducible factor-pathway and stimulation of angiogenesis by application of prolyl hydroxylase inhibitors. FASEB J. 9, 1186–1188.
S.-K. Bae et al. / FEBS Letters 580 (2006) 4105–4113 [19] Gassmann, M., Kvietikova, I., Rolfs, A. and Wenger, R.H. (1997) Oxygen- and dioxin-regulated gene expression in mouse hepatoma cells. Kidney Int. 2, 567–574. [20] Tian, H., McKnight, S.L. and Russell, D.W. (1997) Endothelial PAS domain protein 1 (EPAS1), a transcription factor selectively expressed in endothelial cells. Genes Dev. 11, 72–82. [21] Ema, M., Taya, S., Yokotani, N., Sogawa, K., Matsuda, Y. and Fujii-Kuriyama, Y. (1997) A novel bHLH-PAS factor with close sequence similarity to hypoxia-inducible factor 1alpha regulates the VEGF expression and is potentially involved in lung and vascular development. Proc. Natl. Acad. Sci. USA 94, 4273–4278. [22] Jia, S.H., Li, Y., Parodo, J., Kapus, A., Fan, L., Rotstein, O.D. and Marshall, J.C. (2004) Pre-B cell colony-enhancing factor inhibits neutrophil apoptosis in experimental inflammation and clinical sepsis. J. Clin. Invest. 9, 1318–1327. [23] Stephenson, G.D. and Rose, D.P. (2003) Breast cancer and obesity: an update. Nutr. Cancer 1, 1–16. [24] Schneider, B.P. and Miller, K.D. (2005) Angiogenesis of breast cancer. J. Clin. Oncol. 8, 1782–1790. [25] Bouloumie, A., Drexler, H.C., Lafontan, M. and Busse, R. (1998) Leptin, the product of Ob gene, promotes angiogenesis. Circ. Res. 10, 1059–1066. [26] Hausman, G.J. and Richardson, R.L. (2004) Adipose tissue angiogenesis. J. Anim. Sci. 3, 925–934. [27] Caro, J.F., Sinha, M.K., Kolaczynski, J.W., Zhang, P.L. and Considine, R.V. (1996) Leptin: the tale of an obesity gene. Diabetes 11, 1455–1462. [28] van der Veer, E., Nong, Z., O’Neil, C., Urquhart, B., Freeman, D. and Pickering, J.G. (2005) Pre-B-cell colony-enhancing factor regulates NAD+-dependent protein deacetylase activity and promotes vascular smooth muscle cell maturation. Circ. Res. 1, 25–34. [29] Goonewardene, T.I., Sowter, H.M. and Harris, A.L. (2002) Hypoxia-induced pathways in breast cancer. Microsc. Res. Tech. 1, 41–48. [30] Agorreta, J., Zulueta, J.J., Montuenga, L.M. and Garayoa, M. (2005) Adrenomedullin expression in a rat model of acute lung injury induced by hypoxia and LPS. Am. J. Physiol. Lung Cell. Mol. Physiol. 3, L536–L545. [31] Ye, S.Q., Simon, B.A., Maloney, J.P., Zambelli-Weiner, A., Gao, L., Grant, A., Easley, R.B., McVerry, B.J., Tuder, R.M. and
4113
[32]
[33] [34]
[35]
[36]
[37]
[38]
[39]
[40]
Standiford, T., et al. (2005) Pre-B-cell colony-enhancing factor as a potential novel biomarker in acute lung injury. Am. J. Respir. Crit. Care Med. 4, 361–370. Blancher, C., Moore, J.W., Talks, K.L., Houlbrook, S. and Harris, A.L. (2000) Relationship of hypoxia-inducible factor (HIF)-1alpha and HIF-2alpha expression to vascular endothelial growth factor induction and hypoxia survival in human breast cancer cell lines. Cancer Res. 60, 7106–7113. Ameri, K., Burke, B., Lewis, C.E. and Harris, A.L. (2002) Regulation of a rat VL30 element in human breast cancer cells in hypoxia and anoxia: role of HIF-1. Br. J. Cancer 87, 1173–1181. Sowter, H.M., Raval, R.R., Moore, J.W., Ratcliffe, P.J. and Harris, A.L. (2003) Predominant role of hypoxia-inducible transcription factor (Hif)-1alpha versus Hif-2alpha in regulation of the transcriptional response to hypoxia. Cancer Res. 63, 6130– 6134. Zelzer, E., Levy, Y., Kahana, C., Shilo, B.Z., Rubinstein, M. and Cohen, B. (1998) Insulin induces transcription of target genes through the hypoxia-inducible factor HIF-1alpha/ARNT. EMBO J. 17, 5085–5094. Feldser, D., Agani, F., Iyer, N.V., Pak, B., Ferreira, G. and Semenza, G.L. (1999) Reciprocal positive regulation of hypoxiainducible factor 1alpha and insulin-like growth factor 2. Cancer Res. 16, 3915–3918. Richard, D.E., Berra, E. and Pouyssegur, J. (2000) Nonhypoxic pathway mediates the induction of hypoxia-inducible factor 1alpha in vascular smooth muscle cells. J. Biol. Chem. 35, 26765–26771. Spinella, F., Rosano, L., Di Castro, V., Natali, P.G. and Bagnato, A. (2002) Endothelin-1 induces vascular endothelial growth factor by increasing hypoxia-inducible factor-1alpha in ovarian carcinoma cells. J. Biol. Chem. 31, 27850–27855. Moon, E.J., Jeong, C.H., Jeong, J.W., Kim, K.R., Yu, D.Y., Murakami, S., Kim, C.W. and Kim, K.W. (2004) Hepatitis B virus X protein induces angiogenesis by stabilizing hypoxiainducible factor-1alpha. FASEB J. 2, 382–384. Hellwig-Burgel, T., Rutkowski, K., Metzen, E., Fandrey, J. and Jelkmann, W. (1999) Interleukin-1beta and tumor necrosis factoralpha stimulate DNA binding of hypoxia-inducible factor-1. Blood 5, 1561–1567.
Hypoxic induction of human visfatin gene is directly mediated by hypoxia-inducible factor-1 Soo-Kyung Baeb,e,1, Su-Ryun Kima,d,1, Jong Gab Kimb,e, Jee Yeon Kimb, Tae Hyeon Kooc, Hye-Ock Janga, Il Yuna, Mi-Ae Yood, Moon-Kyoung Baea,* a
College of Dentistry and Research Institute for Oral Biotechnology, Pusan National University, Pusan 602-739, South Korea b College of Medicine, Pusan National University, Pusan 602-739, South Korea c Department of Biology, Korea Science Academy, Pusan 614-822, South Korea d Department of Molecular Biology, Pusan National University, Pusan 609-735, South Korea e Medical Research Center for Ischemic Tissue Regeneration, Pusan National University, Pusan 602-739, South Korea Received 4 May 2006; revised 19 June 2006; accepted 23 June 2006 Available online 30 June 2006 Edited by Robert Barouki
Abstract Visfatin has been originally identified as a growth factor for early stage B cells and recently known as an adipokine. Here, we report that hypoxia induces the visfatin mRNA and protein levels in MCF7 breast cancer cells. We also demonstrate that induction of visfatin gene is regulated by hypoxia-inducible factor-1a (HIF-1a). Moreover, 5 0 -flanking promoter region of human visfatin gene contains two functional HIF responsive elements (HREs), activating the expression of visfatin. Mutation of these HREs in the visfatin promoter abrogates activation of a luciferase reporter gene driven by visfatin promoter under hypoxia. Taken together, our results demonstrate that visfatin is a new hypoxia-inducible gene of which expression is stimulated through the interaction of HIF-1 with HRE sites in its promoter region. 2006 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. Keywords: Visfatin; Pre-B-cell colony-enhancing factor; Hypoxia; Hypoxia-inducible factor-1a; Breast cancer cells
1. Introduction Adipokines are biologically active polypeptides growth factors and cytokines that appear to be preferentially produced by adipose tissue [1,2]. They exert their effects on the normal and neoplastic breast tissues by endocrine and paracrine mechanisms; in most cases, they are also secreted by cancerous epithelial cells, providing for autocrine regulation [3]. They can stimulate cancer epithelial cell proliferation, invasion, and angiogenesis which is essential for breast cancer development and progression [3]. Visfatin (pre-B-cell colony-enhancing factor) is firstly isolated from peripheral blood lymphocyte and recently characterized as a novel adipokine [4,5]. In addition, visfatin is
*
Corresponding author. Fax. +82 51 245 8388. E-mail address: [email protected] (M.-K. Bae). 1
Denotes co-first authors.
expressed on the various normal, inflamed and tumor tissues [5–9]. More recently, acute mechanical and multiple inflammatory stimuli increased visfatin expression [6,7], but precise mechanism(s) underlying this visfatin induction remains poorly resolved. Furthermore, molecular mechanism of visfatin gene regulation in tumor progression has not been elucidated yet. Hypoxia detected in the central regions of solid tumors can be a leading cause of angiogenesis, a fundamental determinant of malignant tumor progression [10,11]. A key factor in this process is hypoxia-inducible factor (HIF), which regulates transcription of hypoxia-activated genes and consists of HIF1a and HIF-1b heterodimer. The a subunit of HIF-1 is rapidly degraded under normoxic condition and stabilized under hypoxia, while HIF-1b is expressed constitutively [12]. Since the production of many adipokines including leptin, IL-6, VEGF, and heparin-binding epidermal growth factor-like growth factor are tightly regulated by hypoxia [13–17], we hypothesized that hypoxic microenvironment within tumor mass activates the expression of visfatin. In this present study, we investigated whether the expression of visfatin is upregulated in response to hypoxia and further examined whether the HIF-1 is involved in hypoxia-induced activation of human visfatin gene in breast cancer MCF7 cell line. Herein, our data provide evidence that hypoxic induction of visfatin gene is mediated by transactivation of HIF-1 through HIF-1 consensus binding site (HRE) in the proximal promoter of visfatin gene.
2. Materials and methods 2.1. Reagents and antibodies Cobalt chloride and L -mimosine were from Sigma. Mouse monoclonal anti-HIF-1a and a-tubulin antibodies were purchased from BD transduction laboratories and Biogenex, respectively. Rabbit and mouse monoclonal anti-visfatin antibodies were from AdipoGen. Alexa 594-conjugated goat anti-mouse IgG was purchased from Molecular Probes. 2.2. Cell culture MCF7 cells, Hepa-1c4 and Hepa-1c1c7 cells were maintained in DMEM supplemented with 10% FBS (Invitrogen) and 1% antibiotics. For hypoxic condition, cells were incubated at 5% CO2 level with 1% O2 balanced with N2 in hypoxic chamber (Forma Scientific).
0014-5793/$32.00 2006 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.febslet.2006.06.052
4106 2.3. Plasmids and constructs A genomic DNA fragment of the human visfatin gene [6] containing 2 kb of 5 0 -flanking region was prepared by PCR amplification of human genomic DNA. A 1949 bp PCR product was obtained and subcloned into a pCR2.1/TA vector (Invitrogen). From this construct, a full-length (p(1949)luc) and a deletion mutant of visfatin gene (p(708)luc) were amplified by PCR and subcloned into pGL3 luciferase reporter vector (Promega), respectively. All constructs were confirmed by automatic DNA sequencing analysis. 2.4. Real-time PCR Real-time PCR quantification was performed using a SYBR Green approach (Light Cycler; Roche Applied Science, Penzberg, Germany). Cycling parameters consisted of 1 cycle of 95 C for 10 min, followed by amplification for 30 cycles of 95 C for 10 s, 57 C for 5 s, and 72 C for 7 s. Subsequently, a melting curve program was applied with continuous fluorescence measurement. The entire cycling process including data analysis took less than 1 h and was monitored using the LightCycler software program (version 4.0). 2.5. Immunocytochemistry Cells cultured on a coverglass were fixed in 4% paraformaldehyde for 10 min, blocked with 0.5% Triton X-100/PBS for 5 min, and then labeled with appropriate primary antibodies and Alexa 594-conjugated secondary antibody. Coverslips were mounted in Vectastain containing DAPI (Vector Laboratories). Cells were analyzed using fluorescent microscopy (Nikon). 2.6. RNA interference experiment MCF7 cells were transfected with siRNA specific for HIF-1a (antisense: 5 0 -UCAAACACACUGUGUCCAG-3 0 ; sense: 5 0 -CUGGACACGUGUGUUUGA-3 0 ) at a concentration of 100 nM using the Lipofectamine 2000 (Invitrogen). After transfection, the cells were exposed to a normoxic or hypoxic atmosphere for the indicated times. 2.7. Transient transfection and reporter gene analysis Cells were plated on 24-well plates and transfected with the luciferase constructs and pCMV-b-gal using Lipofectamine 2000 (Invitrogen). On the following day, the cells were incubated with hypoxic condition for 20 h and then solubilized. Cell lysates were analyzed for b-galactosidase enzyme assay and for luciferase activity using assay kit (Promega) and luminometer (Turner Designs). Each extract was assayed at least three times, and relative luciferase activity was calculated as RLU/b-galactosidase. 2.8. Site-directed mutagenesis Six putative HIF-1 binding sites (HRE: 5 0 -RCGTG-3 0 ) within p(708)luc were targeted for mutagenesis. The mutations were made with a QuickChange site-directed mutagenesis kit (Stratagene) and checked by automatic DNA sequencing. 2.9. Chromatin immunoprecipitation assay Chromatin immunoprecipitation (ChIP) analysis was performed with the ChIP kit (Upstate Biotechnology) according to the manufacturer’s instructions with minor modifications and anti-HIF-1a monoclonal antibody (Novus). Chromatins were immunoprecipitated from 1 · 106 cells incubated under normoxia and hypoxia for 16 h, respectively. Sequences of promoter-specific primers included the visfatin promoter region 383 to 258 as follows: sense, 5 0 -AGAGCTGGCGTCTGGGAG-3 0 ; and anti-sense, 5 0 -GCCTTCACCCCGTCACCC-3 0 . 2.10. Immunohistochemistry Human breast cancer tissue specimens were prepared and thin sections (4 lm) from selected areas were used. In brief, after deparaffinization and blocking of endogenous peroxidase activity, antigen retrieval was routinely performed using PBS containing 0.04% alkaline protease solution (Promega). The primary monoclonal antibodies used in this study were mouse anti-pan-visfatin and rabbit anti-HIF-1a. Tissues were stained according to the standard avidin–biotin method using ABC reagent (Vectastain Elite kit, Vector Labs.) and counterstained with Harris hematoxylin.
S.-K. Bae et al. / FEBS Letters 580 (2006) 4105–4113
3. Results 3.1. Hypoxia increases human visfatin mRNA and protein levels To examine the effects of hypoxia on visfatin gene expression, MCF7 cells were incubated under hypoxic condition for periods of 16 h and 24 h. As shown in Fig. 1A, hypoxic treatment produced significant increases in visfatin mRNA levels. Using real-time RT-PCR, we also quantified the enhanced expression level of visfatin mRNA under hypoxia (Fig. 1B). Next, to analyze whether hypoxic condition could affect visfatin protein levels, Western immunoblot analysis was performed. As shown in Fig. 1C, 1% O2 and cobalt chloride, hypoxic mimic agent, increased visfatin protein levels in a time-dependent manner. The increased level of visfatin protein by hypoxia was confirmed by immunocytochemistry. As shown in Fig. 1D, a stronger staining of visfatin protein was observed in hypoxic MCF7 cells than in normoxic cells. These results indicated that hypoxia significantly upregulates visfatin mRNA and protein levels. 3.2. Hypoxic induction of visfatin mRNA and protein is dependent on HIF-1 HIF-1 is a key transcription factor of the transcriptional responses to hypoxia. To determine the crucial role of HIF-1a in hypoxia-mediated upregulation of visfatine gene, we transfected HIF-1a expression vector into normoxic MCF7 cells. As shown in Fig. 2A, the overexpressed HIF-1a increased the visfatin protein level even under normoxic condition, which shows similar increase in the expression of visfatin under hypoxia. The stability of HIF-1a protein is known to be markedly downregulated by prolyl hydroxylase domain protein (PHD) that hydroxylates HIF-1a under normoxia, followed by its polyubiquitination and proteasomal degradation. We found that treatment of normoxic MCF7 cells with L -mimosine, a well-recognized PHD inhibitor [18], induced visfatin mRNA levels (Fig. 2B). To confirm whether endogenous visfatin gene is one of direct HIF-1 target genes, we also used specific siRNA against HIF-1a. Transfection of MCF7 cells with siRNA against HIF-1a markedly reversed hypoxia-induced visfatin expression (Fig. 2C). Thus, these data demonstrated that HIF-1a is directly involved in the induction of the visfatin gene expression. 3.3. Transcriptional activation of the visfatin gene promoter by hypoxia We next focused on detailed analysis of human visfatin promoter to find out functional HIF-1 binding sites. Therefore, we searched putative HIF-1 binding sequences in 5 0 -flanking region of the human visfatin gene [6]. An extensive analysis of 5 0 visfatin promoter sequence (between positions 1949 and +1), using the computer database TRANSFAC, revealed the presence of several putative HIF-1 consensus-binding sites. The relative position and orientation of these elements are indicated at the top of Fig. 3. Based on this information, we cloned two fragments containing up to 1949 bp and up to 708 bp of the 5 0 -flanking region of the human visfatin gene, and then constructed luciferase reporter vectors, named as p(1949)luc and p(708)luc, respectively. These constructs were then introduced into MCF7 cells and incubated with hypoxia. As shown in Fig. 3A, reporter activities of both constructs under hypoxia were significantly increased compared to that of pGL3 alone. Because the transcriptional activity of p(708)luc
S.-K. Bae et al. / FEBS Letters 580 (2006) 4105–4113
4107
Fig. 1. Effect of hypoxia on the expression of visfatin gene. (A) MCF7 cells were incubated under 21% O2 or 1% O2 for the indicated times. RT-PCR analysis was performed using specific primers for human visfatin. b-actin served as an internal control. (B) MCF7 cells were incubated under normoxic or hypoxic conditions for 16 h or 24 h. Total RNAs were isolated and then analyzed by real-time PCR using visfatin primers. The expression level under normoxia (0 h hypoxia) was set to 100%, and the values are normalized to actin mRNA levels. (C) MCF7 cells were incubated for the indicated times under 1% O2 or treated with 100 lM cobalt chloride. Visfatin protein levels were examined by Western blotting using antivisfatin. a-tubulin served as loading control. (D) Photographs showed immunocytochemically staining cells with anti-pan visfatin antibody. The signals were detected using Alexa 594-conjugated secondary antibody (red).
under hypoxia is very similar to that of p(1949)luc, we expected that potential functional HREs exist on fragment from 708 to +1, and then we used this fragment (from 708 to +1) in the following experiments. To test whether exogenous HIF1a overexpression can cause the same stimulatory effects on visfatin promoter, HIF-1a expression vector was added with p(708)luc. As expected, transfection of HIF-1a expression vector stimulated reporter activity even in normoxic condition: this was further enhanced by the hypoxia (Fig. 3B). To determine whether the functional HIF-1 complex is required for transcriptional activation of visfatin by hypoxia, we took advantage of the mouse hepatoma Hepa-1c4 cells. This cell line is derived from parental Hepa-1c1c7 cells and deficient in HIF1b subunit, which fails to form functional HIF-1 heterodimer [19]. The p(708)luc reporter construct was transiently transfec-
ted into Hepa-1c4 cells and Hepa-1c1c7 cells under normoxic or hypoxic conditions. The luciferase activities indicated a substantial induction in hypoxic Hepa-1c1c7 cells, whereas no significant increase was observed in Hepa-1c4 cells in response to hypoxia (Fig. 3C). HIF-1a and HIF-2a are two highly related pHLH/PAS homologous transcription factors that undergo increased function at low oxygen levels [20]. Transcriptionally active heterodimer composed of HIF-2a and ARNT, like HIF-1a and ARNT, binds to HRE and activates hypoxia-induced genes [21]. To check whether HIF-2a might be involved in hypoxia-induced visfatin promoter activation, we performed a luciferase assay using an p(708)luc reporter and HIF-2a. The p(708)luc activity in the presence of HIF-1a was significantly increased, whereas no increase was observed in the presence of HIF-2a expression vector (Fig. 3D). Thus, functional
4108
S.-K. Bae et al. / FEBS Letters 580 (2006) 4105–4113
Fig. 2. The role of HIF-1a in the regulation of visfatin gene expression. (A) MCF7 cells were transfected with HIF-1a expression vector and left untreated or exposed to 1% O2 for 24 h. Anti-visfatin or HIF-1a Western blotting from protein extracts was performed. a-tubulin served as a loading control (left panel). The right panel showed densitometric analysis assessing relative visfatin protein expression level in at least three times independent experiments. The density of the control bands (21% O2 for 24 h) was set to 100%. (B) Cell lysates were prepared from MCF7 cells treated with 400 or 800 lM L -mimosine for 16 h. RT-PCR analysis was performed using specific primers for visfatin or b-actin. (C) MCF7 cells were transfected with siRNAs specific for HIF-1a or non-specific siRNA for negative control. After transfection, the cells were incubated under normoxic (N) or hypoxic (H) conditions for 24 h and subjected to Western blot analysis against visfatin or HIF-1a (left). The relative visfatin protein expression level was measured in at least three times independents followed by densitometry. The expression under normoxia transfected with control siRNA was set to 100% (right).
HIF-1 is necessary to activate the transcription of the visfatin gene in response to hypoxia. 3.4. Identification of functional HRE in human visfatin gene promoter To determine which elements within the region from 708 to +1 of visfatin promoter are actually involved in hypoxiamediated visfatin gene induction, specific mutations were individually introduced into the core sequence of the 8 putative HRE motifs within p(708)luc construct. As shown in Fig. 4A, although wild-type p(708)luc construct supported the hypoxic induction of luciferase activity by 6-fold, this activation was decreased when the HRE-2, -4 or -5 motifs were mutated. Moreover, the enhanced luciferase activity under hypoxic condition was abrogated markedly by 30% when the mutated HRE-4 (Mut4), or -5 (Mut5) constructs were transfected. Interestingly, HRE sites of mutated HRE-4 and -5 motifs shared complete conservation with the mouse and rat visfatin promoter (data not shown). We further demonstrated that double mutations
of both HRE 4 and 5 located between 358 and 346 sites completely abolished the hypoxic induction of p(708)luc construct (Fig. 3B), indicating that these two HRE motifs are essential and sufficient for the hypoxic visfatin response. Next, to obtain direct evidence for the interaction between HIF-1a and the visfatin promoter, we used ChIP assay to measure the HIF-1a recruitment to the visfatin promoter. As shown in Fig. 4C, HIF-1a bound effectively to the visfatin promoter under hypoxic treatment. In contrast, non-specific binding was unaffected by hypoxia, as demonstrated by results using control IgG as the primary antibody for immunoprecipitation. 3.5. Expression of visfatin and HIF-1a proteins in human breast carcinoma We next examined if there is a correlation between visfatin and HIF-1a protein expression in human neoplastic breast tissues. We carried out immunostaining of 22 human breast cancer tissues, applying anti-visfatin and anti-HIF-1a to sections of human breast carcinoma. As shown in Fig. 5, visfatin (21
S.-K. Bae et al. / FEBS Letters 580 (2006) 4105–4113
4109
Fig. 3. Visfatin promoter sequences mediate transcriptional responses to hypoxia and HIF-1. (A) The location and orientation of putative HRE sites within 5 0 -flanking region of the human visfatin gene defined by the core sequence 5 0 -RCGTG-3 0 were indicated at the top by the arrows. 1949-kb and 708-kb fragment of the 5 0 -flanking region of the human visfatin gene were subcloned into upstream of a luciferase reporter genes, respectively. MCF7 cells were transiently transfected with luciferase reporter vectors as indicated and subsequently incubated under hypoxia for 24 h. (B) MCF7 cells were transfected with p(708)luc, pBOS-hHIF-1a, pBOS-hARNT, or pEF-BOS as indicated and then incubated under hypoxia for 20 h. (C) Hepa1c1c7 (wild-type) and the HIF-1b/ARNT-deficient variant Hepa-1c4 cells were transfected with p(708)luc. Subsequently, the cells were incubated at 21% O2 (N: normoxia) or 1% O2 (H: hypoxia) for 20 h. (D) MCF7 cells were transfected with p(708)luc and HIF-1a or HIF-2a expression vector as indicated and then incubated for 20 h. All data (A–D) expressed as percentage ± S.E. from at least three separate occasions with triplicate. RLA means a relative luciferase activity.
of 22 cases; 95.4%) and HIF-1a (15 of 22 cases; 68.2%) were highly expressed in a malignant epithelium of nearly all of the cases of human breast neoplasia. Interestingly, the expres-
sion of HIF-1a was positively associated with visfatin, suggesting the correlation between visfatin and HIF-1a in actual breast tumors.
4110
S.-K. Bae et al. / FEBS Letters 580 (2006) 4105–4113
Fig. 4. Specificity of HIF-1 dependent activation and binding to the visfatin promoter sequences. (A) The regions of the mutated HRE-1–6 (Mut1–6) are indicated by triangle at the top. The mutations in two base pairs in each of the HIF-1 binding sites are indicated in bold. MCF7 cells were transfected with wild-type and mutated promotor constructs. After transfection, the cells were cultured at 1% oxygen for 20 h before the assessment of luciferase activity. (B) MCF7 cells were transfected with wild-type and mutated HRE-4, -5, -4/5 (double mutation in both HRE-4 and -5 sites), and then the cells were incubated at 1% O2 for additional 20 h. Each experiment was performed in triplicate on at least three separate occasions. (C) MCF7 cells were cultured under normoxia (N) or hypoxia (H) conditions for 16 h as described and subjected to ChIP assay. Anti-HIF-1 a polyclonal antibody (a-HIF-1a) or mouse IgG (a-IgG) was used to precipitate sonicated chromatin. Sonicated cell lysate (2 ll) was used as an input control. The gel pictures are representative of three independent experiments. IP, immunoprecipitation.
4. Discussion Visfatin is a secreted protein that is preferentially produced in visceral adipose tissue and known as a novel adipokine
exerting insulin-mimetic effects on various insulin-sensitive tissues [5]. The cDNA sequence of visfatin gene is identical to that of pre-B cell colony-enhancing factor (PBEF) which was originally cloned as a cytokine-like growth factor, expressed
S.-K. Bae et al. / FEBS Letters 580 (2006) 4105–4113
4111
Fig. 5. Immunostaining for visfatin and HIF-1a in human breast tumor. Sections of human breast cancer tissues were stained with antibodies that recognize the visfatin and HIF-1a (bottom, 200·). Sections were stained without primary anti-visfatin and anti-HIF-1a antibodies as controls (top, 200·). Samples were counterstained with hematoxylin to reveal morphology.
by lymphocyte, enhancing the effects of IL-7 and stem cell factor on early stage B cells [4]. In addition, visfatin has been reported to be expressed not only in the various normal organ tissues, but also in inflamed tissue under inflammatory stimuli and tumor tissues such as doxorubicin-responsive breast cancer and colorectal cancer [4,6–9,22]. Although it was reported using computer-based analysis of nucleotide sequences that the proximal promoter region of the human visfatin contains a number of putative binding motifs for transcription factors [6], the molecular basis and detailed regulatory mechanism of visfatin gene expression have not been elucidated yet. In this report, we showed for the first time that the human visfatin gene is transcriptionally activated in response to hypoxia through a mechanism that involves binding of the HIF-1 to functional HRE sites located within the proximal promoter region. In support of this conclusion, we found that: (i) low oxygen tension and CoCl2 increase visfatin mRNA and protein levels in breast cancer cell line, (ii) exogenous overexpression of HIF-1a markedly induces visfatin expression, whereas knockdown of HIF-1a using siRNA suppressed the hypoxic induction of visfatin gene, (iii) the 5 0 -flanking region of the human visfatin gene contains HREs that are necessary and sufficient for hypoxia-induced transcriptional activation of visfatin promoter-luciferase reporter gene, and (iv) the immunopositive region for visfatin is positively associated with HIF-1a in human breast tumor tissue. Obesity is a risk factor for breast cancer in post-menopausal women [23]. Circulating levels of adipokines are high in obesity and stimulate the growth of breast cancer cells; some of these are potentially involved in the regulation of angiogenesis which process is important for progression of breast cancer [24]. For example, leptin, one of typical adipokines, induces the proliferation of endothelial cells and elicits a marked enhancement of
angiogenesis [25]. In addition, the leptin is necessary to regulate adipose tissue expansion, which is one of the few sites of active angiogenesis in adult [26,27]. Based on these reports, we raised the possibility that visfatin may also regulate angiogenesis. As expected, we found that visfatin protein accelerated the formation of new blood vessel in chick chorioallantoic membranes (unpublished data), suggesting that visfatin-mediated angiogenesis may be involved not only in enlargement of tumor mass but also in excessive growth of adipose tissue for the settlement of obesity. We are now investigating about molecular mechanism(s) for the angiogenic activities of visfatin and elucidating the relationship between hypoxia and visfatin gene induction in normal cell types including adipose cells and normal breast cell lines. Recently, it is reported that visfatin promotes the efficient acquisition of a stable and mature smooth muscle cell phenotype that is essential for developing and remodeling of blood vessel [28]. Therefore, visfatin may act as a regulator for the maintenance of vascular homeostasis in vascular endothelial and smooth muscle cells. Our observations add the human visfatin gene to a list of genes regulated by hypoxia via the HIF-1-dependent pathway. Hypoxia, a common consequence of solid tumor growth in breast cancer and other cancers, serves to propagate a cascade of molecular pathways which include angiogenesis, glycolysis, and alterations in microenvironmental pH [29]. Consistent with a stimulatory effect of a local hypoxic environment on the expression of target genes in tumors, we showed marked expression of visfatin gene in both hypoxia-treated breast cancer cells and human breast tumor tissues. Besides the role as a tumor-promoting factor, hypoxia is one of key factors in pathogenesis of acute lung injury [30]. Recently, it was reported that pre-B-cell colony-enhancing factor (also known as visfatin) is a potential novel biomaker in acute lung injury [31].
4112
Taking into account these reports, we consider that visfatin may participate in response to hypoxia in several pathological situations. We showed that HIF-1 and HIF-2 are not equally effective on the p(708)luc visfatin promoter, as HIF-1a protein was able to induce the visfatin promoter at higher levels than HIF-2a (Fig. 3D). The activation control reporter plasmid pSV40pro-EpoHRE-Luc, containing the four HRE core sequence of erythropoietin gene, was also tested in reporter assay. However, either exogenous HIF-2a or HIF-1a induced transcription of the control HRE reporter plasmid at similar levels (data not shown), indicating the specificity of HIF-1 for the activation of visfatin promoter. In addition, several reports showed that basal HIF-2a mRNA and protein level under hypoxia were low or absent in the various breast cancer cell lines, whereas these cells dramatically expressed the high-level HIF-1a protein in response to hypoxia [32,33]. Recently, it has been reported that the expression of hypoxia-induced genes was critically dependent on HIF-1a but not HIF-2a in breast cancer cells [34]. Based on these observations, we strongly suggest that HIF-1a protein, rather than HIF-2a, may be predominantly expressed in human breast cancer cells and preferentially bind to HREs of the visfatin promoter, leading to the activation of the human visfatin gene. Hypoxia is not the only condition that stabilizes HIF-1a and activates HIF-1 transcriptional activity. Several hormones, growth factors and viral proteins, including insulin and insulin-like growth factor [35,36], angiotensin II, thrombin and platelet-derived growth factor [37], endothelin-1 [38], and Hepatitis B virus X protein [39], have been shown to increase the level of HIF-1a in various cell types. In addition, inflammatory cytokines, such as interleukin 1b and tumor necrosis factor a, also induce HIF-1 activity in normoxic cells [40]. Considering that multiple inflammatory stimuli (including lipopolysaccharide, interleukins, and tumor necrosis factor a) significantly increased visfatin expression [6], it would be of interest to investigate whether these stimuli could regulate visfatin gene expression in some types of cells via hypoxia-independent HIF-1 pathways. In Fig. 5, the expressional correlation between visfatin and HIF-1a genes was observed in neoplastic human breast tissues, supporting our result that visfatin is highly expressed under the regulation of HIF-1a protein. However, since expressional ratio of visfatin (21 of 22 cases; 95.4%) was higher than that of HIF-1a (15 of 22 cases; 68.2%), we cannot exclude possibility that HIF-1-independent pathway may be also involved in the regulation of visfatin gene expression. In conclusion, the data presented here are consistent with the visfatin gene being a hypoxia-inducible gene. Moreover, we demonstrated that hypoxia mediates the upregulation of visfatin gene via HIF-1a and HIF-1-dependent transcriptional activity, as described for several other genes regulated by low oxygen availability. Our data may provide a molecular explanation for patho-physiological regulation of visfatin in breast tumorigenesis and a novel approach as a target for the development of new therapeutic strategies. Acknowledgements: This work was supported by the research grant from a Vascular System Research Center grant from KOSEF, Korea Research Foundation Grant funded by Korea Government (MOEHRD, basic Research Promotion Fund) (KRF-2005-204C00051) (to M.-K. Bae) and MRC program of MOST/KOSEF (R13-2005-009) (to S.-K. Bae).
S.-K. Bae et al. / FEBS Letters 580 (2006) 4105–4113
References [1] Koerner, A., Kratzsch, J. and Kiess, W. (2005) Adipocytokines: leptin – the classical, resistin – the controversical, adiponectin – the promising, and more to come. Best Pract. Res. Clin. Endocrinol. Metab. 4, 525–546. [2] Trayhurn, P. and Wood, I.S. (2004) Adipokines: inflammation and the pleiotropic role of white adipose tissue. Br. J. Nutr. 3, 347–355. [3] Rose, D.P., Komninou, D. and Stephenson, G.D. (2004) Obesity, adipocytokines, and insulin resistance in breast cancer. Obes. Rev. 3, 153–165. [4] Samal, B., Sun, Y., Stearns, G., Xie, C., Suggs, S. and McNiece, I. (1994) Cloning and characterization of the cDNA encoding a novel human pre-B-cell colony-enhancing factor. Mol. Cell. Biol. 2, 1431–1437. [5] Fukuhara, A., Matsuda, M., Nishizawa, M., Segawa, K., Tanaka, M., Kishimoto, K., Matsuki, Y., Murakami, M., Ichisaka, T. and Murakami, H., et al. (2005) Visfatin: a protein secreted by visceral fat that mimics the effects of insulin. Science 5708, 426–430. [6] Ognjanovic, S., Bao, S., Yamamoto, S.Y., Garibay-Tupas, J., Samal, B. and Bryant-Greenwood, G.D. (2001) Genomic organization of the gene coding for human pre-B-cell colony enhancing factor and expression in human fetal membranes. J. Mol. Endocrinol. 2, 107–117. [7] Ye, S.Q., Zhang, L.Q., Adyshev, D., Usatyuk, P.V., Garcia, A.N., Lavoie, T.L., Verin, A.D., Natarajan, V. and Garcia, J.G. (2005) Pre-B-cell-colony-enhancing factor is critically involved in thrombin-induced lung endothelial cell barrier dysregulation. Microvasc. Res. 3, 142–151. [8] Hufton, S.E., Moerkerk, P.T., Brandwijk, R., de Bruine, A.P., Arends, J.W. and Hoogenboom, H.R. (1999) A profile of differentially expressed genes in primary colorectal cancer using suppression subtractive hybridization. FEBS Lett. 1–2, 77–82. [9] Folgueira, M.A., Carraro, D.M., Brentani, H., Patrao, D.F., Barbosa, E.M., Netto, M.M., Caldeira, J.R., Katayama, M.L., Soares, F.A. and Oliveira, C.T., et al. (2005) Gene expression profile associated with response to doxorubicin-based therapy in breast cancer. Clin. Cancer Res. 20, 7434–7443. [10] Bergers, G. and Benjamin, L.E. (2003) Tumorigenesis and the angiogenic switch. Nat. Rev. Cancer 6, 401–410. [11] Vaupel, P. (2004) The role of hypoxia-induced factors in tumor progression. Oncologist, 10–17. [12] Kallio, P.J., Wilson, W.J., O’Brien, S., Makino, Y. and Poellinger, L. (1999) Regulation of the hypoxia-inducible transcription factor 1alpha by the ubiquitin-proteasome pathway. J. Biol. Chem. 10, 6519–6525. [13] Grosfeld, A., Andre, J., Hauguel-De Mouzon, S., Berra, E., Pouyssegur, J. and Guerre-Millo, M. (2002) Hypoxia-inducible factor 1 transactivates the human leptin gene promoter. J. Biol. Chem. 45, 42953–42957. [14] Ambrosini, G., Nath, A.K., Sierra-Honigmann, M.R. and FloresRiveros, J. (2002) Transcriptional activation of the human leptin gene in response to hypoxia. Involvement of hypoxia-inducible factor 1. J. Biol. Chem. 37, 34601–34609. [15] Jeong, H.J., Chung, H.S., Lee, B.R., Kim, S.J., Yoo, S.J., Hong, S.H. and Kim, H.M. (2003) Expression of proinflammatory cytokines via HIF-1alpha and NF-kappaB activation on desferrioxamine-stimulated HMC-1 cells. Biochem. Biophys. Res. Commun. 4, 805–811. [16] Forsythe, J.A., Jiang, B.H., Iyer, N.V., Agani, F., Leung, S.W., Koos, R.D. and Semenza, G.L. (1996) Activation of vascular endothelial growth factor gene transcription by hypoxia-inducible factor 1. Mol. Cell. Biol. 9, 4604–4613. [17] Jin, K., Mao, X.O., Sun, Y., Xie, L., Jin, L., Nishi, E., Klagsbrun, M. and Greenberg, D.A. (2002) Heparin-binding epidermal growth factor-like growth factor: hypoxia-inducible expression in vitro and stimulation of neurogenesis in vitro and in vivo. J. Neurosci. 13, 5365–5373. [18] Warnecke, C., Griethe, W., Weidemann, A., Jurgensen, J.S., Willam, C., Bachmann, S., Ivashchenko, Y., Wagner, I., Frei, U., Wiesener, M. and Eckardt, K.U. (2003) Activation of the hypoxia-inducible factor-pathway and stimulation of angiogenesis by application of prolyl hydroxylase inhibitors. FASEB J. 9, 1186–1188.
S.-K. Bae et al. / FEBS Letters 580 (2006) 4105–4113 [19] Gassmann, M., Kvietikova, I., Rolfs, A. and Wenger, R.H. (1997) Oxygen- and dioxin-regulated gene expression in mouse hepatoma cells. Kidney Int. 2, 567–574. [20] Tian, H., McKnight, S.L. and Russell, D.W. (1997) Endothelial PAS domain protein 1 (EPAS1), a transcription factor selectively expressed in endothelial cells. Genes Dev. 11, 72–82. [21] Ema, M., Taya, S., Yokotani, N., Sogawa, K., Matsuda, Y. and Fujii-Kuriyama, Y. (1997) A novel bHLH-PAS factor with close sequence similarity to hypoxia-inducible factor 1alpha regulates the VEGF expression and is potentially involved in lung and vascular development. Proc. Natl. Acad. Sci. USA 94, 4273–4278. [22] Jia, S.H., Li, Y., Parodo, J., Kapus, A., Fan, L., Rotstein, O.D. and Marshall, J.C. (2004) Pre-B cell colony-enhancing factor inhibits neutrophil apoptosis in experimental inflammation and clinical sepsis. J. Clin. Invest. 9, 1318–1327. [23] Stephenson, G.D. and Rose, D.P. (2003) Breast cancer and obesity: an update. Nutr. Cancer 1, 1–16. [24] Schneider, B.P. and Miller, K.D. (2005) Angiogenesis of breast cancer. J. Clin. Oncol. 8, 1782–1790. [25] Bouloumie, A., Drexler, H.C., Lafontan, M. and Busse, R. (1998) Leptin, the product of Ob gene, promotes angiogenesis. Circ. Res. 10, 1059–1066. [26] Hausman, G.J. and Richardson, R.L. (2004) Adipose tissue angiogenesis. J. Anim. Sci. 3, 925–934. [27] Caro, J.F., Sinha, M.K., Kolaczynski, J.W., Zhang, P.L. and Considine, R.V. (1996) Leptin: the tale of an obesity gene. Diabetes 11, 1455–1462. [28] van der Veer, E., Nong, Z., O’Neil, C., Urquhart, B., Freeman, D. and Pickering, J.G. (2005) Pre-B-cell colony-enhancing factor regulates NAD+-dependent protein deacetylase activity and promotes vascular smooth muscle cell maturation. Circ. Res. 1, 25–34. [29] Goonewardene, T.I., Sowter, H.M. and Harris, A.L. (2002) Hypoxia-induced pathways in breast cancer. Microsc. Res. Tech. 1, 41–48. [30] Agorreta, J., Zulueta, J.J., Montuenga, L.M. and Garayoa, M. (2005) Adrenomedullin expression in a rat model of acute lung injury induced by hypoxia and LPS. Am. J. Physiol. Lung Cell. Mol. Physiol. 3, L536–L545. [31] Ye, S.Q., Simon, B.A., Maloney, J.P., Zambelli-Weiner, A., Gao, L., Grant, A., Easley, R.B., McVerry, B.J., Tuder, R.M. and
4113
[32]
[33] [34]
[35]
[36]
[37]
[38]
[39]
[40]
Standiford, T., et al. (2005) Pre-B-cell colony-enhancing factor as a potential novel biomarker in acute lung injury. Am. J. Respir. Crit. Care Med. 4, 361–370. Blancher, C., Moore, J.W., Talks, K.L., Houlbrook, S. and Harris, A.L. (2000) Relationship of hypoxia-inducible factor (HIF)-1alpha and HIF-2alpha expression to vascular endothelial growth factor induction and hypoxia survival in human breast cancer cell lines. Cancer Res. 60, 7106–7113. Ameri, K., Burke, B., Lewis, C.E. and Harris, A.L. (2002) Regulation of a rat VL30 element in human breast cancer cells in hypoxia and anoxia: role of HIF-1. Br. J. Cancer 87, 1173–1181. Sowter, H.M., Raval, R.R., Moore, J.W., Ratcliffe, P.J. and Harris, A.L. (2003) Predominant role of hypoxia-inducible transcription factor (Hif)-1alpha versus Hif-2alpha in regulation of the transcriptional response to hypoxia. Cancer Res. 63, 6130– 6134. Zelzer, E., Levy, Y., Kahana, C., Shilo, B.Z., Rubinstein, M. and Cohen, B. (1998) Insulin induces transcription of target genes through the hypoxia-inducible factor HIF-1alpha/ARNT. EMBO J. 17, 5085–5094. Feldser, D., Agani, F., Iyer, N.V., Pak, B., Ferreira, G. and Semenza, G.L. (1999) Reciprocal positive regulation of hypoxiainducible factor 1alpha and insulin-like growth factor 2. Cancer Res. 16, 3915–3918. Richard, D.E., Berra, E. and Pouyssegur, J. (2000) Nonhypoxic pathway mediates the induction of hypoxia-inducible factor 1alpha in vascular smooth muscle cells. J. Biol. Chem. 35, 26765–26771. Spinella, F., Rosano, L., Di Castro, V., Natali, P.G. and Bagnato, A. (2002) Endothelin-1 induces vascular endothelial growth factor by increasing hypoxia-inducible factor-1alpha in ovarian carcinoma cells. J. Biol. Chem. 31, 27850–27855. Moon, E.J., Jeong, C.H., Jeong, J.W., Kim, K.R., Yu, D.Y., Murakami, S., Kim, C.W. and Kim, K.W. (2004) Hepatitis B virus X protein induces angiogenesis by stabilizing hypoxiainducible factor-1alpha. FASEB J. 2, 382–384. Hellwig-Burgel, T., Rutkowski, K., Metzen, E., Fandrey, J. and Jelkmann, W. (1999) Interleukin-1beta and tumor necrosis factoralpha stimulate DNA binding of hypoxia-inducible factor-1. Blood 5, 1561–1567.