Epidermal Growth Factor Receptor Transactivation by Intracellular

Home , Retinoid

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

Biochimica et Biophysica Acta 1833 (2013) 2029–2038

Contents lists available at SciVerse ScienceDirect

Biochimica et Biophysica Acta

journal homepage: www.elsevier.com/locate/bbamcr

Epidermal growth factor receptor transactivation by intracellular

prostaglandin E2-activated prostaglandin E2 receptors. Role in retinoic acid receptor-β up-regulation

Ana B. Fernández-Martínez ⁎, Francisco J. Lucio Cazaña

Departamento de Biología de Sistemas, Universidad de Alcalá, Alcalá de Henares, Madrid, Spain

article info abstract

Article history: The pharmacological modulation of renoprotective factor vascular endothelial growth factor-A (VEGF-A) in Received 28 December 2012 the proximal tubule has therapeutic interest. In human proximal tubular HK-2 cells, treatment with Received in revised form 22 April 2013 all-trans retinoic acid or prostaglandin E2 (PGE2) triggers the production of VEGF-A. The pathway involves Accepted 24 April 2013 an initial increase in intracellular PGE , followed by activation of EP receptors (PGE receptors, most Available online 2 May 2013 2 2 likely an intracellular subset) and increase in retinoic acid receptor-β (RARβ) expression. RARβ then up-regulates transcription factor hypoxia-inducible factor-1α (HIF-1α), which increases the transcription Keywords: Intracellular PGE2 and production of VEGF-A. Here we studied the role in this pathway of epidermal growth factor receptor RARß (EGFR) transactivation by EP receptors. We found that EGFR inhibitor AG1478 prevented the increase in HIF-1α VEGF-A production induced by PGE2- and all-trans retinoic acid. This effect was due to the inhibition of EGFR transactivation the transcriptional up-regulation of RARβ, which resulted in loss of the RARβ-dependent transcriptional VEGF up-regulation of HIF-1α. PGE2 and all-trans retinoic acid also increased EGFR phosphorylation and this effect

was sensitive to antagonists of EP receptors. The role of intracellular PGE2 was indicated by two facts;

i) PGE2-induced EGFR phosphorylation was substantially prevented by inhibitor of prostaglandin uptake transporter bromocresol green and ii) all-trans retinoic acid treatment, which enhanced intracellular but

not extracellular PGE2, had lower effect on EGFR phosphorylation upon pre-treatment with cyclooxygenase

inhibitor diclofenac. Thus, EGFR transactivation by intracellular PGE2-activated EP receptors results in the sequential activation of RARβ and HIF-1α leading to increased production of VEGF-A and it may be a target for the therapeutic modulation of HIF-1α/VEGF-A. © 2013 Elsevier B.V. All rights reserved.

1. Introduction cellular energy metabolism, angiogenesis, erythropoiesis and other biological processes [2]. These products include enzymes involved in Critical mediators of cellular adaptation to hypoxia are hypoxia- glucose uptake and metabolism, carbonic anhydrase IX, erythropoie- inducible factors (HIFs), HIF-1 and HIF-2 being the most prominent tin and vascular endothelial growth factor (VEGF) [3]. HIF-1α expres- and best characterized. HIF is a transcription factor that consists of sion is also increased in normoxia: growth and coagulation factors, an oxygen-sensitive α-subunit, HIF-α and a constitutively expressed hormones, stress factors and inﬂammatory mediators such as prosta-

β-subunit, HIF-β. In a hypoxic environment HIF-α is stabilized and glandin E2 (PGE2), cytokines, interleukin-1β, tumor necrosis factor-α, translocates to the nucleus, and, together with the β subunit and tran- nitric oxide and reactive oxygen species increase the expression of scriptional coactivators, it binds to hypoxia-responsive elements HIF-1α [4,5]. They increase HIF-1α transcription and/or translation (HRE) in target genes [1]. HIF-1α supports tissue survival in hypoxia [4,6] but not HIF-1α protein stability. by regulating the expression of gene products that are involved in All-trans retinoic acid (ATRA) is the carboxylic acid form of vitamin A and its major metabolite. The actions of ATRA are generally mediated by binding to retinoic acid receptors (RARs), which are members of the Abbreviations: iPGE2, intracellular prostaglandin E2; EPR, EP receptors; RARβ, retinoic acid receptor-β; HIF-1α, factor hypoxia-inducible factor-1α; VEGF-A, factor nuclear receptor superfamily. RARs are key regulators of multiple phys- vascular endothelial growth factor-A; EGFR, epidermal growth factor receptor; ATRA, iological processes, most signiﬁcantly embryonic development and all-trans retinoic acid; RARE, retinoic acid receptor response element; BG, bromocresol organ homeostasis. At the cellular level, RARs contribute to the regula- green tion of gene networks that control cell growth, differentiation, survival ⁎ Corresponding author at: Departamento de Fisiología, Facultad de Medicina, Universidad and death [7]. Three RAR subtypes (RARα,RARβ and RARγ)have de Alcalá, Alcalá de Henares, 28871 Madrid, Spain. Tel.: +34 918854515; fax: +34 ﬁ 918854590. been identi ed. They act as ligand-regulated transcription factors by E-mail address: [email protected] (A.B. Fernández-Martínez). dimerizing with retinoid X receptors (RXR) and binding to retinoic

0167-4889/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.bbamcr.2013.04.013 2030 A.B. Fernández-Martínez, F.J. Lucio Cazaña / Biochimica et Biophysica Acta 1833 (2013) 2029–2038 acid response elements (RARE), which are located in the regulatory 2.2. Cell culture and treatments regions of their target genes [8].RARβ itself is a RARs target [9]:its promoter region contains a RARE (β-RARE) responsible for the up- Human kidney HK-2 cells, human prostate cancer cells (PC3) and regulation of RARβ induced by ATRA in many different cell types [10]. human renal cell carcinoma (A498) were purchased from American We have recently shown in human renal proximal tubular HK-2 Type Culture Collection (Rockville, MD). HK-2 were maintained in cells that the up-regulation of RARβ by cyclooxygenase (COX)- DMEM/F12 supplemented with 10% foetal bovine serum (FBS), 1% dependent production of PGE2 plays a critical role in the increase in penicillin/streptomycin/amphoterycin B and 1% glycine (Invitrogen, HIF-1α expression induced by ATRA, interleukin-1β, PGE2 analogue Carlsbad, CA) and 1% Insulin–Transferrine–Selenium (Sigma, St. Louis, 16,16-dimethyl-PGE2 (16,16-PGE2) or hypoxia [11,12]. Interestingly, MO). PC3 were maintained in RPMI and A498 cells were maintained the transcriptional inhibitor actinomycin D prevented the increase in MEM Eagle, and supplemented with 10% FBS and 1% penicillin/ in RARβ and HIF-α expression induced in normoxia but not in hypox- streptomycin/amphoterycin B. ia [13]. The culture was performed in a humidiﬁed 5% CO2 environment at It is accepted that, once prostanoids are formed, they are quickly 37 °C. In all experiments, cells were plated at 70–90% conﬂuence and released to the outside of cells by simple diffusion and act as auto- 24 h later they were treated with 1 μM 16,16-PGE2 and/or 10 μM crine or paracrine mediators in the vicinity of their sites of production ATRA, for different periods of time. When appropriated, the following to maintain local homeostasis [14–17]. Therefore, it has generally agents were added 1 h before: 50 μM bromocresol green (inhibitor of been assumed that PGE2 exerts its actions via plasma membrane prostaglandin uptake transporter), 0.3 mM diclofenac (COX inhibi- spanning G-protein coupled EP receptors (EPR: EP1, EP2, EP3 and tor), 10 μM AH6809 (antagonist of EP1-3 receptors) or GW627368X EP4), each EP activating a distinct G protein-coupled signaling path- (antagonist of EP4 receptor), 1 μg/ml actinomycin D (transcription in- way [14,15]. However, the up-regulation of RARβ in HK-2 cells by hibitor), 2 μM LE135 (RARβ antagonist), 10 μM YC1 (HIF-1α inhibitor), HIF-1α-increasing agents was most likely due to intracellular EPR 1 μM AG1478 (inhibitor of EGFR activation) were added 1 h before the activated by intracellular PGE2 (iPGE2) [12]. These data are in good treatment. agreement with the location of functional EPR at the nuclear mem- branes of a variety of cell types and tissues [18–20]. In fact, in HK-2 2.3. Real-time quantitative RT-PCR cells the four EPR are located not only in the cell membrane but also inside the cell [11]. We have also provided data on the regulation of Total cell RNA from HK-2 cells was isolated with TriReagent (Sigma,

RARβ expression by iPGE2 in other cell lines, which indicates that it St. Louis, MO). One microgram of total RNA, extracted by using the is not a phenomenon restricted to HK-2 cells [12]. It is important to TaqMan® Gene Expression Master Mix (Applied Biosystems), was take into account that the classical downstream plasma membrane reverse transcribed in a total reaction volume of 25 μlbymeansofincu-

EPR-linked signaling cascades pathways activated by PGE2 may not bation at 42 °C for 30 min using random hexamer primers. For quanti- necessarily be the same ones linked to iPGE2-activated intracellular tative PCR, mRNAs were ampliﬁed and quantiﬁed by TaqMan probes EPR. For instance: it has been proposed that iPGE2, acting through in- (Applied Biosystems) (Table 1). As template, cDNA from the target tracellular EP4 receptors, maintains epidermal growth factor receptor and housekeeping gene GAPDH were prepared in separate tubes for (EGFR) in an activated state in cardiac myocytes [21]. This activation each primer master mixture reaction. The comparative threshold cycle of EGFR through a molecule that does not, itself, bind EGFR is called (CT) method was used to calculate the relative expression [22]. transactivation.

The elucidation of the mechanism through which iPGE2-activated 2.4. Protein isolation and Western-blotting EPR increase RARβ expression has important implications. For instance, the up-regulation of the RARβ gene by ATRA presumably HK-2 cells were stimulated and washed twice with ice-cold PBS and plays a critical role in amplifying the ATRA response [10]. In our then harvested, scraped into ice-cold PBS, and then pelleted by centrifu- more speciﬁc context, we have shown that the increase in the expres- gation at 500 ×g, for 5 min, at 4 °C. Cells were kept on ice for 30 min in sion of HIF-1α through this pathway enhances the production of a solution containing 50 mM Tris–HCl (pH 7.5), 150 mM NaCl, 1% Triton VEGF-A, a putative renoprotective factor [12]. In the present work X-100, 0.5% sodium deoxycholate and protease inhibitors. Thereafter, we have investigated the role of EGFR in the iPGE2-dependent path- the cells were pelleted by centrifugation at 4000 ×g, for 5 min, at way leading to the increase in VEGF-A production under normoxic 4 °C. Proteins from cell lysates were denatured by heating. Then, they conditions. We have found that the transactivation of EGFR by were resolved by 10% SDS-PAGE, and blotted onto a nitrocellulose iPGE2-activated EPR (upon treatment with 16,16-PGE2 or ATRA) membrane (BioTrace/NT), for 1 h, in 50 mM Tris–HCl, 380 mM glycine, plays a crucial role in the transcriptional up-regulation of RARβ and 0.1% SDS, and 20% methanol. Blots were incubated overnight at 4 °C that the increase in RARβ expression led, in turn, to the transcription- with rabbit anti-RARβ (1:1000), mouse anti-HIF-1α (1:1000) and al up-regulation of HIF-1α and, eventually, to the increase in the pro- rabbit anti-phospho-EGFR (1:500) antibodies. After treatment for 1 h duction of VEGF-A. at room temperature with the corresponding secondary antiserum (1:4000), the signals were detected with enhanced chemiluminescence reagent using β-actin antibody (1: 25000) as loading control. 2. Material and Methods

2.5. Determination of PGE2 formation 2.1. Reagents

PGE2 was determined in culture medium and cell lysates using a 16,16-PGE2, bromocresol green, diclofenac, AH6809 and GW627368X commercially available enzyme immunoabsorbent assay (EIA) kit were purchased from Cayman Chemical (Ann Arbor, MI). AG1478, actinomycin D, YC1 and ATRA were purchased from Sigma (St. Louis, β Table 1 MO). RAR antagonist LE135 was a generous gift from Prof. Hiroyuki TaqMan® probes table. Kagechika (Tokyo Medical and Dental University, Japan). Primary antibodies against RARβ and phospho-EGFR were purchased from Santa Gene Applied Biosystems ref. Cruz Biotechnology (Santa Cruz, CA). Antibody against HIF-1α was pur- HIF-1α Hs00936368m1 chased from BD Biosciences (Palo Alto, CA) and anti-β-actin antibody RARβ Hs00977140m1 was from Sigma Chemical Co. (St. Louis, MO). GAPDH Hs03929097g1 A.B. Fernández-Martínez, F.J. Lucio Cazaña / Biochimica et Biophysica Acta 1833 (2013) 2029–2038 2031

(Cayman Chemical Company, Ann Arbor, MI) following the manu- 3. Results facturer's protocol.

3.1. PGE2-induced RARβ up-regulation is due to increased RARβ mRNA transcription by iPGE2. Role of RARE and EPR 2.6. Determination of VEGF secretion Transcriptional mechanisms are responsible for the up-regulation of 4 HK-2 cells were placed in 24 well plates (7.5 × 10 cells/well) for RARβ by ATRA [7]. Therefore, we hypothesized that the up-regulation of 24 h. Then they were treated as indicated in Fig. 4c and d and the medi- RARβ by iPGE2 was due to an increase in the expression of RARβ mRNA. − um was removed and kept at 80 °C. VEGF was analyzed by ELISA assay In order to test this hypothesis, we first investigated whether inhibition using a human VEGF Human Development Kit (Tebu-bio, Le-Perray-en- of transcription prevented PGE2-induced RARβ protein up-regulation. Yvelines, France). To this end, the expression of RARβ protein and mRNA was assessed by Western blot and Q-RT-PCR, respectively, in HK-2 cells which were first incubated for 1 h with or without transcription inhibitor actinomy- 2.7. Plasmids, siRNA, transient transfection and luciferase assay cin D and then with 16,16-PGE2.16,16-PGE2-induced increase in RARβ protein and mRNA expression was abolished by actinomycin D (Fig. 1a Transient transfection with luciferase reporter plasmid tk-β-RARE- and b), which suggested that transcriptional mechanisms were respon- Luc (the plasmid, which contains a part of the RARE sequence from sible for the up-regulation of RARβ by 16,16-PGE . Finally, when cells the promoter of the human RARβ gene, was a generous gift of Prof. 2 were incubated first with 16,16-PGE and then with actinomycin D Ronald Evans, The Salk institute, CA, USA) or luciferase reporter plasmid 2 for up to 4 h in order to determine the RARβ mRNA half-life, we p9HIF1-Luc (the plasmid, which contains nine copies of a part of the found no differences between 16,16-PGE -treated cells and control human HIF binding sequence of the 5′ human VEGF gene promoter 2 cells (Fig. 1b, inset). These results indicate that transcriptional mecha- was generously gifted by Dr. Manuel Ortiz de Landazuri, Servicio de nisms, but not post-transcriptional ones, are involved in 16,16-PGE - Inmunología, Hospital de La Princesa, Madrid, Spain) or luciferase re- 2 induced RARβ up-regulation. porter plasmid RARβ 2-tk-luc reporter (which was made by cloning We next analyzed the effect of 16,16-PGE on the activity of the the RARβ promoter in a pre-existing tk-luc plasmid, was a generous 2 RARβ gene promoter. HK-2 cells were transiently transfected with ei- gift Rene Bernard, Division of Molecular Carcinogenesis and Center for ther tk-β-RARE-Luc (RARE construct from the RARβ gene) or a con- Biomedical Genetics, The Netherlands Cancer Institute, Amsterdam, struct of the whole RARβ promoter. The treatment with 16,16-PGE The Netherlands) was performed as follows: 3 × 105 cells per well 2 resulted in an increase in the luciferase activity of the RARE and were plated in 6-well plates 24 h before transfection. Cells were incu- RARβ promoter constructs (Fig. 1c) as well as in the expression of bated 12 h at 37 °C with 1 ml OptiMEM (Invitrogen, CA) containing RARβ mRNA (Fig. 1d). Furthermore, the time-course of the activation complexes of 5 μg lipofectamine (Invitrogen, CA), 1 μghumantk-ß- of RARE-Luc in HK-2 cells (Fig. 1c, inset), was compatible with its RARE-Luc or p9HIF1-Luc or RARβ2-tk-luc reporter and 1 μg renilla lucif- participation in the up-regulation of RARβ mRNA by 16,16-PGE . In- erase reporter as an internal control. Transfected cells were next incu- 2 creased luciferase activity upon treatment with 16,16-PGE was also bated with complete growth medium for 24 h and then pretreated 2 observed in human prostate cancer cells PC3 and human renal cell with different drugs for 1 h and treated either 16,16-PGE or ATRA for 2 carcinoma A498 cells (Fig. 1c). This suggests that the induction of 4h.Finally, firefly luciferase activity was measured with a Lumat RARE activity by 16,16-PGE might be a widespread effect of the LB9506 luminometer (Berthold Technologies, Herts, UK) and normal- 2 prostanoid and not a phenomenon restricted to HK-2 cells. ized against renilla luciferase activity by using the dual-luciferase To confirm that the transcriptional effects of 16,16-PGE on RARβ reporter assay system (Promega, Madison, WI). 2 expression were actually due to iPGE , we analyzed the effect of the For inhibiting HIF-1α and RARβ expression, we used HIF-1α siRNA 2 inhibition of prostaglandin uptake by pre-treatment for 1 h with sc-44225 and RARβ siRNA (sc-29466) (Santa Cruz Biotechnologies), 50 μM bromocresol green (BG). BG prevented (Fig. 1e) the increase in respectively, and scramble siRNA AM4637 (Applied Biosystems) as a RARE-Luc activity, RARβ promoter activity (inset) and RARβ mRNA control. HK-2 cells at 70% confluence were transfected with 100 nM expression induced by 16,16-PGE . RARβ,HIF-1α siRNA or scramble siRNA using Lipofectamine 2000 2 Finally, we assessed the effect of antagonist of EP1-EP3 receptors reagent. Twenty four hours after transfection, cells were used for the AH6809 [23] or antagonist of EP4 receptor GW627368X [24] on experiments. 16,16-PGE2-induced increase in RARE-Luc activity and expression of RARβ mRNA. As shown in Fig. 1f, preincubation for 1 h with these 2.8. Immunofluorescence antagonists blocked the increase in RARE-Luc activity and the up- regulation of RARβ mRNA induced by 16,16-PGE2. Cells were fixed for 20 min with ice-cold methanol following 1 h of In summary, the results shown in Fig. 1 support the view that the β blocking at room temperature. Blocking solution (4 % BSA in PBS) was transcriptional up-regulation of RAR by iPGE2-activated EPR (most further used for dilution of primary and secondary antibodies. Next likely through RARE activation) is responsible for 16,16-PGE2-induced β cells were incubated at 4 °C with primary anti-phospho EGFR antibody increase in RAR protein expression. diluted 1:50 for overnight following PBS washing and for 1 h with secondary antibody (Alexa FLUO 488 goat 1:1000) following PBS washing. 3.2. iPGE2, in an EPR antagonist sensitive manner, mediates the β Finally coverslips were mounted with ProLong Gold antifade reagent transcriptional increase of RAR expression induced by ATRA with DAPI (Invitrogen). The detection was performed by confocal laser scan microscopy LEICA TCS-SL (Heidelberg, Germany). We next analyzed the implication of iPGE2 in the transcriptional up-regulation of RARβ by ATRA.

In HK-2 cells, iPGE2 increases upon treatment with ATRA (Fig. 2a), 2.9. Statistical analysis while the release of PGE2 to the extracellular medium, compared to control cells, actually diminishes. Consequently, ATRA-induced in-

Each experiment was repeated at least three times. The results are crease in iPGE2 would not be sensitive to BG. Instead, it is adequately expressed as the mean ± SD. They were subjected to one way analysis prevented by a COX inhibitor such is diclofenac (Fig. 2a). of variance (ANOVA) following by the Bonferroni's test for multiple We next assessed the effect on ATRA-induced increase in RARβ comparisons. The level of signiﬁcance was set at P b 0.05. mRNA and RARE-Luc activity of pre-incubation for 1 h with 0.3 mM 2032 A.B. Fernández-Martínez, F.J. Lucio Cazaña / Biochimica et Biophysica Acta 1833 (2013) 2029–2038 A.B. Fernández-Martínez, F.J. Lucio Cazaña / Biochimica et Biophysica Acta 1833 (2013) 2029–2038 2033

Fig. 2. iPGE2, in an EPR antagonist-sensitive manner, mediates the transcriptional increase of RARβ expression induced by ATRA. a) Treatment with ATRA increases the intracellular content in PGE2 but not the release of PGE2 to the extracellular medium. HK-2 cells were incubated for 1 h with 0.3 mM diclofenac or in control conditions and then they were treated with 10 μM ATRA for 24 h. PGE2 levels were measured by EIA in the culture medium and in cell lysates. b) ATRA increases RARβ mRNA expression in a diclofenac- and EPR antagonist-sensitive manner. HK-2 cells were incubated for 1 h with 0.3 mM diclofenac, 10 μM AH6809 or with 10 μM GW627368X and then with 10 μM ATRA for 4 h. c) ATRA increases RARE activity in a diclofenac- and EPR antagonist-sensitive manner. HK-2 cells (left panel), PC3 cells and A498 cells (right panel) were transiently transfected with RARE-Luc plasmid construct. Then, they were pre-incubated with diclofenac or EPR antagonists and incubated with ATRA as in b). d) ATRA and 16,16-PGE2 do not synergize to increase RARE-Luc activity. HK-2 cells were transfected as in c) and treated for 4 h with ATRA or/and 16,16-PGE2. General information: 1) Reporter gene experiments: Luciferase activity was normalized by Renilla luciferase activity and expressed as relative luminescence units (RLU). 2) Bars and error bars in graphs: Each bar represents the mean ± SD of 3 different experiments.*Pb 0.05 vs. control.

COX inhibitor diclofenac, 10 μM antagonist of EP1-EP3 receptors the activity of RARE-Luc. Our results (Fig. 2d) excluded this pos- AH6809 or 10 μM antagonist of EP4 receptor GW627368X: the three sibility, which suggests that ATRA activates RARE through the agents tested blocked both increases (Fig. 2b and c, left panel). Of same mechanism responsible for 16,16-PGE2-induced RARE-Luc note, we confirmed that diclofenac, AH6809 and GW627368X also activity. prevented ATRA-induced increase in RARE-Luc activity in PC3 Since ATRA is the physiological activator of RARE, the results and A498 cell lines (Fig. 2c, right panel). We finally studied in HK-2 shown in Fig. 2 confirm the critical role of iPGE2 in the transcriptional cells whether ATRA and 16,16-PGE2 synergized to over-increase up-regulation of RARβ.

Fig. 1. PGE2-induced RARβ up-regulation is due to increased RARβ mRNA transcription by iPGE2. Role of RARE and EPR. a) Inhibition of transcription prevents PGE2-induced RARβ protein up-regulation. HK-2 cells were incubated for 1 h with/without 1 μg/ml actinomycin D (actD) and then for 8 h with/without 1 μM 16,16-PGE2. b) 16,16-PGE2-induced RARβ mRNA up-regulation is prevented by inhibition of transcription. Cells were incubated as above for 4 h. Total cell RNA was extracted and RARβ mRNA levels were analyzed by

Q-RT-PCR. Inset: Cells were incubated for 4 h with/without 1 μM 16,16-PGE2 and treated with 1 μg/ml act D for 0 to 4 h to analyze the decay of RARβ mRNA. c) 16,16-PGE2 increases the activity of human retinoic acid-responsive element (RARE)-Luc and RARβ-Luc constructs. HK-2 cells, human prostate cancer cells (PC3) and human renal cell carcinoma

(A498) were transiently transfected with the plasmid constructs and incubated for 4 h with or without 1 μM 16,16-PGE2. Inset: Time-course of RARE activity in 16,16-PGE2-treated

HK-2 cells. d) 16,16-PGE2 induces RARβ mRNA up-regulation. HK-2 cells were incubated with 1 μM 16,16-PGE2 for up to 8 h and RARβ mRNA levels were analyzed by Q-RT-PCR. e) Inhibition of prostaglandin uptake prevents 16,16-PGE2-induced increase in the activity of RARE-Luc and RARβ-Luc constructs and in RARβ mRNA expression. HK-2 cells transfected with the constructs were incubated with bromocresol green (50 μM/1 h) and then with PGE2 (1 μM/4 h). f) Antagonists of EP receptors block the increase in RARE activity and RARβ mRNA induced by 16,16-PGE2. HK-2 cells were preincubated for 1 h with 10 μM AH6809 (an antagonist of EPR1-3) or with 10 μM GW627368X (an antagonist of EPR4) and then with 1 μM 16,16-PGE2 for 4 h. General information: 1) Western blot experiments: Each experiment was repeated three times and a representative photograph of the results is shown. 2) Reporter gene experiments: Luciferase activity was normalized by Renilla luciferase activity and expressed as relative luminescence units (RLU). 3) Bars and error bars in graphs: Each bar represents the mean ± SD of 3 different experiments. *P b 0.05 vs. control. 2034 A.B. Fernández-Martínez, F.J. Lucio Cazaña / Biochimica et Biophysica Acta 1833 (2013) 2029–2038

Fig. 3. EP receptor-dependent EGFR transactivation is responsible for iPGE2-induced RARβ up-regulation. a) ATRA and 16,16-PGE2 induce EGFR phosphorylation. HK-2 cells were incubated for up to 45 min with 10 μMATRAor1μMPGE2 and phosphorylation of EGFR was assessed by Western blot or immunoﬂuorescence analysis (photographs, taken after 30–45 min are shown). b) 16,16-PGE2- and ATRA-induced EGFR phosphorylation are prevented by inhibition of the increase in iPGE2- and antagonism of EPR. HK-2 cells were preincubated for 1 h with 50 μM bromocresol green, 0.3 mM diclofenac, 10 μMAH6809or10μM GW627368X and then for 30 min with either 1 μM 16,16-PGE2 or 10 μM ATRA. Normalized density ratio of EGFR-P over

β-actin is indicated for each band. c) ATRA- and 16,16-PGE2-induced increase in RARE-Luc activity and RARβ mRNA and protein expression are prevented by inhibitor of EGFR phosphorylation

AG1478. Cells were pre-incubated for 1 h with 1 μM AG1478 and then with 10 μMATRAor1 μM 16,16-PGE2 for 4 h. d) ATRA- and 16,16-PGE2-induced increase in RARE-Luc activity in PC3 and A498 cell lines is prevented by AG1478. Cells transfected with RARE-Luc plasmid construct were treated as in c). General information: 1) Western blot: Each experiment was repeated three times and a representative photograph of the results is shown. 2) Reporter gene experiments: Luciferase activity was normalized by Renilla luciferase activity and expressed as relative lumi- nescenceunits(RLU).3)Barsanderrorbarsingraphs:Eachbarrepresentsthemean±SDof3differentexperiments.*Pb 0.05 vs. control. **P b 0.05 vs. GW627368X. A.B. Fernández-Martínez, F.J. Lucio Cazaña / Biochimica et Biophysica Acta 1833 (2013) 2029–2038 2035

3.3. EP receptor-dependent EGFR transactivation is responsible for indicated that the two agents prevented the PGE2-induced increase iPGE2-induced RARβ up-regulation in HIF-1α mRNA. Then we examined the involvement of EGFR in the iPGE2-induced increase in HIF-1α mRNA expression by incubat- We next investigated the role of EGFR in the up-regulation of ing cells with AG1478 prior to treatment with 16,16-PGE2 or ATRA. RARβ. To this aim, we first studied the time-course of the phosphor- AG1478 prevented 16,16-PGE2- and ATRA-induced increase in the ylation of EGFR in HK-2 cells incubated with ATRA or 16,16-PGE2 for expression of HIF-1α mRNA (Fig. 4b). Taking into account that up to 45 min. As seen in Fig. 3a, upper panel, EGFR phosphorylation 16,16-PGE2 increases the expression of HIF-1α protein in an actino- increased from the very first minutes with both treatments and mycin D-sensitive manner (Fig.1a), the results shown in Fig. 4a and remained in high values after 45 min incubation. These results were b strongly suggest that up-regulation of HIF-1α is due to EGFR-, confirmed by immunofluorescence analysis: phosphorylated EGFR RARβ-dependent increase in HIF-1α mRNA by iPGE2. was ubiquitous in control HK-2 cells but its nuclear location, as well We next asked whether iPGE2-induced HIF-1α up-regulation results as its overall expression, increased upon treatment with ATRA or in activation of the HRE in the VEGF-A gene promoter. We used HK-2

16,16-PGE2 (Fig. 3a, lower panel). These changes happened in a man- cells transiently transfected with p9HIF1-Luc (a plasmid which ner time-dependent (for clarity we only show the microphotographs contained part of the human HIF binding sequence of the 5′ human taken after 45 or 30 min treatment). VEGF gene promoter). Fig. 4c shows that the treatment for 4 h with

To analyze the role of iPGE2 in the phosphorylation of EGFR, HK-2 1 μM16,16-PGE2 determined a strong increase in luciferase activity, cells were first pre-incubated for 1 h with either the inhibitor of pros- which was abolished by BG and YC1 (an inhibitor of HIF-1α [25]). BG taglandin uptake transporter BG or COX inhibitor diclofenac prior to and transfection with HIF-1α siRNA also prevented the increase in the treatment with 16,16-PGE2 or ATRA, respectively, and cell levels of production of VEGF-A (quantified by ELISA) in 16,16-PGE2-treated phosphorylated EGFR were assessed by Western blot analysis. Our re- cells. Finally, to ensure that iPGE2-induced HRE activation and VEGF-A sults (Fig. 3b) indicated that BG and diclofenac prevented 16,16-PGE2- production were dependent on EGFR and RARβ, we studied in 16,16- and ATRA-induced EGFR phosphorylation (although the prevention PGE2-treated cells the effect of inhibition of EGFR transactivation with by BG was not 100% effective). When cells were preincubated with an- AG1478 or inhibition of RARβ with RARβ siRNA or RARβ antagonist tagonist of EP1-EP3 receptors AH6809 or with EP4 receptor antagonist LE135. Fig. 4dconfirms the dependency of iPGE2-induced HRE activa- GW627368X, 16,16-PGE2- and ATRA-induced EGFR phosphorylation tion and VEGF-A production on the activity of EGFR and RARβ (siRNA were also prevented (GW627368X did not fully prevent ATRA- RARβ did not fully prevent 16,16-PGE2-induced VEGF-A production, induced EGFR phosphorylation, though). These results indicate that which was probably due to the fact that RARβ expression was only iPGE2-dependent activation of EPR is involved in the mechanism of reduced by 67% instead of being fully suppressed). 16,16-PGE2- and ATRA-induced EGFR transactivation. In summary, the results presented in Fig. 4 depict a scenario in To confirm the involvement of transactivation of EGFR in the which iPGE2, acting through EGFR and RARβ, up-regulates HIF-1α iPGE2-dependent increase of RARβ expression, we examined the ef- mRNA. This effect eventually leads to HIF-1α-dependent HRE activation fect in HK-2 cells of the inhibitor of EGFR activation AG1478 (1 μM, and further increase in production of renoprotective factor VEGF-A in

1 h pre-incubation) on ATRA- or 16,16-PGE2-induced increase in HK-2 cells. RARE-Luc activity (in cells previously transfected with reporter luciferase plasmid tk-β-RARE-Luc) and in RARβ mRNA and protein 4. Discussion expression. Our results (Fig. 3c) indicated that the inhibitor of EGFR activation AG1478 prevented the whole sequence of events involved We have previously found in HK-2 cells and other cell lines that in ATRA- or 16,16-PGE2-induced increase in RARβ expression (and intracrine PGE2-EPR signaling plays a critical role in the up-regulation therefore the RARβ-dependent increase in HIF-1α expression) in of RARβ after treatment with PGE2 or ATRA [11,12]. Enhanced RARβ ex- HK-2 cells. Taken together, the results shown in Fig 3a–c indicate pression, in turn, leads to an increase in both the expression of HIF-1α that transactivation of EGFR by iPGE2-activated EPR mediates the and the production of HIF-1α-regulated renoprotective factor VEGF-A increase in RARβ expression in HK-2 cells. [11,12]. In the present work we have studied the role of EGFR and tran- Since the ﬁrst step in the transcriptional up-regulation of RARβ by scriptional mechanisms in the regulation of VEGF-A production by

16,16-PGE2 and ATRA involves activation of RARE by iPGE2-activated iPGE2. Our results indicated that transactivation of EGFR by iPGE2- EPR (Figs. 1 and 2), we also analyzed the effect of AG1478 in other cell activated EPR (presumably an EPR subset located intracellularly [11]) lines which were transfected with plasmid tk-β-RARE-Luc. As shown plays a critical role because it mediates the increase in RARβ expression in Fig. 3d induction of RARE-Luc by 16,16-PGE2 and ATRA in PC3 cells induced by PGE2 or ATRA. The mechanism involved RARβ mRNA and A498 cells was prevented by AG1478, which suggested that up-regulation, most likely through activation of the RARE in the RARβ transactivation of EGFR might be universally required for the activa- promoter (Fig. 5). Finally, the increase in VEGF-A production was tion of RARE leading to increased expression of RARβ. found to be due to RARβ-dependent transcriptional up-regulation of HIF-1α, which most likely resulted in activation of the HRE in the 3.4. Activation of EGFR and RARβ, leading to transcriptional VEGF-A promoter (Fig. 5). up-regulation of HIF-1α, is responsible for iPGE2-induced To the best of our knowledge this is the first time ever in which production of renoprotective factor VEGF-A. iPGE2 has been shown to enhance RARβ mRNA expression. EGFR/ EPR-dependent RARE activation upon treatment with 16,16-PGE2 or Taking into account the results shown in Figs. 1 and 3, we hypoth- ATRA was also demonstrated in other cell lines. Therefore, trans- esized that iPGE2 enhances the production of renoprotective factor activation of EGFR by iPGE2-activated EPR is probably an integral part VEGF-A through an EGFR-, RARβ-dependent transcriptional increase (though a newly described one) of the physiological pathway for in HIF-1α expression, leading to activation of the HRE in the VEGF-A ATRA-induced RARE activation in the RARβ promoter leading to RARβ gene promoter. In order to test this hypothesis, we first analyzed mRNA up-regulation. There are two relevant questions regarding the the effect of 16,16-PGE2 on the expression of HIF-1α mRNA and studies on RARE activity. The first one is why treatment with EP1/EP2/ found that it increased by 69.1 ± 9.1% after 4 h incubation (Fig. 4a, EP3 receptor antagonist AH6809 or with EP4 receptor antagonist inset). We next sought to confirm that the transcriptional increase GW627368X produced a complete inhibition of the activation of the in the expression of HIF-1α was dependent on iPGE2 and RARβ.To RARE-Luc construct from the RARβ gene promoter. Because, if all EPR this aim, HK-2 cells were incubated with BG or LE135 (RARβ antago- were implicated, one could expect that after AH6809 treatment there nist) prior to incubation with 16,16-PGE2. Our results (Fig. 4a) should be still response through EP4. Moreover, it could be argued 2036 A.B. Fernández-Martínez, F.J. Lucio Cazaña / Biochimica et Biophysica Acta 1833 (2013) 2029–2038

Fig. 4. Activation of EGFR and RARβ, leading to transcriptional up-regulation of HIF-1α, is responsible for iPGE2-induced production of renoprotective factor VEGF. a)16,16-PGE2- induced increase in HIF-1a mRNA expression is prevented by inhibition of prostaglandin uptake or antagonism of RARβ. HK-2 cells were preincubated with bromocresol green or

LE135, and incubated with 1 μM 16,16-PGE2 for 4 h (or for 0–4 h in the inset). b) 16,16-PGE2- and ATRA-induced increase in HIF-1α mRNA expression are prevented by AG1478.

HK-2 cells were incubated for 1 h with AG1478 and then with 16,16 PGE2 or ATRA for 4 h. c) Inhibition of prostaglandin uptake and HIF-1α prevent 16,16-PGE2-induced increase in HRE-Luc activity and VEGF-A production. Left panel: HK-2 cells were transfected with p9HIF1-Luc, preincubated for 1 h with 50 μM bromocresol green or 10 μM YC1 and then, incubated with 16,16-PGE2 for 4 h. Right panel: Cells were transfected with HIF-1α siRNA, or control siRNA (scramble) or they were pretreated with bromocresol green. Thereafter cells were incubated with/without 16,16-PGE2 for 48 h and the released VEGF-A was measured by ELISA. d) 16,16-PGE2-induced increase in HRE-Luc activity and VEGF-A production are prevented by inhibitors of EGFR phosphorylation and RARβ activity. Left panel: Cells were transfected with p9HIF1-Luc, preincubated for 1 h with 1 μM AG1478 or 2 μM

LE135 and then, incubated with 1 μM 16,16-PGE2 for 4 h. Right panel: Cells were transfected with RARβ siRNA, or control siRNA (scramble) or they were pretreated for 1 h with

1 μM AG1478 or 2 μM LE135. Thereafter cells were incubated with or without 16,16-PGE2 for 48 h. General information: 1) Reporter gene experiments: Luciferase activity was normalized by Renilla luciferase activity and expressed as relative luminescence units (RLU). 2) Bars and error bars in graphs: Each bar represents the mean ± SD of 3 different experiments.*Pb 0.05 vs. control. that after GW627368X treatment, response through EP1-3 receptors activation but also for basal RARE activity. The fact that the pre- should be detected. However, we do not think that all EPRs are implicat- incubation with EGFR inhibitor AG1478 does not reduce the activity of ed in the activation of the RARE. We rather think that RARE activation RARE in PC3 cells treated with ATRA or PGE2 to the level found in cells needs the simultaneous stimulus from at least two EPRs: one of them, which were only treated with AG1478 (comparing column 4 with 5 of course, is EP4 and the other is EP1, EP2 and/or EP3. This would ex- and 6), suggests that iPGE2 also controls the activity of RARE in PC3 plain why treatment with AH6809 or GW627368X results in the com- cells through EGFR-independent mechanisms. Further experiments plete inhibition of RARE activation. The need of simultaneous stimulus should be done to test this hypothesis. from at least two EPRs to get a biological response is not unusual. For As indicated above, 16,16-PGE2 itself is able to increase the lucifer- instance, it has been previously shown that treatment with AH6809 or ase activity of the RAREβ and RARβ promoter constructs in 3 different

AH23848 (an EP4 receptor antagonist) suppress PGE2-induced phos- cell lines without any concurrent treatment with ATRA. This is a strik- phorylation of Akt in HUVEC cells [26]. We will focus now on the second ing ﬁnding because activation of RARE requires the binding of ATRA- question, which arises from the fact that on PC3 cells AG1478 lowers activated RAR-RXR heterodimers [7]. Data from cultured keratinocytes RARE activity well below the control value. This suggests that in PC3 may help to explain these results. These cells maintain steady state cells, EGFR activity is responsible not only for iPGE2-induced RARE intracellular ATRA concentrations in the 20–50 nM range [27] when A.B. Fernández-Martínez, F.J. Lucio Cazaña / Biochimica et Biophysica Acta 1833 (2013) 2029–2038 2037

cells could help to identify this pathway, if we assume that their intracellular ATRA content is similar to that of cultured keratinocytes (there- by allowing the activation of RAR-RXR). In these conditions, the fact that

16,16-PGE2 treatment increases the activity of RAREβ and RARβ promoter constructs to the same extent that ATRA treatment, suggests

that full activation of RAR-RXR requires iPGE2-induced EGFR transactivation. Further studies are needed to conﬁrm this hypothesis.

Regarding the transactivation of EGFR by iPGE2, in recent years it has been reported transactivation of plasma membrane EGFR by PGE2- activated EPR [29].Theseﬁndings are just a particular case of activation of EGFR by plasma membrane G-protein coupled receptors (GPCR) [30–32]. Our results suggest that transactivation of EGFR is not due to

plasma membrane EPR activated by extracellular PGE2 but to iPGE2- activated intracellular EPR. However, intracellular EPR might be also GPCR. In fact, it has been proposed that the distinct localization of EPR and other GPCR may dictate different signaling pathways for the same receptor [33]. On the other hand, two particular results involving partial prevention of EGFR transactivation in our experimental setting deserve a more detailed analysis. In the ﬁrst place, EP4 receptor antagonist GW627368X did not fully prevent ATRA-induced EGFR phosphorylation

(while it abolished 16,16-PGE2-induced EGFR phosphorylation, as shown in Fig. 3b). This suggests that ATRA transactivates EGFR through additional mechanisms to the ones mediated by EPR. The other partial prevention of EGFR phosphorylation was found in cells which were treated with the inhibitor of prostaglandin uptake BG prior to treatment

with 16,16-PGE2 (Fig. 3b). This result should be conciliated with the fact that the residual activation of EGFR was not able to activate the RARE in the RARβ promoter (and, therefore, to trigger RARβ-dependent HRE activation leading to HIF-1α-dependent increase in VEGF-A production). Our hypothesis to conciliate these results relies on the facts that the expression of phosphorylated EGFR (Fig. 3a) was ubiquitous in HK-2 cells (instead of being restricted to the plasma cell membrane)

and it increased upon treatment with 16,16-PGE2.Weproposethat treatment with 16,16-PGE2 transactivates EGFR located at the cell membrane as well as EGFR located inside the cell, but only transactivation of intracellular EGFR results in RARE activation. In this way, the residual activation of EGFR after treatment with BG would be due to transactivation of EGFR located at the cell membrane and, therefore, it would not result in RARE activation. However, speciﬁc experiments should verify this hypothesis. Whatever the mechanisms involved in the transactivation of EGFR

by ATRA and PGE2, the resulting increase in EGFR kinase activity appears to be critical for RARβ up-regulation since ATRA- or 16,16-

PGE2-induced increase in RARβ expression was prevented by inhibition of EGFR kinase activity with AG1478 (Fig. 3). Regarding the pathway downstream EGFR, histone H3 is one possible target. The phosphorylation of histone H3 is functionally liked to the activation β Fig. 5. EGFR transactivation by iPGE2-activated EPR results in RARE activation in the RARβ of RAR promoter by retinoids, so that it has been proposed as a mo- β promoter leading to increased RAR expression. Role in PGE2-andATRA-induced lecular signature of this activation [34]. Since EGFR phosphorylates α transcriptional up-regulation of HIF-1 and production of renoprotective factor VEGF-A and activates a number of important signaling molecules including ki- (model of a proposed pathway). nases [29], it is conceivable that the phosphorylation of histone H3

may result from the transactivation of EGFR by iPGE2-activated EPR. It is important to conﬁrm this hypothesis in further studies because grown in medium containing 25 nM retinol (originating from 5% fetal up-regulation of RAR genes by ATRA presumably plays a critical role bovine serum) and no detectable ATRA (less than nM). ATRA was syn- in amplifying the genomic response to ATRA [10]. thesized from substrates derived from intracellular ester stores of reti- Current data provide evidence for the regulation of VEGF expression nol [27]. On the other hand, treatment with ATRA determines a by EGFR via the MAPK and PI3K signaling cascades and at least three dif- dose-dependent increase in the activity of a RARE construct in cultured ferent transcription factors, signal transducer and activator of transcrip- keratinocytes and reaches its maximum (ﬁve-fold over control values) tion 3 (STAT3), Sp1 and HIF [35]. Our results here are in good agreement at 10 μM ATRA [28]. These results are not in good agreement with the with this evidence since they indicate that the activation of EGFR by fact that ATRA binds RARs with Ki b20 nM [8]. It indicates that activa- 16,16-PGE2 or ATRA (Fig. 3) results in AG1478-sensitive increased ex- tion of RAR-RXR by endogenous ATRA is suboptimal and that full activa- pression of HIF-1α/VEGF (Fig. 4). HIF-1α protein accumulates during tion of RAR-RXR requires intracellular ATRA concentrations well above hypoxia through a very well know transcription-independent pathway. those needed to bind RARs. This suggests that, at least in cultured It involves the stabilization of HIF-1α protein through inhibition of keratinocytes, treatment with ATRA may activate an unknown signaling proline hydroxylation at Pro402 and Pro564, which prevents pVHL- pathway to reach the full activation of RAR-RXR. Our results in HK-2 dependent HIF-1α ubiquitination leading to proteasomal degradation 2038 A.B. Fernández-Martínez, F.J. Lucio Cazaña / Biochimica et Biophysica Acta 1833 (2013) 2029–2038

[36].However,PGE - and ATRA-induced increase in HIF-1α expression [14] Y. Sugimoto, S. Narumiya, Prostaglandin E receptors, J. Biol. Chem. 282 (2007) 2 – α 11613 11617. under normoxia is not due to the increase HIF-1 protein stability [15] K. Tsuboi, Y. Sugimoto, A. Ichikawa, Prostanoid receptor subtypes. Prostaglandins [11,13]. Since the effect of PGE2 and ATRA on HIF-1α expression is other lipid, Mediat. 68–69 (2002) 535–556. prevented by actinomycin D, it is more likely that transcriptional mech- [16] B.S. Chan, J.A. Satriano, M. Pucci, V.L. Schuster, Mechanism of prostaglandin E2 α transport across the plasma membrane of HeLa cells and Xenopus oocytes expressing anisms are involved in the iPGE2-dependent up-regulation of HIF-1 the prostaglandin transporter “PGT”, J. Biol. Chem. 273 (1998) 6689–6697. induced by these agents [11,13]. This suggestion is reinforced by the [17] V.L. Schuster, Prostaglandin transport. Prostaglandins other lipid, Mediat. 68–69 – fact that treatment with PGE2 resulted in the up-regulation of HIF-1α (2002) 633 647. mRNA in a BG-sensitive manner (Fig. 4a). Because the RARβ inhibitor [18] M. Bhattacharya, K.G. Peri, G. Almazan, A. Ribeiro-da-Silva, H. Shichi, Y. Durocher, M. Abramovitz, X. Hou, D.R. Varma, S. Chemtob, Nuclear localization of prosta- LE135 also prevented PGE2-induced increase in HIF-1α mRNA, the re- glandin E2 receptors, Proc. Natl. Acad. Sci. U.S.A. 95 (1998) 15792–15797. sults shown in Fig. 4a are in good agreement with the view that iPGE2 [19] M. Bhattacharya, K. Peri, A. Ribeiro-da-Silva, G. Almazan, H. Shichi, X. Hou, D.R. enhances HIF-1α expression through RARβ-dependent transcriptional Varma, S. Chemtob, Localization of functional prostaglandin E2 receptors EP3 and EP4 in the nuclear envelope, J. Biol. Chem. 274 (1999) 15719–15724. mechanisms. [20] F. Jr Gobeil, I. Dumont, A.M. Marrache, A. Vazquez-Tello, S.G. Bernier, D. Abran, X. In the final stage of the pathway studied by us, iPGE2-induced Hou, M.H. Beauchamp, C. Quiniou, A. Bouayad, S. Choufani, M. Bhattacharya, S. HIF-1α up-regulation leads to increased production of VEGF-A, presum- Molotchnikoff, A. Ribeiro-Da-Silva, D.R. Varma, G. Bkaily, S. Chemtob, Regulation of eNOS expression in brain endothelial cells by perinuclear EP(3) receptors, Circ. ably through activation of the HRE in its gene promoter (Figs. 4c, d, and Res. 90 (2002) 682–689. 5). VEGF-A has protective effects in human renal proximal tubular cells [21] M. Mendez, M.C. LaPointe, PGE2-induced hypertrophy of cardiac myocytes involves [37], in chronic renal disease [38] and in experimental models of renal EP4 receptor-dependent activation of p42/44 MAPK and EGFR transactivation, Am. J. – – Physiol. 288 (2005) H2111 H2117. disease [39 41]. The regulation of ATRA- and 16,16-PGE2-induced [22] K.J. Livak, T.D. Schmittgen, Analysis of relative gene expression data using VEGF-A production by iPGE2-dependent EGFR transactivation opens real-time quantitative PCR and the 2(-Delta Delta C(T)) Method, Methods 25 the way to new pharmacological approaches for the therapeutic modu- (2001) 402–408. [23] N.N. Orie, L.H. Clapp, Role of prostanoid IP and EP receptors in mediating lation of HIF-1α/VEGF-A. vasorelaxant responses to PGI2 analogues in rat tail artery: evidence for Gi/o modulation via EP3 receptors, Eur. J. Pharmacol. 654 (2011) 258–265. [24] R.J. Wilson, G.M. Giblin, S. Roomans, S.A. Rhodes, K.A. Cartwright, V.J. Shield, J. Acknowledgements Brown, A. Wise, J. Chowdhury, S. Pritchard, J. Coote, L.S. Noel, T. Kenakin, C.L. Burns-Kurtis, V. Morrison, D.W. Gray, H. Giles, GW627368X ((N-{2-[4-(4, 9-diethoxy- 1-oxo-1, 3-dihydro-2H-benzo[f]isoindol-2-yl)phenyl]acetyl} benzenesulphonamide): This work was supported by grants SAF2008-01767 and SAF2011- a novel, potent and selective prostanoid EP4 receptor antagonist, Br. J. Pharmacol. 26838 from the Spanish Ministerio de Ciencia e Innovación, S2010/ 148 (2006) 326–339. BMD-2378 from the Comunidad de Madrid-programa de Actividades [25] J.J. Lou, Y.L. Chua, E.H. Chew, J. Gao, M. Bushell, T. Hagen, Inhibition of hypoxia- inducible factor-1alpha (HIF-1alpha) protein synthesis by DNA damage inducing I+D en Biociencias 2010 and POII10-0034-0322 from the Junta de agents, PLoS One (2010) e10522, (doi:10.13715). Comunidades de Castilla-La Mancha. Ana Belén Fernández Martinez is [26] S. Jain, G. Chakraborty, R. Raja, S. Kale, G.C. Kundu, Prostaglandin E2 regulates the recipient of a post-doctoral fellowship from the Junta de Comunidades tumor angiogénesis in prostate cancer, Cancer Res. 68 (2008) 7750–7759. de Castilla-La Mancha. [27] R.K. Randolph, M. Simon, Characterization of retinol metabolism in cultured human epidermal keratinocytes, J. Biol. Chem. 268 (1993) 9198–9205. [28] B.J. Aneskievich, E. Fuchs, Terminal differentiation in keratinocytes involves positive as well as negative regulation by retinoic acid receptors and retinoid X References receptors at retinoid response elements, Mol. Cell. Biol. 12 (1992) 4862–4871. [29] C. Liebmann, EGFR receptor activation by GPCRs: an universal pathway reveals [1] D. Lando, D.J. Peet, D.A. Whelan, J.J. Gorman, M.L. Whitelaw, Asparagine hydrox- different versions, Mol. Cell. Endocrinol. 331 (2011) 22–231. ylation of the HIF transactivation domain: a hypoxic switch, Science 295 (2002) [30] O.M. Fischer, S. Hart, A. Gschwind, A. Ullrich, EGFR signal transactivation in cancer 858–861. cells, Biochem. Soc. Trans. 31 (2003) 1203–1208. [2] S. Kaluz, M. Kaluzova, E.J. Stanbridge, Regulation of gene expression by hypoxia: [31] R. Wetzker, F. Bohmer, D Transactivation joins multiple tracks to the ERK/MAPK integration of the HIF-transduced hypoxic signal at the hypoxia-responsive element, cascade, Nat. Rev. Mol. Cell Biol. 4 (2003) 651–657. Clin. Chim. Acta 395 (2008) 6–13. [32] A. Gschwind, E. Zwick, N. Prenzel, M. Leserer, A. Ullrich, Cell communication [3] G. Semenza, Signal transduction to hypoxia-inducible factor 1, Biochem. networks: epidermal growth factor receptor transactivation as the paradigm for Pharmacol. 64 (2002) 993–998. interreceptor signal transmission, Oncogene 20 (2001) 1594–1600. [4] J. Zhou, B. Brüne, Cytokines and hormones in the regulation of hypoxia inducible [33] T. Zhu, F. Gobeil, A. Vazquez-Tello, M. Leduc, L. Rihakova, M. Bossolasco, G. Bkaily, factor-1alpha (HIF-1alpha), Cardiovasc. Hematol. Agents Med. Chem. 4 (2006) K. Peri, D.R. Varma, R. Orvoine, S. Chemtob, Intracrine signaling through lipid me- 189–197. diators and their cognate nuclear G-protein-coupled receptors: a paradigm based [5] Y.J. Jung, J.S. Isaacs, S. Lee, J. Trepel, L. Neckers, IL-1beta-mediated up-regulation of on PGE2, PAF, and LPA1 receptors, Can. J. Physiol. Pharmacol. 84 (2006) 377–391. HIF-1alpha via an NFkappaB/COX-2 pathway identifies HIF-1 as a critical link [34] B. Lefebvre, K. Ozato, P. Lefebvre, Phosphorylation of histone H3 is functionally between inflammation and oncogenesis, FASEB J. 17 (2003) 2115–2117. linked to retinoic acid receptor beta promoter activation, EMBO Rep. 3 (2002) [6] S. Galbán, M. Gorospe, Factors interacting with HIF-1alpha mRNA: novel thera- 335–340. peutic targets, Curr. Pharm. Des. 15 (2009) 3853–3860. [35] A. Larsen, D. Ouaret, K. El Oudrani, A. Petitprez, Targeting EGFR and VEGF(R) [7] J.E. Balmer, R. Blomhoff, Gene expression regulation by retinoic acid, J. Lipid Res. pathway cross-talk in tumor survival and angiogenesis, Pharmacol. Ther. 131 43 (2002) 1773–1808. (2011) 80–90. [8] S.M. Thacher, J. Vasudevan, R.A.S. Chandraratna, Therapeutic applications for [36] J.I. Bardos, M. Ashcroft, Negative and positive regulation of HIF-1: a complex ligands of retinoid receptors, Curr. Pharm. Des. 6 (2000) 25–58. network, Biochim. Biophys. Acta 1755 (2005) 107–120. [9] D.J. Mangelsdorf, K. Umesono, R.M. Evans, The retinoid receptors, in: M.B. Sporn, [37] J. Kanellis, S. Fraser, M. Katerelos, D.A. Power, Vascular endothelial growth factor is a A.B. Roberts, D.S. Goodman (Eds.), The Retinoids: Biology, Raven Press, Chemistry survival factor for renal tubular epithelial cells, Am. J. Physiol. 278 (2000) F905–F915. and Medicine, New York, 1994, pp. 319–349. [38] Y. Kitamoto, K. Matsuo, K. Tomita, Different response of urinary excretion of VEGF [10] B. Hoffmann, J.M. Lehmann, X.K. Zhang, T. Hermann, M. Husmann, G. Graupner, M.A. in patients with chronic and acute renal failure, Kidney Int. 59 (2001) 385–386. Pfahl, Retinoic acid receptor-specific element controls the retinoic acid receptor-β [39] Y.G. Kim, S.I. Suga, D.H. Kang, J.A. Jefferson, M. Mazzali, K.L. Gordon, K. Matsui, S. promoter, Mol. Endocrinol. 4 (1990) 1727–1736. Breiteneder-Geleff, S.J. Shankland, J. Hughes, D. Kerjaschki, G.F. Schreiner, R.J. [11] A.B. Fernández-Martínez, M.I. Arenas, F.J. Lucio, Retinoic acid increases Johnson, Vascular endothelial growth factor accelerates renal recovery in experi- hypoxia-inducible factor-1α through intracrine prostaglandin E(2) signaling in mental thrombotic microangiopathy, Kidney Int. 58 (2000) 2390–2399. human renal proximal tubular cells HK-2, Biochim. Biophys. Acta 182 (2012) [40] D.H. Kang, J. Hughes, M. Mazzali, G.F. Schreiner, R.J. Johnson, Impaired angiogen- 672–683. esis in the remnant kidney model: II. Vascular endothelial growth factor admin- [12] A.B. Fernández-Martínez, M.I. Arenas, V. Moreno, F.J. Lucio, Intracrine prostaglandin istration reduces renal fibrosis and stabilizes renal function, J. Am. Soc. Nephrol. E(2) signalling regulates hypoxia-inducible factor-1α expression through retinoic 12 (2001) 1448–1457. acid receptor-β, Int. J. Biochem. Cell Biol. 44 (2012) 2185–2193. [41] Y. Masuda, A. Shimizu, T. Mori, T. Ishiwata, H. Kitamura, R. Ohashi, M. Ishizaki, G. [13] A.B. Fernández-Martínez, M.I. Arenas, I. Sánchez, M.L. Garcia, V. Moreno, E. Asano, Y. Sugisaki, N. Yamanaka, Vascular endothelial growth factor enhances Aguado, F.J. Lucio, Mutual regulation of hypoxic and retinoic acid related signalling glomerular capillary repair and accelerates resolution of experimentally induced in tubular proximal cells, Int. J. Biochem. Cell Biol. 43 (2011) 1198–1207. glomerulonephritis, Am. J. Pathol. 159 (2001) 599–608.