Published OnlineFirst April 20, 2012; DOI: 10.1158/1541-7786.MCR-11-0507

Molecular Cancer Angiogenesis, Metastasis, and the Cellular Microenvironment Research

TGF-b1 Induces Endothelial Cell by Shifting VEGF Activation of p38MAPK from the Prosurvival p38b to Proapoptotic p38a

Giovanni Ferrari1, Vitaly Terushkin2, Martin J. Wolff2, Xiaodong Zhang2, Cristina Valacca2, Paolo Poggio1, Giuseppe Pintucci2, and Paolo Mignatti2,3

Abstract TGF-b1 and VEGF, both angiogenesis inducers, have opposing effects on vascular endothelial cells. TGF-b1 induces apoptosis; VEGF induces survival. We have previously shown that TGF-b1 induces endothelial cell expression of VEGF, which mediates TGF-b1 induction of apoptosis through activation of p38 mitogen-activated protein kinase (MAPK). Because VEGF activates p38MAPK but protects the cells from apoptosis, this finding suggested that TGF-b1 converts p38MAPK signaling from prosurvival to proapoptotic. Four isoforms of p38MAPK —a, b, g, and d—have been identified. Therefore, we hypothesized that different p38MAPK isoforms control endothelial cell apoptosis or survival, and that TGF-b1 directs VEGF activation of p38MAPK from a prosurvival to a proapoptotic isoform. Here, we report that cultured endothelial cells express p38a, b, and g. VEGF activates p38b, whereas TGF-b1 activates p38a. TGF-b1 treatment rapidly induces p38a activation and apoptosis. Subsequently, p38a activation is downregulated, p38b is activated, and the surviving cells become refractory to TGF-b1 induction of apoptosis and proliferate. silencing of p38a blocks TGF-b1 induction of apoptosis, whereas downregulation of p38b or p38g expression results in massive apoptosis. Thus, in endothelial cells p38a mediates apoptotic signaling, whereas p38b and p38g transduce survival signaling. TGF-b1 activation of p38a is mediated by VEGF, which in the absence of TGF-b1 activates p38b. Therefore, these results show that TGF-b1 induces endothelial cell apoptosis by shifting VEGF signaling from the prosurvival p38b to the proapoptotic p38a. Mol Cancer Res; 10(5); 605–14. 2012 AACR.

Introduction and vascular permeability (1, 4). Protein kinase B (Akt) and Angiogenesis, the formation of capillaries from preexisting mitogen-activated protein kinases (MAPK) are components blood vessel, is mediated by a variety of and growth of the signaling mechanism activated by VEGFR-2 (5). factors with paracrine or autocrine modes of action. VEGF TGF-b1, the prototype member of a superfamily of 5 and TGF-b1, potent angiogenesis inducers, act on vascular multifunctional growth factors, is a potent proliferation endothelial cells through different mechanisms (1–3). inhibitor for most cell types, and an important regulator of tissue morphogenesis (3, 6). TGF-b1 induces vessel for- VEGF (VEGF A), the prototype member of a family of 4 – growth factors (VEGF A–D), upregulates endothelial cell mation in vitro and in vivo (7 11); however, it inhibits proliferation and migration and protects endothelial cells endothelial and migration (12) and down- from apoptosis (1). VEGF exerts its activity through 2 regulates VEGFR-2 expression (13, 14). Notably, TGF-b1 tyrosine kinase receptors, VEGFR-1 (flt-1) and VEGFR-2 induces endothelial cell apoptosis (12, 15) by inhibiting fl expression of the antiapoptotic protein Bcl-2 (16) and ( k-1). VEGFR-2 has been implicated in endothelial cell MAPK proliferation and survival, and VEGFR-1 in activating p38 MAPK (p38 ;ref.17). Thus, although both VEGF and TGF-b1 induce angio- genesis, they have opposing effects on endothelial cells. Authors' Affiliations: 1Division of Cardiovascular Surgery, Department of Remarkably, TGF-b1 is a potent inducer of apoptosis, Surgery, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania; Departments of 2Cardiothoracic Surgery and 3Cell Biology, whereas VEGF protects the cells from apoptosis. It has New York University School of Medicine, New York, New York therefore been proposed that TGF-b1 induces angiogenesis fl Note: Supplementary data for this article are available at Molecular Cancer in vivo through an indirect mechanism, by recruiting in am- Research Online (http://mcr.aacrjournals.org/). matory cells that in turn secrete VEGF and/or other angio- Corresponding Author: Paolo Mignatti, New York University School of genesis inducers. However, TGF-b1 induces endothelial cell Medicine, 550 First Avenue, NBV16W15, New York, NY 10016. Phone: 212- expression of VEGF in vitro and in vivo (18–20) and, 263-1478; Fax: 212-263-3161; E-mail: [email protected] surprisingly, VEGF is required for induction of endothelial doi: 10.1158/1541-7786.MCR-11-0507 cell apoptosis through VEGFR-2 activation of p38MAPK 2012 American Association for Cancer Research. (19, 20).

www.aacrjournals.org 605

Downloaded from mcr.aacrjournals.org on September 29, 2021. © 2012 American Association for Cancer Research. Published OnlineFirst April 20, 2012; DOI: 10.1158/1541-7786.MCR-11-0507

Ferrari et al.

p38MAPK, a class of MAPK activated by environmental considered as time 0. Western blotting was carried out as stresses, growth factors and cytokines, controls cell functions described (19). Bromodeoxyuridine (BrdU) uptake was including proliferation and apoptosis, differentiation, and carried out as described (19). senescence (21, 22). Four isoforms of p38MAPK —a, b, g, Immunoprecipitation. Cells were lysed in 10 mmol/L and d—have been identified in mammalian cells. Despite Tris-HCl pH 7.4 containing 150 mmol/L NaCl, 1% Triton more than 60% and more than 90% X-100, 1 mmol/L Pefabloc (Roche), 1 mmol/L leupeptin, 1 MAPK identity within their kinase domains, the p38 isoforms mmol/L Na3VO4, and 2 mmol/L CaCl2. (plus 100 mmol/L show notable differences in tissue expression, upstream peroxyvanadate for phosphorylation studies). One hundred activators, and downstream effectors. p38a and p38b are micrograms of cell extract protein was precleared at 4C for ubiquitous; p38g and p38d are tissue specific (22). In 30 minutes with 0.5 mg of nonimmune IgG coupled to 10 addition, the p38 isoforms have been described in different mL of protein AþG agarose beads (Santa Cruz Biotechnol- cell compartments, in which they can have opposing effects ogy). Precleared extracts were centrifuged for 60 seconds at on the same substrate (21). The specific function(s) of the 300 g, and the supernatant was immunoprecipitated individual isoforms in physiology and pathology are largely overnight at 4C with 10 mL of protein AþG agarose beads unknown. The genetic deficiency of p38a in mice results in and 0.75 mg of per 100 mg of protein. After washing embryonic lethality, with aberrant placental development 3 times with lysis buffer, the beads were boiled in reducing and abnormal angiogenesis in the yolk sac and embryo (23). Laemmli buffer for 5 minutes and loaded onto SDS/PAGE In contrast, disruption of the other isoforms generates no gels. apparent phenotype (21). Numerous studies have implicat- siRNA transfection. Subconfluent HUVE cells were ed p38MAPK in induction of endothelial cell apoptosis by a transiently transfected as described (19) and used for the variety of agents (24–27). However, the p38 isoforms experiments 48 hours after transfection. involved have not been characterized. Reverse transcription PCR. The following primers Here, we report that in vascular endothelial cells, p38a were synthesized by IDT DNA technologies based on the mediates proapoptotic signaling from inducers of apoptosis published sequences: glyceraldehyde-3-phosphate dehydro- such as TGF-b1, whereas p38b relays survival signaling from genase: (forward 50-CCC ACT CTT CCA CCT TCG-30; prosurvival factors including VEGF, and that TGF-b1 reverse 5-TCC TTG GAG GCC ATG TAG GCC AT-30); induces endothelial cell apoptosis by converting VEGF p38a: (forward 50-GCA GGG ACC TTC TCA TAG AT- signaling from the prosurvival p38b to the proapoptotic 30; reverse 50-GAG GGA TAG CCT CAG ACC-30); p38b: p38a. (forward 50-CTG CAA GGA AAG GCC CTC-30; reverse 50-CAG GCA ATG CCT CAC TGC-30); p38g: (forward 50-GAT TAC TGG GAA GAT CCT G-30; reverse 50-CGT Materials and Methods CAC AGA GCC GTC TCC-30); p38d: (forward: 50-GAC Materials ACT CTT CAA GGG CAA G-30; reverse 50-GCC ATC Human purified or recombinant TGF-b1, recombinant AAT CAC TGC AGC-30). cDNA was synthesized from 1 human VEGF, and to p38a, p38g, and p38d mg of total RNA with SuperScript II RT (Invitrogen) and were purchased from R&D Systems; antibodies to human oligo-dT 30 primer. cDNA (2 mL) was amplified by PCR as cleaved caspase 3 and cleaved PARP from described. Technologies; mouse and rabbit nonimmune immunoglob- Proximity ligation assay. HUVE cells grown on gela- ulin G (IgG) from Sigma-Aldrich; and antibodies to p38, tin-coated glass coverslips were treated with TGF-b1 (1 ng/ a-tubulin, and cyclin E from Santa Cruz Biotechnology, mL) or control medium for 6 or 72 hours and fixed. The cells Inc.. Human recombinant fibroblast -2 (FGF- were then incubated with isoform-specific antibodies to 2) was purchased from Gibco BRL; antibody to active p38 p38a or p38b together with antibody to phospho-p38 from Promega Corp.; monoclonal antibody to p38b from overnight at 4C with gentle agitation. PLUS and MINUS Zymed Laboratories; recombinant p38a from R&D Sys- secondary Proximity ligation assay (PLA; ref. 29) probes tems; p38 b from Biovision; and p38g from Abnova. against rabbit and mouse IgG (Olink Bioscience) were added, and the cells were incubated at 37C for 1 hour Cells and media with gentle agitation, followed by incubation with ligation Bovine capillary endothelial cells (BCE) were as described mix for 30 minutes at 37C. Amplification mix was then (28) and used at passages 6 to 15. Human umbilical vein applied for 100 minutes at 37C. The coverslips were endothelial (HUVE; Clonetics) cells were grown in medium mounted on microscope slides with Doulink Mounting (EBM2; Clonetics) containing 2% fetal calf serum (FCS) Medium with DAPI, and the cells photographed under a and the growth supplements provided by the company, and fluorescence microscope. were used at passages 3 to 5. BCE or HUVE cells were In vitro angiogenesis assay in 3D collagen gel. Colla- starved overnight in their medium supplemented with 0.5% gen type I (Becton Dickinson) was placed into 24-well tissue donor calf serum (DCS) or FCS, respectively, after which culture plates (40 mL per well) and allowed to gel for 30 either TGF-b1 (1 ng/mL) or VEGF (30 ng/mL) or FGF-2 minutes at 37C. HUVE cells (2.5 104 cells per well) were (10 ng/mL) was added and incubation was continued for the seeded into the wells and incubated at 37C for 1 hour in indicated time. The time of addition of TGF-b1 was growth medium. Subsequently, the medium was removed

606 Mol Cancer Res; 10(5) May 2012 Molecular Cancer Research

Downloaded from mcr.aacrjournals.org on September 29, 2021. © 2012 American Association for Cancer Research. Published OnlineFirst April 20, 2012; DOI: 10.1158/1541-7786.MCR-11-0507

Control of Endothelial Cell Apoptosis by p38MAPK Isoforms

and 40 mL per well of collagen was added. After incubation at 37C for 30 minutes, 100 mL of medium supplemented with A B HUVEC 0.5% FCS and 1 ng/mL of TGF-b1 was added, and the BCE MCF-7 HUVEC plates were incubated at 37 C (30). TGF-b1 (1 ng/mL) was p38α added every other day. After 5 to 7 days, the culture medium p38β p38 α p38 β p38 γ was removed; the cultures were washed with PBS, stained p38 δ with toluidine blue, and photographed under an inverted p38γ microscope. The number of tube-like structures forming p38δ anastomoses was counted as described (20). C D 15 min 6 h Statistical analysis The Student t tests on the equality of means were carried CONTROL FGF-2 VEGF TGF- β 1 TGF- β 1 VEGF r p38 γ r p38 α out using Stata 8. r p38 β Coomassie p38α p38α p38β Results p38β p38γ

We have previously shown that TGF-b1 induces endo- p38γ δ-Tubulin thelial cell apoptosis in vitro and in vivo through VEGF/ VEGFR-2 activation of p38MAPK (19, 20). In the absence of MAPK TGF-b1, VEGF activates p38 but protects endothelial Figure 1. Endothelial cell expression of p38 isoforms. A, RT-PCR analysis cells from apoptosis (19). Therefore, our finding raised the of HUVE cell expression of p38MAPK isoforms. GAPDH is shown as a question: how does TGF-b1 convert VEGF signaling from loading control in the bottom. B, Western blotting analysis of HUVE and MAPK BCE cell extracts expression of p38MAPK isoforms. Human MCF-7 anti- to proapoptotic? Because p38 exists in 4 isoforms mammary carcinoma cells are shown as a control for p38d. The blot was with different biological functions, we hypothesized that probed sequentially with the different antibodies. C, recombinant (r) MAPK TGF-b1 shifts VEGF activation of p38 from 1 isoform purified p38a, p38b, and p38g analyzed by Western blotting with to another. antibodies to the 3 isoforms. Coomassie blue staining of the SDS/PAGE To investigate this hypothesis, we characterized endothe- gel is shown as a loading control. D, Western blot analysis of p38 isoform expression in HUVE cells incubated with either FGF-2 (10 ng/mL) or VEGF lial cell expression of the p38 isoforms, and the effect of (30 ng/mL) or TGF-b1 (1 ng/mL) for the indicated times. Cells receiving no TGF-b1 and VEGF on their activation. By reverse tran- addition (CONTROL) were used as control. The blot was probed scriptase PCR (RT-PCR) endothelial cells expressed p38a, sequentially with antibodies to p38a, b,andg and to a-tubulin as a b, and g mRNAs, whereas p38d mRNA was barely detect- loading control. These experiments were repeated twice with able (Fig. 1A). Western blotting with p38MAPK isoform- comparable results. specific antibodies showed expression of the corresponding proteins (Fig. 1B and C). that treatment with FGF-2 or VEGF selectively activated Neither VEGF nor TGF-b1 treatment of endothelial cells p38b, whereas TGF-b1orUVB irradiation activated p38a. significantly altered p38a, b,org expression (Fig. 1D), Western blotting analysis of caspase 3 activation, a marker of suggesting that the activity of these kinases is regulated at apoptosis, showed that TGF-b1 and UVB induced cell posttranscriptional level. Isoform-specific antibodies to death, whereas FGF-2 and VEGF had no such effect (Fig. phosphorylated p38 do not exist because the isoforms' 2B). Therefore, these findings suggested that in endothelial phosphorylation sequences are identical. Therefore, we cells p38a mediates apoptotic signaling, whereas p38b relays immunoprecipitated extracts of TGF-b1- or VEGF-treated prosurvival signaling. Our previous studies (19, 20) have cells with antibodies to the individual p38 isoforms and shown that in endothelial cells TGF-b1 activation of analyzed the immunoprecipitates by Western blotting with p38MAPK is abolished by downregulation of VEGFR-2, an antibody that recognizes all phosphorylated isoforms. showing that endothelial cell VEGF mediates p38MAPK The results (Fig. 2A) showed that untreated endothelial cells activation by TGF-b1. In addition, TGF-b1 does not induce had no active p38a but showed comparable levels of active apoptosis in endothelial cells that do not express VEGF in p38b and p38g. Treatment with TGF-b1 selectively acti- response to TGF-b1 (19). Because in the absence of TGF-b1 vated p38a. Conversely, VEGF selectively activated p38b. VEGF activates p38b, our results suggested that TGF-b1 FGF-2, which protects endothelial cells from apoptosis, induces endothelial cell apoptosis by shifting VEGF activa- MAPK strongly upregulated p38b activation; in contrast, UVB tion of p38 from the b to the a isoform. irradiation, which induces apoptosis, activated p38a. The apoptotic effect of TGF-b1 is rapid (3–12 hours) and Although the anti-p38a and p38b antibodies did not followed by refractoriness of the surviving cells to TGF-b1 show cross-reactivity by Western blotting (Fig. 1C), they induction of apoptosis up to 96 hours (20). We found that might cross-react by immunoprecipitation. Therefore, we TGF-b1 treatment of endothelial cells for 72 to 96 hours carried out reverse experiments in which cell extracts were resulted in increased BrdU uptake and cyclin E expression immunoprecipitated with antibody to phospho-p38 and the between 24 and 96 hours (Figs. 3 and 4), coincident with the immunoprecipitates analyzed by Western blotting with formation of capillary-like structures (Fig. 3; refs. 20, 7). antibodies to p38a or p38b. The results (Fig. 2B) showed These structures are likely representative of precapillary

www.aacrjournals.org Mol Cancer Res; 10(5) May 2012 607

Downloaded from mcr.aacrjournals.org on September 29, 2021. © 2012 American Association for Cancer Research. Published OnlineFirst April 20, 2012; DOI: 10.1158/1541-7786.MCR-11-0507

Ferrari et al.

AB

VEGF FGF-2 TGF-β1UV Control FGF-2 VEGF TGF- β 1 UV IP –+ –+ –+ –+WB ERK1/2 (p) p38 p38α Tot p38 cl-Csp3 (p) p38 (p) p38 p38β Tot p38 (p) p38 IP (p) p38 WB p38β p38γ Tot p38 IP (p) p38 WB p38α

Figure 2. Control of p38 isoform activation in endothelial cells. A, endothelial cells were incubated with either FGF-2 (10 ng/mL) or VEGF (30 ng/mL) or TGF- b1 2 (1 ng/mL) for 6 hours, or irradiated with UVB (20 mJ/cm ). Cell extracts were immunoprecipitated with antibodies to p38a, b,org and analyzed by Western blotting with the corresponding antibodies and with antibody to phospho-p38 [(p)p38]. B, endothelial cells were treated with the indicated reagentsas described above. Cell extracts were immunoprecipitated with antibody to phospho-p38 and analyzed by Western blotting with antibodies to p38a or p38b (bottom). Extracts were also analyzed by Western blotting for cleaved caspase-3 (cl-Csp3), an apoptosis marker, and p38 activation [(p)p38] (top). ERK1/2, loading control.

cords and do not reflect endothelial to mesenchymal tran- Consistent with our previous findings (19, 20), Western sition as the cells did not express a-smooth muscle actin blotting analysis with an antibody that recognizes all phos- (Supplementary Fig. S1). A low level of apoptosis (caspase-3 phorylated p38 isoforms showed that TGF-b1 induced activation) was also detected between 72 and 96 hours of p38MAPK activation between 3 and 9 hours of treatment TGF-b1 treatment, concurrent with cell proliferation (Figs. (Fig. 4A), concomitantly with the onset of apoptosis, and 3 and 4). The level of BrdU uptake induced by TGF-b1 was subsequently between 72 and 96 hours, a time when comparable with that induced by 10% FCS and occurred in addition of TGF-b1 fails to induce apoptosis (20) and cell cells grown in starvation medium (0.5% serum) for 72 to 96 proliferation occurs (Figs. 3 and 4). This observation sug- hours, a condition that induces endothelial cell apoptosis gested that the early induction of p38MAPK activation (3–6 (31). Thus, TGF-b1 has a dual effect on endothelial cells: hours) involves the a isoform, which is associated with rapid and transient induction of apoptosis, followed by apoptosis, and the late activation (72–96 hours) entails the sustained proliferation of the surviving cells. b isoform, which mediates survival signaling. Therefore, we On the basis of these findings, we hypothesized that the analyzed the cell extracts by immunoprecipitation with cell proliferation and refractoriness to apoptosis that follow antibody to phospho-p38MAPK followed by Western blot- TGF-b1–induced apoptosis are mediated by a p38MAPK ting with p38a or p38b antibodies. The results (Fig. 4A and isoform(s) other than the one that mediates apoptosis. B) showed that p38a activation occurred between the initial Therefore, we analyzed p38MAPK isoform activation in 6 to 24 hours of treatment, coincident with apoptosis, while endothelial cells treated with TGF-b1 for different times. p38b activation decreased. Subsequently, the level of active p38a decreased, and p38b activation increased between 48 and 96 hours, concurrently with cell proliferation (Fig. 3 16 and 4) and refractoriness to TGF-b1 induction of apoptosis 14 (20). Activation of p38a—to a level lower than at 6 hours— 12 10 also occurred at 72 to 96 hours, coincident with the low level 8 of caspase-3 activation observed at this time of incubation. 6 To confirm these findings, we used the in situ proximity 4 ligation assay (29) with antibodies to p38a or p38b together 2 % BrdU-positive cells % BrdU-positive with phospho-p38 antibodies, which allowed the detection 0 0 6 24 48 72 96 FCS of phosphorylated p38a and p38b in individual cells. The h results showed that cells treated with TGF-b1 for 6 hours stained positively for p38a but not for p38b. Conversely, TGF-b1 treatment for 72 hours resulted in activation of p38b but not of p38a (Fig. 5). Thus, altogether these results indicated that p38a mediates TGF-b1 induction of apo- ptosis, whereas p38b is associated with TGF-b1–induced cell proliferation and refractoriness to apoptosis. We therefore used siRNAs to selectively downregulate the Figure 3. TGF-b1 induction of endothelial cell proliferation and capillary p38a, b, and g isoforms and characterized their effect on morphogenesis. BrdU uptake by BCE cells treated with TGF-b1(1ng/mL) for the indicated times. Cells incubated with 10% FCS are shown as apoptosis. Transient transfection with siRNA to one p38 positive control. The cells were photographed at the indicated times to isoform downregulated the corresponding isoform without show morphologic changes. Magnification, 100. affecting expression of the other isoforms (Fig. 6A), showing

608 Mol Cancer Res; 10(5) May 2012 Molecular Cancer Research

Downloaded from mcr.aacrjournals.org on September 29, 2021. © 2012 American Association for Cancer Research. Published OnlineFirst April 20, 2012; DOI: 10.1158/1541-7786.MCR-11-0507

Control of Endothelial Cell Apoptosis by p38MAPK Isoforms

The morphologic analysis of the siRNA-transfected cells A + TGF-β1 showed effects consistent with the biochemical characteri- 9672482460 h zation of apoptosis (Fig. 6C). As described (7, 20), confluent Cl-Csp 3 endothelial cells treated with TGF-b1 for 72 to 96 hours formed structures that resemble precapillary cords (Fig. 6C, a Cyclin E and b). However, TGF-b1–treated cells transfected with fl (p) p38 p38a siRNA retained a con uent monolayer with no apparent morphologic changes (Fig. 6C, c). In contrast, (p) 38α transfection with p38b or p38g siRNAs caused massive cell death resulting in sparse, disrupted cord-like structures (Fig. (p) p38β 6C, d and e). Notably, downregulation of these p38 isoforms p38 blocked the proliferative effect of TGF-b1 that follows the transient induction of apoptosis (Figs. 3 and 4), and induced B massive apoptosis at a time of TGF-b1 treatment (72–96 1.2 hours) when endothelial cells are refractory to TGF-b1 Caspase-3 1 induction of apoptosis (20). To investigate the relative contribution of p38a and p38b 0.8 Cyclin E to blood vessel formation, we used an in vitro angiogenesis assay (Fig. 7). For this purpose, HUVE cells were either 0.6 mock transfected or transiently transfected with siRNAs to 0.4 the a, b,org isoforms of p38 and grown in 3D collagen gels (arbitary units) in the presence of 1 ng/mL of TGF-b1. Within 5 to 7 days, Appoptosis/proliferation 0.2 mock transfected cells formed a network of cord-like struc- 0 tures as described (30). In contrast, with cells transfected with the siRNAs to the p38MAPK isoforms this effect was 1 dramatically blocked. Cells transfected with p38a siRNA showed very rudimentary cord-like structures larger than 0.8 p38α p38β those obtained with mock-transfected cells and consisting of 0.6 flattened cells that in several areas maintained a confluent monolayer. Conversely, cells transfected with siRNAs to 0.4 p38 activation

(arbitary units) p38b or p38g showed sparse, disrupted capillary-like struc- 0.2 tures that failed to form a network (Fig. 7). Numerous areas of the collagen gels only contained isolated clusters of 0 Hours globular cells. Thus, the p38MAPK isoforms have opposing roles in TGF-b1 induction of in vitro angiogenesis, which mirror Figure 4. TGF- b1 induction of endothelial cell apoptosis and proliferation their effects on endothelial cell apoptosis and survival. correlates with activation of distinct p38 isoforms. A, Western blotting a b analysis of caspase 3 cleavage, cyclin E expression, and activation of Because p38 activation by TGF- 1 is mediated by VEGF total p38 (all isoforms), p38a, and p38b in endothelial cells treated with (19, 20), which in the absence of TGF-b1 activates p38b, TGF-b1 (1 ng/mL) for the indicated times. Total p38 [(p)p38] activation was analyzed by Western blotting with antibody to phospho-p38. Activation of p38a and p38b [(p)p38a and (p)p38] was characterized by immunoprecipitation with antibody to phospho-p38 followed by Western blotting with antibodies to p38a or p38b. Total p38MAPK (p38) is shown as a loading control. Representative blots from triplicate experiments are shown. B, densitometric analysis of the Western blot bands shown in A. that the inhibitory action of the siRNA did not reflect nonspecific effects of the siRNAs or the transfection condi- tions. Transient transfection with p38a siRNA or a mixture of siRNAs to all p38 isoforms abolished TGF-b1 induction of apoptosis. Conversely, transfection with p38b or p38g siRNA induced apoptosis both in untreated and TGF-b1– treated cells (Fig. 6B). Thus, in endothelial cells, p38a transduces apoptotic signaling by apoptosis inducers such Figure 5. PLA of p38a and p38b activation by TGF-b1. HUVE cells treated as TGF-b1, and p38b transduces survival signaling by with TGF-b1 (1 ng/mL) for the indicated times or with control medium for prosurvival factors such as FGF-2 or VEGF. p38g also 72 hours. Phosphorylation of the p38 isoforms is revealed by fluorescent dots. The cells were counterstained with DAPI to evidence the nuclei. mediates survival signaling but its activation is not controlled Magnification: 600. The graph shows mean SE of dots per cell by VEGF or TGF-b1. counted in 20 to 30 cells from 3 independent experiments.

www.aacrjournals.org Mol Cancer Res; 10(5) May 2012 609

Downloaded from mcr.aacrjournals.org on September 29, 2021. © 2012 American Association for Cancer Research. Published OnlineFirst April 20, 2012; DOI: 10.1158/1541-7786.MCR-11-0507

Ferrari et al.

blocks endothelial cell apoptosis by decreasing p38MAPK phosphorylation (27, 35). However, the p38MAPK isoforms involved in endothelial cell apoptosis have not been char- acterized. In cell types other than endothelial cells, p38MAPK activation in response to a variety of stimuli induces either proliferation or growth arrest/apoptosis, depending on the cell type (21, 36–38). The specific p38 isoforms involved in these processes remain largely unidentified. In human fibro- blasts activation of p38g, but not of other p38 isoforms, is required for induction of G2 arrest by g-irradiation (39), suggesting that the individual p38 isoforms can have differ- ent effects on cell fate. Both p38a and b have been impli- cated in the pathogenesis of myocardial hypertrophy, where they may play different roles; p38b being more potent in inducing hypertrophy, and p38a in other functions such as apoptosis (40). Similarly, p38a but not p38b is required for ischemic preconditioning of the myocardium (41). Our study is the first to identify opposing roles for the a and b isoforms of p38MAPK in the control of endothelial cell survival and apoptosis. Figure 6. Control of endothelial cell apoptosis and survival by p38a, b, and Our data also show several features of the mechanism by g. A, Western blotting analysis of p38 isoform expression in HUVE cells which TGF-b1 controls endothelial cell death and prolif- transfected with siRNAs to either p38a or b or g. A blot with an antibody to all p38 isoforms (Tot. p38) is shown as a loading control. B, HUVE cells eration. Under most culture conditions, TGF-b1 is a potent transfected with siRNAs to either p38a or b or g, or with a mixture thereof inhibitor of endothelial cell proliferation and migration, and (Mix), or mock transfected (control) and incubated with TGF- b1(1ng/mL) an inducer of apoptosis (42, 43). However, TGF-b1 also or control medium for 6 hours were analyzed by Western blotting with stimulates endothelial cell proliferation in vitro and in vivo antibody to cleaved PARP (PARP p85) or to total PARP as a loading control. C, confluent HUVE cells grown in gelatin-coated dishes and (10, 43, 44). We found that the apoptotic effect of TGF-b1 either non transfected and untreated (a) or transfected with transfection reagent alone (b) or siRNAs to either p38a (c) or b (d) or g (e). The cells were photographed after 96 hours incubation in the absence (a) or presence (b–d) of TGF- b1 (1 ng/mL). Magnification, 100. Histogram. The results A were quantitated as described (20). Mean SE of triplicate samples are shown. , P < 0.05 (sample vs. a). These experiments were repeated 3 times with comparable results. our results show that TGF-b1 converts VEGF signaling from activation of the prosurvival p38b to the proapoptotic –siRNA p38α siRNA p38a.

Discussion We have previously shown that TGF-b1 induces endo- thelial cell apoptosis through VEGF/VEGFR-2–mediated activation of p38MAPK (19). Because in the absence of TGF- p38β siRNA p38γ siRNA MAPK b1, VEGF activates p38 but protects endothelial cells B from apoptosis, our finding suggested that TGF-b1 modifies 70 –siRNA* VEGF signaling from anti- to proapoptotic. The data 60 reported in this article show that TGF-b1 induces endothe- 50 MAPK lial cell apoptosis by shifting VEGF activation of p38 40

from the b to the a isoform. In endothelial cells, the b 30 p38α siRNA p38β siRNA p38γ siRNA isoform of p38MAPK mediates prosurvival signaling induced 20 by growth factors such as VEGF and FGF-2, which protect 10 endothelial cells from apoptosis. Conversely, p38a mediates # Capillary-like structures 0 apoptotic signaling from inducers of endothelial cell apo- ptosis such as TGF-b1 and UVB. p38g also mediates Figure 7. In vitro angiogenesis assay in 3D collagen gel. A, HUVE cells endothelial cell survival, but is not controlled by VEGF or transfected with the indicated siRNAs or mock transfected (-siRNA), and TGF-b1. grown in 3D collagen gels for 5 days in the presence of TGF-b1 (1 ng/mL). MAPK Magnification, 100. B, results of the experiments shown in A quantitated Numerous studies implicate p38 in apoptosis. In MAPK as described under Materials and Methods. Mean SE of triplicate endothelial cells, p38 activation mediates apoptosis samples are shown. , P < 0.05 (sample vs. a). These experiments were induction by a variety of agents (24–27, 32–34), and VEGF repeated 3 times with comparable results.

610 Mol Cancer Res; 10(5) May 2012 Molecular Cancer Research

Downloaded from mcr.aacrjournals.org on September 29, 2021. © 2012 American Association for Cancer Research. Published OnlineFirst April 20, 2012; DOI: 10.1158/1541-7786.MCR-11-0507

Control of Endothelial Cell Apoptosis by p38MAPK Isoforms

on endothelial cells is rapid and transient, and is followed by tion to prevent nonspecific effects. Indeed, in our studies, we a long period during which the surviving cells are refractory did not observe nonspecific apoptotic effects or p38MAPK to apoptosis induction by TGF-b1 (20) and proliferate. activation induced by our siRNAs. Our VEGFR-1 and These opposing effects of TGF-b1 are mediated by the VEGFR-2 siRNAs did not induce apoptosis (19, 20), nor selective activation of 2 different p38 isoforms. The rapid did our siRNAs to p38a or a mixture of siRNAs to p38a, b, induction of apoptosis results from downregulation of p38b and g (Fig. 6B). Similarly, our siRNAs to VEGFR-1 and and upregulation of p38a activation. Subsequently, down- VEGFR-1, or control siRNA to lamin A did not induce regulation of p38a and parallel, sustained increase in p38b p38MAPK activation (19). Therefore, our siRNAs did not activation provide survival signaling. On the basis of our have nonspecific effects on p38MAPK activation or endothe- data, we cannot conclude whether TGF-b1 induction of lial cell apoptosis mediated by TLR3. endothelial cell proliferation is also mediated by p38b Our results are in contrast with previous data showing that because siRNA-mediated downregulation of p38 b expres- pharmacologic inhibition of p38MAPK blocks endothelial cell sion results in rapid and massive apoptosis. Our previous apoptosis (19, 27, 35). Because p38b and p38g mediate findings that TGF-b1 induces prolonged upregulation of survival signaling, their inhibition should result in apoptosis. endothelial cell VEGF and FGF-2 expression and activates This discrepancy is explained by the fact that all currently extracellular signal-regulated kinase (ERK)1/2 by a VEGF/ available inhibitors of p38MAPK inhibit both p38a and VEGFR-2-dependent mechanism (19) suggest that the p38b, and therefore abolish both pro- and antiapoptotic proliferative effect of TGF-b1 could be mediated by signaling. In addition, the apoptotic effect of p38a activa- ERK1/2, while the sustained activation of p38b provides tion seems to be prevalent. Indeed, siRNA-mediated down- endothelial cells with survival signaling. Thus, TGF-b1 regulation of all the p38 isoforms expressed by endothelial controls both endothelial cell apoptosis and survival through cells blocks TGF-b1 induction of apoptosis as effectively as the selective activation of the a and b isoforms of p38MAPK, downregulation of p38a alone. In contrast, selective inhi- respectively. bition of p38b or p38g expression results in apoptosis (Fig. The effect of TGF-b1 on endothelial cells is mediated by 2 6B and C). type-I receptors (TGF-bRI): –like kinase On the basis of our data, we cannot conclude whether the (ALK)-5, the predominant type I receptor that mediates proliferative effect of TGF-b1 on endothelial cells is medi- TGF-b1 responses, and ALK-1, whose expression is restrict- ated by VEGF or other endothelial cell-derived growth ed to endothelial cells. ALK-5 and ALK-1 mediate TGF-b factors. TGF-b1-induced upregulation of VEGF lasts at signaling through distinct Smad proteins: Smad2/Smad3 least 24 h (19), a timing coincident with the onset of cell and Smad1/Smad5, respectively (11). Although TGF-b1 proliferation. Although TGF-b1 downregulates endothelial primarily activates Smad2 and Smad3, in primary endothe- cell expression of VEGFR-2, this effect does not exceed 40– lial cells it also activates Smad1 and Smad5 through a 50% (13, 14). In addition, TGF-b1 upregulates endothelial heterodimeric receptor complex consisting of ALK5 and cell expression of FGF-2 (19), which can also provide a ALK1 (45). The ALK5 and ALK1 pathways mediate oppos- mitogenic stimulus. Therefore, both VEGF and FGF-2 ing effects; ALK5 inhibits, whereas ALK1 stimulates cell could mediate TGF-b1 induction of endothelial cell prolif- migration and proliferation (46). In addition, ALK5, but not eration through sustained activation of the ERK1/2 or other ALK1, mediates TGF-b1 induction of endothelial cell signaling pathways. apoptosis (47). However, ALK-1/Smad1 signaling mediates The coordinated mechanism of control of endothelial cell TGF-b1 induction of VEGF expression in endothelial cells apoptosis, survival, and proliferation by TGF-b1—VEGF (18), which is required for TGF-b1 induction of apoptosis activation of p38a and b has profound effects on in vitro (19). Based on our data, we cannot conclude whether ALK1 angiogenesis. Downregulation of p38a-mediated apoptosis and ALK5 activate different p38 isoforms. It is tempting blocks the formation of capillary-like structures in 3D to speculate that signaling by both receptors is required collagen gel. Similarly, induction of excess apoptosis by for TGF-b1 induction of apoptosis, and that downregula- downregulation of p38b—mediated survival signaling tion of ALK5 is responsible for the ensuing cell proliferation blocks endothelial cell sprouting and organization into a (Fig. 4; ref. 20). The relative contribution of TGFbRs capillary-like network. These findings are consistent with to activation of the different isoforms of p38MAPK and our previous observation that inhibition of endothelial cell control of endothelial cell apoptosis warrants further apoptosis blocks angiogenesis in vitro and in vivo (20). investigation. We have previously shown that VEGF mediates TGF-b1 Our conclusions are partly based on results of siRNA- induction of endothelial cell apoptosis through activation of mediated gene silencing experiments. It has been reported p38MAPK (19). This conclusion was based on our findings that siRNAs can inhibit angiogenesis by a nonspecific effect that inhibition of VEGF or downregulation of VEGFR-2 mediated by toll-like receptor 3 (TLR3), a double-stranded blocks TGF-b1 activation of p38MAPK, and inhibition of RNA receptor that generates apoptotic signaling in endo- p38MAPK expression abrogates TGF-b1 induction of apo- thelial cells (48, 49). This effect requires siRNAs more than ptosis (19). Our data show that VEGF activates the pro- 21 nucleotides in length, and has been obtained with survival p38b, whereas TGF-b1 activates the proapoptotic nonchemically modified molecules. We used 19-bp siRNAs p38a. Because TGF-b1 activates p38a through VEGF, with dinucleotide 30 DNA overhangs, a chemical modifica- which in the absence of TGF-b1 activates p38b, our findings

www.aacrjournals.org Mol Cancer Res; 10(5) May 2012 611

Downloaded from mcr.aacrjournals.org on September 29, 2021. © 2012 American Association for Cancer Research. Published OnlineFirst April 20, 2012; DOI: 10.1158/1541-7786.MCR-11-0507

Ferrari et al.

VEGF VEGF TGF-β1 VEGF VEGF TGF-β1 VEGF VEGF

flt-1 flt-1 flt-1 flk-1 flk-1 flk-1

P P P P P P

P p38MAPK β α P P p38MAPK α β p38MAPK β α

Survival Apoptosis Survival

Figure 8. Control of endothelial cell apoptosis and survival by TGF- b1–VEGF interaction. In the absence of TGF- b1, VEGF induces activation of p38 b through VEGFR-2 and promotes endothelial cell survival (left). TGF- b1 induces VEGF expression. Crosstalk between VEGF- and TGF- b1–specific signaling pathways converts VEGF/VEGFR-2 activation of p38b into activation of the proapoptotic p38a. Activation of p38a is required for TGF- b1 induction of apoptosis; however, parallel TGF- b1 signaling may also be necessary. After the initial induction of apoptosis, endothelial cells become refractory to the apoptotic effect of TGF- b1, and VEGF/VEGFR-2 signaling reverts to activation of the prosurvival p38 b (right).

show that TGF-b1 shifts VEGF signaling from p38b to effect of TGF-b1 on apoptosis is transient (Fig. 4) and p38a. followed by refractoriness of the cells to TGF-b1induc- Our results therefore generate the following model for the tion of apoptosis (20), inhibition of p38b and/or p38g control of endothelial cell apoptosis and survival by TGF-b1 signaling results in sustained induction of apoptosis by and VEGF (Fig. 8). In the absence of TGF-b1, VEGF TGF-b1 (Fig. 6). Presently, all synthetic inhibitors of p38 binding to VEGFR-2 induces endothelial cell activation block both the a and the b isoforms. Novel inhibitors that of p38b and cell survival (Fig. 8, left). TGF-b1 induces selectively inhibit p38b activity (or activation) could block endothelial cell expression of VEGF and switches VEGF/ the prosurvival effect of TGF-b1(andVEGF)onendo- VEGFR-2–activated signaling to the proapoptotic p38a. thelial cells which follow apoptosis induction. These This effect involves crosstalk between TGF-b1 and VEGF inhibitors could thus transform the transient apoptotic signaling pathways that can occur at multiple levels upstream effect of TGF-b1 into a sustained, massive effect. Mice of p38MAPK. VEGF/VEGFR-2–mediated activation of genetically deficient in p38b show no major phenotype if p38a is necessary for TGF-b1 induction of endothelial cell unchallenged (50), indicating that selective inhibition of apoptosis; however, parallel TGF-b1–initiated signaling p38b would represent a feasible and safe approach to acting in concert with p38a may also be required. Subse- antitumor angiogenesis therapy. quently (Fig. 8, right), the cells become refractory to b a b TGF- 1, p38 activation is downregulated and p38 acti- Disclosure of Potential Conflicts of Interest vation upregulated. This effect results in abrogation of No potential conflicts of interest were disclosed. apoptosis and increased prosurvival signaling. The signaling mechanism(s) that control the selective Grant Support activation of the a or b isoform of p38 are largely This work was supported by NIH grants R01 HL070203, R01 HL070203-03S1, 5R01 CA136715, R21 RAG033735A; and funds from Department of Cardiothoracic unknown, as is the crosstalk between TGF-b1andVEGF Surgery, NYU School of Medicine, and Dr. Leo A. Shifrin and Roslyn Myers Breast signaling that mediates the shift of VEGF activation of Cancer Discovery Fund (P. Mignatti), R21 AG028785 (G. Pintucci), RC1 HL10035 p38 from the b to the a isoform. Understanding these (G. Ferrari). The costs of publication of this article were defrayed in part by the payment of page mechanisms can shed light on the complex control of charges. This article must therefore be hereby marked advertisement in accordance with vascular endothelial cell survival and death during angio- 18 U.S.C. Section 1734 solely to indicate this fact. genesis. Importantly, our findings provide indications for the development of novel pharmacologic approaches to Received October 19, 2011; revised February 29, 2012; accepted March 14, 2012; inhibit tumor angiogenesis. We found that, whereas the published OnlineFirst April 20, 2012.

References 1. Ferrara N. Vascular endothelial growth factor. Arterioscler Thromb 3. Bierie B, Moses HL. Transforming growth factor beta (TGF-beta) and Vasc Biol 2009;29:789–91. inflammation in cancer. Growth Factor Rev 2010;21:49–59. 2. Tian M, Neil JR, Schiemann WP. Transforming growth factor-beta and 4. Ferrara N. Vascular endothelial growth factor: basic science and the hallmarks of cancer. Cell Signal 2011;23:951–62. clinical progress. Endocr Rev 2004;25:581–611.

612 Mol Cancer Res; 10(5) May 2012 Molecular Cancer Research

Downloaded from mcr.aacrjournals.org on September 29, 2021. © 2012 American Association for Cancer Research. Published OnlineFirst April 20, 2012; DOI: 10.1158/1541-7786.MCR-11-0507

Control of Endothelial Cell Apoptosis by p38MAPK Isoforms

5. Koch S, Tugues S, Li X, Gualandi L, Claesson-Welsh L. Signal trans- 25. Chai W, Liu Z. p38 mitogen-activated protein kinase mediates palmi- duction by vascular endothelial growth factor receptors. Biochem J tate-induced apoptosis but not inhibitor of nuclear factor-kappaB 2011;437:169–83. degradation in human coronary artery endothelial cells. Endocrinology 6. Santibanez JF, Quintanilla M, Bernabeu C. TGF-beta/TGF-beta recep- 2007;148:1622–8. tor system and its role in physiological and pathological conditions. 26. Kumar P, Miller AI, Polverini PJ. p38 MAPK mediates gamma-irradi- Clin Sci 2011;121:233–51. ation-induced endothelial cell apoptosis, and vascular endothelial 7. Choi ME, Ballermann BJ. Inhibition of capillary morphogenesis and growth factor protects endothelial cells through the phosphoinositide associated apoptosis by dominant negative mutant transforming 3-kinase-Akt-Bcl-2 pathway. J Biol Chem 2004;279:43352–60. growth factor-beta receptors. J Biol Chem 1995;270:21144–50. 27. Gratton JP, Morales-Ruiz M, Kureishi Y, Fulton D, Walsh K, Sessa WC. 8. Madri JA, Pratt BM, Tucker AM. Phenotypic modulation of endothelial Akt down-regulation of p38 signaling provides a novel mechanism of cells by transforming growth factor-beta depends upon the compo- vascular endothelial growth factor-mediated cytoprotection in endo- sition and organization of the extracellular matrix. J Cell Biol thelial cells. J Biol Chem 2001;276:30359–65. 1988;106:1375–84. 28. Seghezzi G, Patel S, Ren CJ, Gualandris A, Pintucci G, Robbins ES, 9. Roberts AB, Sporn MB, Assoian RK, Smith JM, Roche NS, Wakefield et al. -2 (FGF-2) induces vascular endothelial LM, et al. Transforming growth factor type beta: rapid induction of growth factor (VEGF) expression in the endothelial cells of forming fibrosis and angiogenesis in vivo and stimulation of collagen formation capillaries: an autocrine mechanism contributing to angiogenesis. J in vitro. Proc Natl Acad Sci U S A 1986;83:4167–71. Cell Biol 1998;141:1659–73. 10. Yang EY, Moses HL. Transforming growth factor beta 1-induced 29. Soderberg O, Gullberg M, Jarvius M, Ridderstrale K, Leuchowius KJ, changes in cell migration, proliferation, and angiogenesis in the chick- Jarvius J, et al. Direct observation of individual endogenous protein en chorioallantoic membrane. J Cell Biol 1990;111:731–41. complexes in situ by proximity ligation. Nat Methods 2006;3: 11. Orlova VV, Liu Z, Goumans MJ, Ten Dijke P. Controlling angiogenesis 995–1000. by two unique TGF-beta type I receptor signaling pathways. Histol 30. Koh W, Stratman AN, Sacharidou A, Davis GE. In vitro three dimen- Histopathol 2011;26:1219–30. sional collagen matrix models of endothelial lumen formation during 12. Pollman MJ, Naumovski L, Gibbons GH. Vascular cell apoptosis: cell vasculogenesis and angiogenesis. Methods Enzymol 2008;443: type-specific modulation by transforming growth factor-beta1 in 83–101. endothelial cells versus smooth muscle cells. Circulation 1999;99: 31. Gerber H-P, Dixit V, Ferrara N. Vascular endothelial growth factor 2019–26. induces expression of the antiapoptotic proteins Bcl-2 and A1 in 13. Minami T, Rosenberg RD, Aird WC. Transforming growth factor-beta vascular endothelial cells. J Biol Chem 1998;273:13313–16. 1-mediated inhibition of the flk-1/KDR gene is mediated by a 50- 32. Yue TL, Ni J, Romanic AM, Gu JL, Keller P, Wang C, et al. TL1, a novel untranslated region palindromic GATA site. J Biol Chem 2001;276: -like cytokine, induces apoptosis in endothelial 5395–402. cells. Involvement of activation of stress protein kinases (stress-acti- 14. Mandriota SJ, Menoud PA, Pepper MS. Transforming growth factor vated protein kinase and p38 mitogen-activated protein kinase) and beta 1 down-regulates vascular endothelial growth factor receptor 2/ caspase-3-like protease. J Biol Chem 1999;274:1479–86. flk-1 expression in vascular endothelial cells. J Biol Chem 1996;271: 33. Xu ZR, Hu L, Cheng LF, Qian Y, Yang YM. Dihydrotestosterone 11500–5. protects human vascular endothelial cells from H(2)O(2)-induced apo- 15. Pollman MJ, Naumovski L, Gibbons GH. Endothelial cell apoptosis in ptosis through inhibition of caspase-3, caspase-9 and p38 MAPK. Eur capillary network remodeling. J Cell Physiol 1999;178:359–70. J Pharmacol 2010;643:254–9. 16. Tsukada T, Eguchi K, Migita K, Kawabe Y, Kawakami A, Matsuoka N, 34. Jiang H, Liang C, Liu X, Jiang Q, He Z, Wu J, et al. Palmitic acid et al. Transforming growth factor beta 1 induces apoptotic cell death in promotes endothelial progenitor cells apoptosis via p38 and JNK cultured human umbilical vein endothelial cells with down-regulated mitogen-activated protein kinase pathways. Atherosclerosis 2010; expression of bcl-2. Biochem Biophys Res Commun 1995;210: 210:71–7. 1076–82. 35. Yilmaz A, Kliche S, Mayr-Beyrle U, Fellbrich G, Waltenberger J. p38 17. Hyman KM, Seghezzi G, Pintucci G, Stellari G, Kim JH, Grossi EA, et al. MAPK inhibition is critically involved in VEGFR-2-mediated endothelial Transforming growth factor-beta1 induces apoptosis in vascular cell survival. Biochem Biophys Res Commun 2003;306:730–6. endothelial cells by activation of mitogen-activated protein kinase. 36. Ambrosino C, Nebreda AR. Cell cycle regulation by p38 MAP kinases. Surgery 2002;132:173–79. Biol Cell 2001;93:47–51. 18. Bostrom K, Zebboudj AF, Yao Y, Lin TS, Torres A. Matrix GLA protein 37. Delston RB, Matatall KA, Sun Y, Onken MD, Harbour JW. p38 phos- stimulates VEGF expression through increased transforming growth phorylates Rb on Ser567 by a novel, cell cycle-independent mecha- factor-beta1 activity in endothelial cells. J Biol Chem 2004;279: nism that triggers Rb-Hdm2 interaction and apoptosis. Oncogene 52904–13. 2011;30:588–99. 19. Ferrari G, Pintucci G, Seghezzi G, Hyman K, Galloway AC, Mignatti P. 38. Smeeton J, Zhang X, Bulus N, Mernaugh G, Lange A, Karner CM, et al. VEGF, a prosurvival factor, acts in concert with TGF-beta1 to induce Integrin-linked kinase regulates p38 MAPK-dependent cell cycle arrest endothelial cell apoptosis. Proc Natl Acad Sci U S A 2006;103: in ureteric bud development. Development 2010;137:3233–43. 17260–5. 39. Wang X, Mcgowan CH, Zhao M, He L, Downey JS, Fearns C, et al. 20. Ferrari G, Cook BD, Terushkin V, Pintucci G, Mignatti P. Transforming Involvement of the MKK6-p38gamma cascade in gamma-radiation- growth factor-beta 1 (TGF-beta1) induces angiogenesis through vas- induced cell cycle arrest. Mol Cell Biol 2000;20:4543–52. cular endothelial growth factor (VEGF)-mediated apoptosis. J Cell 40. New L, Han J. The p38 MAP kinase pathway and its biological function. Physiol 2009;219:449–58. Trends Cardiovasc Med 1998;8:220–8. 21. Cuenda A, Rousseau S. p38 MAP-kinases pathway regulation, func- 41. Sicard P, Clark JE, Jacquet S, Mohammadi S, Arthur JS, O'keefe SJ, tion and role in human diseases. Biochim Biophys Acta 2007;1773: et al. The activation of p38 alpha, and not p38 beta, mitogen-activated 1358–75. protein kinase is required for ischemic preconditioning. J Mol Cell 22. Cuadrado A, Nebreda AR. Mechanisms and functions of p38 MAPK Cardiol 2010;48:1324–8. signalling. Biochem J 2010;429:403–17. 42. Sawdey M, Podor TJ, Loskutoff DJ. Regulation of type 1 plasminogen 23. Adams RH, Porras A, Alonso G, Jones M, Vintersten K, Panelli S, et al. activator inhibitor in cultured bovine aortic endothelial Essential role of p38alpha MAP kinase in placental but not embryonic cells. Induction by transforming growth factor-beta, lipopolysaccha- cardiovascular development. Mol Cell 2000;6:109–16. ride, and tumor necrosis factor-alpha. J Biol Chem 1989;264:10396– 24. Ohta T, Eguchi R, Suzuki A, Miyakaze S, Ayuzawa R, Kaji K. Hypoxia- 401. induced apoptosis and tube breakdown are regulated by p38 MAPK 43. Madri JA, Bell L, Merwin JR. Modulation of vascular cell behavior but not by caspase cascade in an in vitro capillary model composed of by transforming growth factors beta. Mol Reprod Dev 1992;32: human endothelial cells. J Cell Physiol 2007;211:673–81. 121–6.

www.aacrjournals.org Mol Cancer Res; 10(5) May 2012 613

Downloaded from mcr.aacrjournals.org on September 29, 2021. © 2012 American Association for Cancer Research. Published OnlineFirst April 20, 2012; DOI: 10.1158/1541-7786.MCR-11-0507

Ferrari et al.

44. Iruela-Arispe ML, Sage EH. Endothelial cells exhibiting angiogenesis in transforming growth factor-beta in human umbilical vein endothelial vitro proliferate in response to TGF-beta 1. J Cell Biochem 1993;52: cells. J Cell Physiol 2002;193:299–318. 414–30. 48. Berge M, Bonnin P, Sulpice E, Vilar J, Allanic D, Silvestre JS, et al. Small 45. Goumans MJ, Valdimarsdottir G, Itoh S, Rosendahl A, Sideras P, interfering RNAs induce target-independent inhibition of tumor growth Ten Dijke P. Balancing the activation state of the endothelium and vasculature remodeling in a mouse model of hepatocellular via two distinct TGF-beta type I receptors. Embo J 2002;21: carcinoma. Am J Pathol 2010;177:3192–201. 1743–53. 49. Kleinman ME, Yamada K, Takeda A, Chandrasekaran V, Nozaki M, 46. Goumans MJ, Valdimarsdottir G, Itoh S, Lebrin F, Larsson J, Mummery Baffi JZ, et al. Sequence- and target-independent angiogenesis sup- C, et al. Activin receptor-like kinase (ALK)1 is an antagonistic mediator pression by siRNA via TLR3. Nature 2008;452:591–7. of lateral TGFbeta/ALK5 signaling. Mol Cell 2003;12:817–28. 50. Beardmore VA, Hinton HJ, Eftychi C, Apostolaki M, Armaka M, Darragh 47. Ota T, Fujii M, Sugizaki T, Ishii M, Miyazawa K, Aburatani H, et al. J, et al. Generation and characterization of p38beta (MAPK11) gene- Targets of transcriptional regulation by two distinct type I receptors for targeted mice. Mol Cell Biol 2005;25:10454–64.

614 Mol Cancer Res; 10(5) May 2012 Molecular Cancer Research

Downloaded from mcr.aacrjournals.org on September 29, 2021. © 2012 American Association for Cancer Research. Published OnlineFirst April 20, 2012; DOI: 10.1158/1541-7786.MCR-11-0507

TGF-β1 Induces Endothelial Cell Apoptosis by Shifting VEGF Activation of p38 MAPK from the Prosurvival p38β to Proapoptotic p38 α

Giovanni Ferrari, Vitaly Terushkin, Martin J. Wolff, et al.

Mol Cancer Res 2012;10:605-614. Published OnlineFirst April 20, 2012.

Updated version Access the most recent version of this article at: doi:10.1158/1541-7786.MCR-11-0507

Supplementary Access the most recent supplemental material at: Material http://mcr.aacrjournals.org/content/suppl/2012/04/20/1541-7786.MCR-11-0507.DC1

Cited articles This article cites 50 articles, 22 of which you can access for free at: http://mcr.aacrjournals.org/content/10/5/605.full#ref-list-1

Citing articles This article has been cited by 3 HighWire-hosted articles. Access the articles at: http://mcr.aacrjournals.org/content/10/5/605.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://mcr.aacrjournals.org/content/10/5/605. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.

Downloaded from mcr.aacrjournals.org on September 29, 2021. © 2012 American Association for Cancer Research.