<<

Published OnlineFirst April 7, 2009; DOI: 10.1158/0008-5472.CAN-08-2524 Published Online First on April 7, 2009 as 10.1158/0008-5472.CAN-08-2524

Research Article

Connective Tissue Confers Drug Resistance in Breast through Concomitant Up-regulation of Bcl-xL and cIAP1

Ming-Yang Wang,1,2,3 Pai-Sheng Chen,1,2 Ekambaranellore Prakash,1 Hsing-Chih Hsu,1,2,3 Hsin-Yi Huang,4 Ming-Tsan Lin,3,5 King-Jen Chang,2,3 and Min-Liang Kuo1,2

1Laboratory of Molecular and Cellular Toxicology, Institute of Toxicology, College of Medicine, and 2Angiogenesis Research Center, National Taiwan University; and Departments of 3Surgery, 4Pathology, and 5Primary Care Medicine, National Taiwan University Hospital, Taipei, Taiwan

Abstract possessing an NH2-terminal signal peptide indicative of their Connective tissue growth factor (CTGF) expression is elevated secreted protein nature. The biological properties of CCN proteins in advanced breast cancer and promotes metastasis. Chemo- include the stimulation of cellular proliferation, migration, adhesion, therapy response is only transient in most metastatic diseases. and formation, as well as the regulation of In the present study,we examined whether CTGF expression and tumorigenesis. CTGF is a secreted growth factor could confer drug resistance in human breast cancer. In breast that can bind to on the cell surface (3) and also serves cancer patients who received neoadjuvant chemotherapy, as an angiogenic factor in collaboration with matrix metalloprotei- CTGF expression was inversely associated with chemotherapy nases (4). Elevated CTGF expression has been observed in breast response. Overexpression of CTGF in MCF7 cells (MCF7/CTGF) (5), pancreatic cancers (6), melanomas (7), and chondro- enhanced clonogenic ability,cell viability,and resistance to sarcomas (8). Although CTGF shows multiple roles in various cancer apoptosis on exposure to doxorubicin and paclitaxel. Reducing types, in breast tumor cells CTGF overexpression has been linked to the CTGF level in MDA-MB-231 (MDA231) cells by antisense tumor size increase and lymph node metastasis (5, 9). Use of in vitro CTGF cDNA (MDA231/AS cells) mitigated this drug resistance and in vivo selection models and large-scale microarray analysis has capacity. CTGF overexpression resulted in resistance to shown that CTGF is critically involved in the formation of osteolytic doxorubicin- and paclitaxel-induced apoptosis by up-regula- bone metastasis in breast cancer (10, 11). In our previous study, tion of Bcl-xL and cellular inhibitor of apoptosis protein 1 CTGF promoted metastasis of breast cancer cells via an a h (cIAP1). Knockdown of Bcl-xL or cIAP1 with specific small v 3-extracellular signal–regulated kinase (ERK)-1/2–dependent, interfering RNAs abolished the CTGF-mediated resistance to S100A4-upregulated pathway. Thus, CTGF may play an important apoptosis induced by the chemotherapeutic agents in role in breast cancer progression and metastasis. MCF7/CTGF cells. Inhibition of extracellular signal–regulated Despite the proven survival benefit of systemic adjuvant kinase (ERK)-1/2 effectively reversed the resistance to apoptosis chemotherapy in breast cancer patients, a significant number of as well as the up-regulation of Bcl-xL and cIAP1 in MCF7/CTGF them still develop metastatic diseases and respond only transiently cells. A neutralizing antibody against integrin AvB3 significantly to conventional treatments leading to eventual mortality. In attenuated CTGF-mediated ERK1/2 activation and up- Greenberg’s report, only 16.6% of metastatic breast cancer patients regulation of Bcl-xL and cIAP1,indicating that the integrin achieved complete response by cytotoxic chemotherapy (12). Even AvB3/ERK1/2 signaling pathway is essential for CTGF functions. with targeted therapy using against HER2/neu The Bcl-xL level also correlated with the CTGF level in breast overexpressing breast cancer, complete response was <10% in cancer patients. We also found that a COOH-terminal domain Slamon’s series (13). It has been proposed that metastatic cancer peptide from CTGF could exert activities similar to full-length cells deregulate apoptotic modulators and that this plays a crucial CTGF,in activation of ERK1/2,up-regulation of Bcl-xL/cIAP1, role during the metastatic process (14–16). Because most of the and resistance to apoptosis. We conclude that CTGF expression conventional chemotherapeutic drugs induce apoptosis of cancer could confer resistance to chemotherapeutic agents through cells, we speculate that an increased resistance to drug-induced augmenting a survival pathway through ERK1/2-dependent apoptosis in metastatic cancer cells might be the cause of poor Bcl-xL/cIAP1 up-regulation. [Cancer Res 2009;69(8):3482–91] response in metastatic breast cancer patients. Taking into account that CTGF plays an important role in breast Introduction cancer metastasis and that metastatic cancer cells usually show resistance to apoptotic stress, we postulated that CTGF expression Connective tissue growth factor (CTGF), also known as CTGF/ would increase drug resistance in breast cancer cells. It is still unclear CCN2, belongs to the CCN family (1), consisting of six members: whether CTGF expression would deregulate apoptotic modulators CTGF, NOVH, CYR61, WISP-1, WISP-2, and WISP-3 (2), all of which or would confer drug resistance in breast cancer cells. We herein examined the effects of CTGF on the survival and chemosensitivity of Note: Supplementary data for this article are available at Cancer Research Online breast cancer cells and the signaling pathways involved therein. (http://cancerres.aacrjournals.org/). M-Y. Wang and P-S. Chen contributed equally to this work. Requests for reprints: Min-Liang Kuo or King-Jen Chang, National Taiwan University, 1 Jen Ai Road, Section I, Room 544, Taipei 100, Taiwan. Phone: 886-2-23123456-5083; Fax: Materials and Methods 886-2-23410217; E-mail: [email protected] or [email protected]. I2009 American Association for Cancer Research. Patients and specimens. Fifty-eight patients with locally advanced doi:10.1158/0008-5472.CAN-08-2524 breast cancer who had received chemotherapy before surgery at National

Cancer Res 2009; 69: (8). April 15, 2009 3482 www.aacrjournals.org

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 2009 American Association for Cancer Research. Published OnlineFirst April 7, 2009; DOI: 10.1158/0008-5472.CAN-08-2524

Connective Tissue Growth Factor and Drug Resistance

Figure 1. Role of CTGF in chemotherapeutic response of human breast cancer. A, immunohistochemical staining for CTGF in breast cancer tissue from surgical resection specimens. CTGF protein was predominantly expressed in the cytoplasm and membrane of carcinoma cells. Staining intensity was scored as indicated. B, association between CTGF expression level and chemotherapy response of patients with breast cancer. C, clonogenic ability of MCF7/neo and MCF7/CTGF. Cells were treated with various concentrations of paclitaxel or doxorubicin for 24 h and then refed with fresh medium. After 2 wk of culture, the colonies were stained with crystal violet and counted. Each experiment was done in triplicate. D, cell viability of MCF7/neo and MCF7/CTGF on exposure to chemotherapy drugs. Cells were cultured in 96-well plates in the presence or absence of increasing concentrations of paclitaxel or doxorubicin until the untreated control cells reached confluence. Cell viability and IC50 values were then determined by using the MTT assay. P < 0.005, Student’s t test.

Taiwan University Hospital were included. Paraffin-embedded, formalin- Jackson ImmunoResearch. Antibody-bound protein bands were detected fixed surgical resection specimens were collected for immunohistochemical with enhanced chemiluminescence reagents (Amersham Pharmacia staining for CTGF. Tumor size, local invasion, and lymph node metastasis Biotech) and photographed with Kodak X-Omat Blue autoradiography film were acquired from pathology reports. Prechemotherapy tumor burden (Perkin-Elmer Life Sciences). The constitutively active mutant of mitogen- was estimated by tumor dimension acquired either by image or direct activated protein kinase/ERK kinase (MEK)-1 construct (a substitution of measurement of the breast tumor. Postchemotherapy tumor dimensions the regulatory phosphorylation sites, Ser218 and Ser222, with aspartic acid were acquired from the final pathology report after surgery. Chemotherapy S218D/S222D mutant) was a gift from Dr. Ruey-Hwa Chen and has been response was determined as a change in tumor volume. described previously (17). The primer sequences for PCR are shown in Cell culture,antibodies,and reagents. Human breast cancer cell Supplementary Table S3. lines MCF7 and MDA231 were obtained from the American Type Immunohistochemical staining. For immunohistochemical staining, Culture Collection. BT474, BT483, MDA-MB-453, and MDA-MB-435 cells with the procedure previously described (18), CTGF, Bcl-xL, and cIAP1 were generously provided by the National Health Research Institute expression were evaluated by our pathologist using a semiquantitative (Chunan, Taiwan). These cell lines were maintained in DMEM or MEM weighted histoscore method. We have randomly chosen five low-power supplemented with 10% fetal bovine serum (Life Technologies, Inc.), views. The immunohistochemistry results were scored by taking the 2 mmol/L L-glutamine (Life Technologies), 100 Ag/mL streptomycin, and percentage of positivity and intensity of staining into account. The a h 100 units/mL penicillin in a humidified 5% CO2 atmosphere. Integrin v 3, intensity score was given as follows: 0, no staining; 1, weak positivity; 2, a-tubulin, p-ERK1/2, ERK1/2, CTGF, Bcl-xL, phosphorylated focal adhesion moderate positivity; and 3, strong positivity. The percentage of positivity kinase (p-FAK), FAK, p-Src, Src, p-AKT, AKT, poly(ADP-ribose) polymerase was multiplied by the intensity score to obtain a final value designated as (PARP), and cellular inhibitor of apoptosis protein 1 (cIAP1) antibodies were the immunohistochemistry score. The sums of five immunohistochemistry purchased from Santa Cruz Biotechnology. CTGF antibody was from BD scores represent the CTGF, Bcl-xL, or cIAP1 expression of breast cancer. Biosciences. Goat anti-rabbit or anti-mouse IgG conjugated with either The sum of an immunohistochemistry score z2 was defined as high FITC or tetramethylrhodamine isothiocyanate were purchased from expression. www.aacrjournals.org 3483 Cancer Res 2009; 69: (8). April 15, 2009

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 2009 American Association for Cancer Research. Published OnlineFirst April 7, 2009; DOI: 10.1158/0008-5472.CAN-08-2524

Cancer Research

Detection of sub-G1 cells by flow cytometry. Adherent and detached X-100/PBS at 37jC for 30 min with 1 mg/mL of RNase A, and stained with cells were collected after trypsin detachment. Cells were washed with 0.5 mL of 50 mg/mL propidium iodide for 10 min. Fluorescence emitted from ice-cold PBS and fixed in 70% ethanol at À20jC for at least 1 h. the propidium iodide-DNA complex was quantified after laser excitation of After fixation, cells were washed twice, incubated in 0.5 mL of 0.5% Triton the fluorescent dye by a FACScan flow cytometer (BD Biosciences).

Figure 2. CTGF increased resistance to apoptosis induced by paclitaxel and doxorubicin in breast cancer cells. A, effect of CTGF on drug-induced apoptosis in MCF7 cells (left) and MDA231 cells (right). Cells were treated with 5 Amol/L paclitaxel or 10 Amol/L doxorubicin for 24 h and then assayed by TUNEL analysis. Top, immunofluorescence photomicrographs of cells undergoing apoptosis (green staining, arrowhead) and the corresponding propidium iodide–counterstained photomicrographs. Bottom, percentage of TUNEL-positive cells. *, P < 0.05. B, detection of cleaved PARP induced by paclitaxel and doxorubicin. Cells were serum starved for 24 h and then treated with 5 Amol/L paclitaxel or 10 Amol/L doxorubicin for 24 h. Apoptosis was examined by PARP cleavage. C, effects of rCTGF on drug-induced apoptosis. Detection of apoptotic cells in MDA231 cells treated with paclitaxel and rCTGF. After serum starvation for 24 h, MDA231/AS cells were treated with 5 Amol/L paclitaxel and 200 ng/mL rCTGF. D, PARP cleavage in MDA231 cells treated with paclitaxel and rCTGF. Columns, mean of triplicate experiments; bars, SD. a, P < 0.05, compared with MDA231/neo without treatment; b, P < 0.05, compared with MDA231/AS without treatment; c, P < 0.05, compared with MDA231/ AS treated with paclitaxel (Student’s t test).

Cancer Res 2009; 69: (8). April 15, 2009 3484 www.aacrjournals.org

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 2009 American Association for Cancer Research. Published OnlineFirst April 7, 2009; DOI: 10.1158/0008-5472.CAN-08-2524

Connective Tissue Growth Factor and Drug Resistance

Terminal deoxyribonucleotidyl transferase–mediated dUTP nick status, HER2/neu, tumor grade, tumor size, and stage between end labeling assay. The terminal deoxyribonucleotidyl transferase– these two groups (Supplementary Table S1). mediated dUTP nick end labeling (TUNEL) technology used was the CTGF enhanced clonogenic ability and viability in response DeadEnd Fluorometric TUNEL System (Promega). For TUNEL staining, all to chemotherapeutic agents in human breast cancer cells. fixation and staining procedures were conducted at room temperature. In our previous study, we found that MCF7 cells expressed low Slides were prepared with 2 Â 105 cells on each slide. Then these slides were treated according to the manufacturer’s instructions. The slides were levels of CTGF, whereas MDA231 cells expressed abundant CTGF counterstained with propidium iodide for 12 h at 4jC. The localized green (9). To confirm that CTGF could confer drug resistance in breast fluorescence of apoptotic cells was detected by fluorescence microscopy. cancer cells, we established MCF7/CTGF cells overexpressing Recombinant CTGF. Suspension FreeStyle 293-F cells (Invitrogen) at a CTGF and MDA231/AS cells in which CTGF expression was density of 1.1 Â 106/mL were transfected with CTGF expression vector with reduced by transfection with antisense-oriented CTGF. As the FreeStyle 293 Expression System (Invitrogen) according to the previously reported, no appreciable difference in ability j manufacturer’s instructions. After culture at 37 C with 8% CO2 and 92% was evident among these cells at 24, 48, and 72 hours after plating air in an orbital shaker for 48 h, conditioned medium was collected. To (9), as measured by trypan blue exclusion assay (data not shown). ultrafilter and concentrate the dilute protein sample, Amicon Ultra-15 Overexpressing CTGF in MCF7 cells significantly increased their devices (Millipore) were used for rapid separation of recombinant CTGF clonogenic ability on exposure to paclitaxel and doxorubicin at (rCTGF) protein without any loss of activity. Small interfering RNA treatment for Bcl-xL. SiRNA for Bcl-xL was various doses (Fig. 1C), whereas decreased CTGF expression in designed and synthesized using the software and Stealth siRNA duplex MDA231 cells drastically decreased their clonogenic ability on oligoribonucleotides from Invitrogen. Of the small interfering RNA (siRNA) exposure to these drugs (Supplementary Fig. S2). Hence, we suggest constructs tested, the greatest knockdown was seen with the oligo targeting that elevated CTGF levels can increase the clonogenic ability of Bcl-xL (5¶-AGGAGAGAAAGUCAACCACCAGCUC-3¶). We used si-GAPDH breast cancer cells exposed to chemotherapeutic agents. (Ambion) as the control siRNA. Inhibition of Bcl-xL expression was assessed To determine the magnitude of CTGF expression affecting by reverse transcription-PCR (RT-PCR) and Western blot analyses following chemosensitivity toward paclitaxel and doxorubicin, the IC50 of transfection of human breast cancer cells with Bcl-xL siRNA (siBcl-xL). proliferation was determined by 3-(4,5-dimethylthiazol-2-yl)-2,5- Briefly, cells were grown in 6-cm dishes and transiently transfected with 50 diphenyltetrazolium bromide (MTT) assay in these breast cancer pmol of siRNA using 8 Ag of TransFast transfection reagent (Promega) in a cells at 24 hours after exposure to paclitaxel or doxorubicin. total transfection volume of 2 mL of serum-free DMEM. After incubation at MCF7/CTGF cells had greater cell viability than the MCF7/neo 37jC with 5% CO2 for 8 h, 2 mL of DMEM containing 20% normal growth medium were added. Samples were then prepared and analyzed by RT-PCR, control cells in response to paclitaxel and doxorubicin (Fig. 1D). Western blot, and invasion assays. Similarly, reducing CTGF expression in MDA231 cells drastically Statistical analysis. All observations were confirmed by at least three reduced cell viability on exposure to these chemotherapeutic F independent experiments. The data are presented as means SD. Student’s agents (Supplementary Fig. S3). The IC50 values of the MCF7/CTGF t test (paired and unpaired) was used to evaluate the statistical significance cells toward doxorubicin and paclitaxel were nearly 7 and of mean values. Levels of statistical significance were P < 0.05 and P < 0.005. 15 times higher than those of the MCF7/neo cells, respectively All P values were two-tailed. (10.43 F 4.02 versus 1.48 F 0.22 Amol/L and 6.30 F 0.89 versus F A 0.41 0.11 mol/L; P < 0.01, t test). The IC50 values of the MDA231/neo cells toward doxorubicin and paclitaxel were Results nearly 17 and 9 times higher than those of the MDA231/AS CTGF expression is inversely associated with chemothera- cells, respectively (1.50 F 0.39 versus 0.09 F 0.01 Amol/L and peutic response in breast cancer patients. CTGF has been 5.13 F 0.87 versus 0.55 F 0.08 Amol/L; P < 0.01, t test; shown to play an important role in breast cancer metastasis and Supplementary Table S2). These results indicate that expression also associated with increased tumor size and lymph node of CTGF enhances resistance to chemotherapeutic agents in metastasis (5, 9). Because highly metastatic cancer cells exhibit a human breast cancer cells. drug-resistant phenotype (14), we speculated that CTGF expression CTGF increased resistance to apoptosis induced by might increase drug resistance in breast cancer cells. To investigate paclitaxel and doxorubicin in breast cancer cells. To confirm the effect of CTGF expression and the chemotherapy response in that CTGF-induced drug resistance is mediated through inhibition breast cancer patients, 58 tumors from breast cancer patients of apoptosis, we exposed cells to paclitaxel and doxorubicin receiving neoadjuvant chemotherapy were analyzed immunohis- and subjected them to apoptosis assays. MCF7/CTGF cells tochemically for CTGF expression from their surgical resection exhibited increased resistance to apoptosis induced by paclitaxel specimen. Patients were divided into two groups: those with and doxorubicin as quantitatively determined by TUNEL assay tumors expressing low levels of CTGF (scores 0 and 1, Fig. 1A) and (Fig. 2A, left). Thirty-seven percent of cells were induced into those with tumors expressing high levels of CTGF (scores 2 and 3, apoptosis in MCF7/neo cells after exposure to 5 Amol/L paclitaxel Fig. 1A). Twenty-three (40%) tumors had low CTGF expression and for 24 hours; in contrast, only 15% were induced in MCF7/CTGF 35 (60%) tumors had high CTGF expression. Chemotherapy cells (P < 0.05, t test). Similarly, the percentage of apoptotic cells response was defined as the quotient of postchemotherapy was also higher in MCF7/neo cells than in MCF7/CTGF cells on tumor burden divided by prechemotherapy tumor burden exposure to 10 Amol/L doxorubicin for 24 hours (41 F 5% versus (Supplementary Fig. S1). Patients with low levels of CTGF 26 F 5%; P < 0.05, t test). Reducing CTGF expression in MDA231 expression had a better chemotherapy response than those with cells greatly increased the percentage of apoptotic cells induced by high levels of CTGF expression (response quotients, 0.16 F 0.05 paclitaxel and doxorubicin. When exposed to paclitaxel and versus 0.91 F 0.28; P < 0.05, t test), indicating that breast cancer doxorubicin, apoptotic cells were 22% and 29% in MDA231/neo patients with a higher CTGF expression had a poorer response to cells versus 47% and 44% in MDA231/AS cells, respectively (Fig. 2A, chemotherapy (Fig. 1B). We have similar results from diagnostic right; P < 0.05, t test). Apoptosis was further verified by PARP biopsy (data not shown). There was no difference in age, hormone cleavage. Paclitaxel and doxorubicin treatment caused robust www.aacrjournals.org 3485 Cancer Res 2009; 69: (8). April 15, 2009

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 2009 American Association for Cancer Research. Published OnlineFirst April 7, 2009; DOI: 10.1158/0008-5472.CAN-08-2524

Cancer Research

Figure 3. Bcl-xL and cIAP1 are functional effectors in CTGF-mediated drug resistance. A, regulation of Bcl-2 and IAP family proteins by CTGF in MCF7 and MDA231 cells. RT-PCR analysis of Bcl-2 and IAP expression in MCF7/neo, MCF7/CTGF, MDA231/neo, and MDA231/AS cells (left). Protein levels of Bcl-xL and cIAP1 were confirmed by Western blot (right). B, top, knockdown of Bcl-xL or cIAP1 by transfection with specific siRNAs in MCF7/neo and MCF7/CTGF cells. Bcl-xL and cIAP1 protein expression were determined by Western blot. Bottom, effect of drug-induced apoptosis on inhibition of Bcl-xL/cIAP1. MCF7/neo and MCF7/CTGF cells were transfected with siRNA toward Bcl-xL, cIAP1, or scramble control. After transfection, these cells were treated with 5 Amol/L paclitaxel for 24 h and then subjected to TUNEL assay. a, P < 0.05, compared with MCF7/neo treated with scramble siRNA; b, P < 0.05, compared with MCF7/CTGF treated with scramble siRNA; c, P < 0.05, compared with MCF7/CTGF treated with paclitaxel and scramble siRNA (Student’s t test). C, top, overexpression of Bcl-xL/cIAP1 expression in MDA231/neo and MDA231/AS cells by transfection with Bcl-xL– or cIAP1-expressing vectors. Bottom, effect of overexpression of Bcl-xL/cIAP1 on drug-induced apoptosis. a, P < 0.05, compared with MDA231/neo transfected with pcDNA3; b, P < 0.05, compared with MDA231/AS transfected with pcDNA3. c, P < 0.05, compared with MDA231/AS transfected with pcDNA3 and treated with paclitaxel (Student’s t test). Columns, mean of triplicate experiments; bars, SD.

PARP cleavage in MCF7/neo and MDA231/AS as compared with Because CTGF is a secreted protein, we examined whether that of MCF7/CTGF and MDA231/neo cells (Fig. 2B). The above exogenous administration of CTGF could reverse the reduced data confirmed that CTGF significantly increased resistance to CTGF expression effect. We treated MDA231/AS cells with drug-induced apoptosis. 100 ng/mL rCTGF, then exposed them to paclitaxel and analyzed

Cancer Res 2009; 69: (8). April 15, 2009 3486 www.aacrjournals.org

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 2009 American Association for Cancer Research. Published OnlineFirst April 7, 2009; DOI: 10.1158/0008-5472.CAN-08-2524

Connective Tissue Growth Factor and Drug Resistance

Figure 4. Role of integrin avh3 in CTGF-induced signal transduction. A, MCF7 cells were serum starved for 24 h and then treated with 10 Ag/mL IgG or integrin  5 avh3–blocking antibodies for 24 h. P-FAK, FAK, p-Src, Src, p-ERK1/2, ERK1, Bcl-xL, and cIAP1 expression were measured by Western blotting. B, cells (5 10 ) were seeded in 6-cm dishes, serum starved for 24 h, and then treated with 10 Ag/mL of integrin avh3–blocking antibody or IgG and 2 Amol/L paclitaxel for 24 h. Cell viability was examined by MTT assay. C, involvement of ERK1/2 activation in the regulation of Bcl-xL/cIAP1 and drug-induced apoptosis. Top, cells were treated with 20 Amol/L PD98059 for 24 h. Bcl-xL and cIAP1 were examined by Western blot. Bottom, cells were pretreated with PD98059 for 24 h before adding paclitaxel. Apoptotic cells were stained with propidium iodide and analyzed by flow cytometry. D, top, MDA231 cells were transfected with constitutively active mutant MEK1. After 24 h, Bcl-xL and cIAP1 were examined by Western blot. Bottom, cells were transfected with constitutively active mutant MEK1. After 24 h, these cells were treated with paclitaxel. Apoptotic cells were detected by staining with propidium iodide and analyzed by flow cytometry. Columns, mean of experiments independently repeated thrice in triplicate; bars, SD. *, P < 0.05; #, nonsignificant (Student’s t test). by apoptosis assays. Treatment of cells with rCTGF reduced MDA231/AS and MDA231/neo cells induced by paclitaxel (Fig. 2D; paclitaxel-induced apoptotic cells from 75% to 49% in MDA231/AS P < 0.05, t test). We previously examined the expression levels of and from 46% to 14% in MDA231/neo cells (Fig. 2C; P < 0.05, CTGF in several breast cancer cell lines and found that the BT483, t test). The rCTGF also drastically reduced PARP cleavage in both T47D, MDA-MB-453, and MCF7 cell lines expressed extremely low www.aacrjournals.org 3487 Cancer Res 2009; 69: (8). April 15, 2009

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 2009 American Association for Cancer Research. Published OnlineFirst April 7, 2009; DOI: 10.1158/0008-5472.CAN-08-2524

Cancer Research or undetectable CTGF mRNA, whereas MDA231 and MDA435 pathway contributed to the resistance to apoptosis induced by cells expressed it in abundant levels (9). Western blots revealed paclitaxel. that protein expression was correlated with mRNA level in all Bcl-xL and cIAP1 are downstream effectors of CTGF. To these cells (Supplementary Fig. S4, top). To investigate the effects investigate the molecular mechanism of the antiapoptotic proper- of endogenous CTGF in other breast cancer cells, we treated ties of CTGF, we examined the expression of Bcl-2 and inhibitor of those cells with CTGF antiserum and examined the effects on apoptosis protein (IAP) family proteins by RT-PCR and Western drug resistance. Treatment with CTGF antiserum did not impair blot. The mRNA and protein levels of Bcl-xL and cIAP1 were cell viability, suggesting that CTGF expression is not essential both up-regulated by CTGF (Fig. 3A). To determine whether the for cell growth. Administration of CTGF antiserum in combination up-regulation of Bcl-xL and cIAP1 were the crucial determinants of with paclitaxel significantly decreased cell viability in MDA231 the resistance to drug-induced apoptosis in these cells, we assessed and MDA435 cells. Conversely, it did not increase the drug the effects of siRNA targeting Bcl-xL and cIAP1 on apoptosis sensitivity in T47D, BT483, and MCF7 cells, all of which showed low sensitivity in MCF7/CTGF cells. The siRNA targeting Bcl-xL or or undetectable CTGF levels (Supplementary Fig. S4, bottom). cIAP1 effectively down-regulated expression of the target proteins, These results suggest that a CTGF-induced outside-in signaling compared with the control siRNA (Fig. 3B, top). Then, the MCF7

Figure 5. CT domain is required for CTGF function in drug resistance. A, Western blot analysis of p-FAK, FAK, p-ERK1/2, ERK1, Bcl-xL, and cIAP1 expression in MCF7 cells stably transfected with CTGF, CT, or neo control. B, cell viability of MCF7/neo, MCF7/CTGF, and MCF7/CT cells on exposure to doxorubicin and paclitaxel. MCF7/neo, MCF7/CTGF, and MCF7/CT cells were cultured in 96-well plates in the presence or absence of increasing concentrations of paclitaxel or doxorubicin. Cell viability was then determined using the MTT assay. C, activation of FAK, Src, and ERK1/2 after rCTGF or rCT treatment. MCF7 cells (5 Â 105) were seeded in 6-cm dishes, serum starved for 24 h, and then treated with serum, 100 ng/mL rCTGF, or equimolar rCT. P-FAK, FAK, p-Src, Src, p-ERK1/2, ERK1, Bcl-xL, and cIAP1 expression were analyzed by Western blot. D, PARP cleavage after treatment with rCTGF or rCT. MCF7 (top) or MDA231/AS (bottom) cells were serum starved for 24 h and then treated with serum, 100 ng/mL rCTGF or equimolar rCT, and 5 Amol/L paclitaxel for 24 h. Cleaved PARP was detected by Western blot.

Cancer Res 2009; 69: (8). April 15, 2009 3488 www.aacrjournals.org

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 2009 American Association for Cancer Research. Published OnlineFirst April 7, 2009; DOI: 10.1158/0008-5472.CAN-08-2524

Connective Tissue Growth Factor and Drug Resistance cells were exposed to paclitaxel after siRNA treatment and were and there was no significant difference between MDA231/AS and further analyzed by TUNEL and MTT assays. Paclitaxel induced MDA231/neo cells (Fig. 4D, bottom). These results indicated that apoptosis in 31% of MCF7/neo cells and 9% of MCF7/CTGF cells. the ERK1/2 pathway is essential for CTGF-mediated drug a h Transfection with siRNA targeting Bcl-xL increased the apoptosis resistance and that the integrin v 3/FAK/ERK1/2 pathway is percentage from 9% to 20% in MCF7/CTGF cells but did not critical for CTGF-induced ERK1/2 activation and subsequent drug increase apoptosis in MCF7/neo cell cultures. The siRNA targeting resistance. cIAP1 had similar effects and increased apoptosis in MCF7/CTGF COOH-terminal domain is critical for CTGF function in cells to 23%. Both siRNAs significantly increased apoptosis in drug resistance. CTGF has four structural modules that resemble MCF7/CTGF cells but did not manifest any effects in MCF7/neo an -like growth factor-binding domain (module 1), a von cells (Fig. 3B, bottom). We also found that both siRNAs had similar Willebrand factor type C repeat (module 2), a type effects on decreasing cell viability after exposure to paclitaxel in I repeat (module 3), and a COOH-terminal domain (CT) that MCF7/CTGF cells (data not shown). According to these results, contains a putative cysteine knot (module 4; Supplementary both Bcl-xL and cIAP1 are important downstream effectors in Fig. S6) A previous study suggests that the CT could bind to a h CTGF-enhanced drug resistance in MCF7 cells. integrin v 3 (30). Hence, we hypothesized that the CT could Furthermore, transfection with Bcl-xL or cIAP1 vectors success- confer drug resistance to breast cancer cells via binding to integrin fully increased the corresponding protein levels in MDA231/AS avh3. We transfected MCF7 cells with plasmids expressing CT cells (Fig. 3C, top), suggesting that down-regulation of Bcl-xL and and found that both full-length CTGF and CT had similar effects cIAP1 contributed to increased paclitaxel-induced apoptosis, on activating FAK and ERK1/2; moreover, expression of CT also whereas restoring the levels of Bcl-xL or cIAP1 decreased apoptosis up-regulated Bcl-xL and cIAP1 in MCF7 cells (Fig. 5A). To examine from 52% to 34% and 26%, respectively, after exposure to paclitaxel whether CT expression could confer drug resistance in breast (Fig. 3C, bottom). These results suggest that CTGF expression cancer cells, we performed an MTT assay in these stable cells after conferred resistance to drug-induced apoptosis in breast cancer exposing them to paclitaxel. As we speculated, MCF7/CTGF and cells via up-regulation of Bcl-xL and cIAP1. MCF7/CT had similar viability profiles after exposure to various A Integrin vB3–dependent ERK1/2 activation is required for concentrations of paclitaxel (Fig. 5B). We further synthesized a CTGF-conferred drug resistance. CTGF exerts a range of diverse recombinant CT peptide (rCT) and found that rCT treatment functions by binding with cell surface integrins, including isoforms activated the FAK/ERK1/2 pathway and up-regulated Bcl-xL and a h a h a h a h a h v 3, IIb 3, M 2, 5 1, and 6 1 (19–22). Most integrins activate cIAP1 expression in MCF7 cells (Fig. 5C). rCT could also reduce FAK and thereby also Src-family kinases, which causes the PARP cleavage dramatically in both MDA231/AS and MCF7 cells phosphorylation of and therefore signaling from p130CAS and induced by paclitaxel (Fig. 5D). The above data indicated that CT paxillin (23). Here, we found that CTGF expression also dramat- exhibits similar activity as full-length CTGF to induce drug ically induced FAK and Src phosphorylation (Fig. 4A). Similar resistance in breast cancer cells. results were also observed with rCTGF (Supplementary Fig. S5). CTGF expression is associated with Bcl-xL expression in Bcl-xL and cIAP1 protein expression were induced in breast cancer patients. Having shown the inverse association of f120 minutes (data not shown). Earlier we have observed that CTGF expression with chemotherapy response in breast cancer integrin avh3 was essential for CTGF-induced ERK1/2 phosphor- patients (Fig. 1), we further investigated whether the CTGF level ylation (9). In addition, integrin avh3–blocking antibody inhibited correlated with Bcl-xL and cIAP1 in breast cancer patients. Hence, FAK and ERK1/2 phosphorylation as well as Bcl-xL and cIAP1 we examined Bcl-xL and cIAP1 expression by immunohistochem- protein expression in MCF7/CTGF cells. Treatment with integrin ical staining in serial histologic sections of the same patient avh3–blocking antibody also increased paclitaxel-induced cell group studied in Fig. 1 (Fig. 6). Among patients who had a high death in MCF7/CTGF cells but not in IgG-treated cells (Fig. 4B). CTGF expression level, 74% also highly expressed Bcl-xL, whereas The ERK1/2 pathway has previously been implicated in cellular in the low CTGF expression group, only 39% expressed high Bcl-xL protection from various apoptotic signals (19, 24–27). Inhibition of (P < 0.05, Fisher’s exact test.). In addition, the CTGF level had a MEK1 activity has been reported to down-regulate Bcl-xL protein tendency to always be associated with the cIAP1 level, although expression and promoter activity and to enhance susceptibility to the P value did not reach statistically significant levels (P = 0.29; cell death induction (28). The MEK/ERK pathway also mediates Table 1). These data agreed with our in vitro results that CTGF tumor necrosis factor-aÀinduced cIAP1 expression in human expression up-regulated Bcl-xL and cIAP1 expression. endothelial cells (29). To examine whether ERK1/2 activation contributed to the resistance to paclitaxel-induced apoptosis in these cells, the MEK1-specific inhibitor PD98059 and a constitu- Discussion tively active mutant MEK1 were used to modulate the ERK1/2 In this study, we show that CTGF overexpression in human phosphorylation status. PD98059 effectively inhibited ERK1/2 breast cancer cells enhances chemoresistance through inhibition of phosphorylation and Bcl-xL and cIAP1 levels in MCF7/CTGF cells apoptosis. The apoptosis inhibition was accompanied by decreased (Fig. 4C, top). The sub-G1 population of MCF7/CTGF cells induced PARP cleavage and a reduction in the sub-G1 population as evinced by paclitaxel significantly increased from 15% to 22% after PD98059 by nuclear condensation and TUNEL assays. CTGF overexpression treatment, whereas the control cells exhibited no change. There resulted in specific up-regulation of Bcl-xL and cIAP1, although the was no sub-G1 population difference in these cells after treatment expression levels of the others remained unchanged. We also a h with PD98059 (Fig. 4C, bottom). On the other hand, transfection showed that activation of ERK1/2 via the integrin v 3/FAK with constitutively active MEK1 caused ERK1/2 rephosphorylation pathway by CTGF is responsible for up-regulation of Bcl-xL and and restored Bcl-xL and cIAP1 expression in MDA231/AS cells cIAP1, and that its inactivation effectively reverses both the (Fig. 4D, top). In addition, transfection with constitutively active resistance to drug-induced apoptosis and the up-regulation of MEK1 in MDA231/AS cells decreased sub-G1 cells from 44% to 26%, Bcl-xL and cIAP1 in CTGF-overexpressing cells. Thus, CTGF www.aacrjournals.org 3489 Cancer Res 2009; 69: (8). April 15, 2009

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 2009 American Association for Cancer Research. Published OnlineFirst April 7, 2009; DOI: 10.1158/0008-5472.CAN-08-2524

Cancer Research

Figure 6. Correlation between CTGF, Bcl-xL, and cIAP1 expression in breast cancer patients. Representative pictures of immunohistochemical staining for CTGF, Bcl-xL, and cIAP1 in breast cancer tissues. Low (scores 0 and 1) and high (scores 2 and 3) expression of CTGF, Bcl-xL, and cIAP1 in breast cancer tumors are shown.

confers drug resistance in breast cancer cells by increasing baculoviral IAP repeat domains (36), which modulate apoptosis by resistance to apoptosis through integrin avh3/FAK/ERK1/2– binding and inactivating caspases (37). In the present study, we mediated up-regulation of the Bcl-xL and cIAP1 survival pathway. proposed a novel mechanism in which CTGF could up-regulate Recently, CTGF has been reported to have an important role in cIAP1 and thereby confer resistance to drug-induced apoptosis. and fibrotic disease, whereas CTGF is also reported Knockdown of Bcl-xL or cIAP1 in MCF7/CTGF cells increased to be critical in tumor development and progression (5, 31–33), as drug-induced apoptosis. Likewise, restoring Bcl-xL and cIAP1 in well as in tumor cell survival (34). In this study, we noticed that MDA231/AS cells reduced drug-induced apoptosis. Thus, we activation of ERK1/2 also up-regulated Bcl-xL and cIAP1, which concluded that both Bcl-xL and cIAP1 play important roles in are considered to be antiapoptotic proteins. The role of Bcl-xL in CTGF-induced drug resistance. exerting resistance to chemotherapy treatment with conventional Using immunohistochemical staining, we examined 58 breast antineoplastic agents was first suggested when it was observed cancer patients receiving neoadjuvant chemotherapy and observed that the expression of Erb-B2 (Her-2/neu) in breast cancer cells a correlation between the CTGF level and a reduction in increased the Bcl-xL level and rendered the cells resistant chemotherapy response. No pathologic complete response was to tamoxifen-induced apoptosis (35). CTGF-induced Bcl-xL found in our patient samples. Three patients in the low CTGF up-regulation is compatible with previous observations that CTGF group had only residual tumor cells and one had no residual tumor is associated with a poor prognosis in breast cancer patients and cells after chemotherapy. In Mauri’s review article, rates of path- the enhanced metastatic ability of breast cancer cells. IAPs are a ologic response are low among these studies (ranging from 4% to family of antiapoptotic proteins characterized by the presence of 29%; ref. 38). Absence of pathological complete response might be due to a limited number of patients involved in this study. Patients with high CTGF had a significantly lower chemotherapy response Table 1. Correlation of Bcl-xL or cIAP1 expression with than those in the low CTGF group. Bcl-xL and cIAP1 expression CTGF expression in breast cancer patients were also examined in these patients by immunohistochemistry in serial sections. Both Bcl-xL and cIAP1 expression levels correlated Low CTGF High CTGF P* with the CTGF level although cIAP1 expression did not reach sta- tistically significant. This supports our in vitro experiment showing Bcl-xL that CTGF up-regulated Bcl-xL and cIAP1 expression. Low 14 9 <0.05 CTGF antiserum reduced cell viability when cotreated with High 9 26 chemotherapeutic agents such as doxorubicin or paclitaxel. Over- cIAP1 Low 15 17 0.29 expression of CTGF in MCF7 cells significantly increased High 8 18 anchorage-independent growth in soft agar, and reducing CTGF expression in MDA231 cells decreased anchorage-independent growth. In Dornho¨fer’s study, a specific monoclonal antibody *Fisher’s exact test. against CTGF could suppress anchorage-independent growth in soft agar and inhibit tumor growth and lymph node metastasis in

Cancer Res 2009; 69: (8). April 15, 2009 3490 www.aacrjournals.org

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 2009 American Association for Cancer Research. Published OnlineFirst April 7, 2009; DOI: 10.1158/0008-5472.CAN-08-2524

Connective Tissue Growth Factor and Drug Resistance mouse xenograft model cells (39). In confor- Disclosure of Potential Conflicts of Interest mity to our previous data that a neutralizing antibody could re- No potential conflicts of interest were disclosed. verse the effects of CTGF overexpression, CTGF neutralizing antibody may not only increase chemotherapy response in breast cancer but may also be able to decrease micrometastasis and Acknowledgments tumor recurrence. Received 7/2/08; revised 1/11/09; accepted 2/6/09; published OnlineFirst 4/7/09. We have shown that CTGF overexpression in breast cancer cells Grant support: National Science Council, Taiwan (NSC95-2314-B-002-198-), the elicits an increase in resistance toward doxorubicin and paclitaxel, National Health Research Institutes, Taiwan (NHRI-EX97-9712BC), China Medical University, Taichung, Taiwan and National Taiwan University (97F008-111), and the as well as the specific up-regulation of Bcl-xL and cIAP1. In view of Ministry of Economic Affairs, Taipei, Taiwan (93-EC-17-A-19-S1-0016). its prominent expression in advanced breast cancer tissues as well The costs of publication of this article were defrayed in part by the payment of page as in breast cancer cell lines, coupled with its effect on the charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. regulation of apoptosis, CTGF is evidently one of the important We thank Dr. Tung-Tien Sun (Department of Urology, New York University School of determinants of chemotherapy sensitivity in breast cancer. Hence, Medicine) for revising the paper; Dr. Ruey-Hwa Chen (Institute of Biological Chemistry, Academia Sinica, Taipei, Taiwan) for providing the active MEK1 plasmid; and Dr. John CTGF may be a novel predictive marker for chemosensitivity and a Barrett [Department of Physiology and Biophysics (R430), University of Miami Miller potential therapeutic target for breast cancer. School of Medicine] for generously providing us Bcl-xL plasmid.

References profound deficiency of apoptosis execution mecha- 27. Pardo OE, Arcaro A, Salerno G, et al. Fibroblast nisms. Cancer Lett 1997;115:185–93. growth factor-2 induces translational regulation of Bcl- 1. Bork P. The modular architecture of a new family of 16. Del Bufalo D, Biroccio A, Leonetti C, Zupi G.. Bcl-2 XL and Bcl-2 via a MEK-dependent pathway: correlation growth regulators related to connective tissue growth overexpression enhances the metastatic potential of a with resistance to etoposide-induced apoptosis. J Biol factor. FEBS Lett 1993;327:125–30. human breast cancer line. FASEB J 1997;11:947–53. Chem 2002;277:12040–6. 2. Perbal B. CCN proteins: multifunctional signalling 17. Brunet A, Pages G, Pouyssegur J, et al. Constitutively 28. Jost M, Huggett TM, Kari C, Boise LH, Rodeck U. regulators. Lancet 2004;363:62–4. active mutants of MAP kinase kinase (MEK1) induce Epidermal -dependent control of 3. Lau LF, Lam SC. The CCN family of angiogenic growth factor-relaxation and oncogenicity when keratinocyte survival and Bcl-xL expression through a regulators: the integrin connection. Exp Cell Res 1999; expressed in fibroblasts. Oncogene 1994;9:3379–87. MEK-dependent pathway. J Biol Chem 2001;276:6320–6. 248:44–57. 18. Chang CC, Shih JY, Jeng YM, et al. Connective 29. Furusu A, Nakayama K, Xu Q, Konta T, Kitamura M. 4. Kondo S, Kubota S, Shimo T, et al. Connective tissue tissue growth factor and its role in lung adenocar- MAP kinase-dependent, NF-nB-independent regulation growth factor increased by hypoxia may initiate cinoma invasion and metastasis. J Natl Cancer Inst of inhibitor of apoptosis protein genes by TNF-a.JCell angiogenesis in collaboration with matrix metallopro- 2004;96:364–75. Physiol 2007;210:703–10. teinases. Carcinogenesis 2002;23:769–76. 19. Babic AM, Chen CC, Lau LF, et al. Fisp12/mouse 30. Gao R, Brigstock DR. Connective tissue growth factor 5. Xie D, Nakachi K, Wang H, Elashoff R, Koeffler HP. connective tissue growth factor mediates endothelial (CCN2) Induces adhesion of rat activated hepatic a h Elevated levels of connective tissue growth factor, WISP- and migration through integrin v 3, stellate cells by binding of its C-terminal domain to a 1, and CYR61 in primary breast cancers associated with promotes endothelial cell survival, and induces integrin vh3 and proteoglycan. J Biol more advanced features. Cancer Res 2001;61:8917–23. angiogenesis in vivo. Mol Cell Biol 1999;19:2958–66. Chem 2004;279:8848–55. 6. Wenger C, Ellenrieder V, Alber B, et al. Expression and 20. Jedsadayanmata A, Chen CC, Kireeva ML, Lau LF, 31. Gurish MF, Boyce JA. Mast cell growth, differen- differential regulation of connective tissue growth factor Lam SC. Activation-dependent adhesion of human tiation, and death. Clin RevAllergy Immunol 2002; in pancreatic cancer cells. Oncogene 1999;18:1073–80. platelets to Cyr61 and Fisp12/mouse connective tissue 22:107–18. 7. Kubo M, Kikuchi K, Nashiro K, et al. Expression of growth factor is mediated through integrin aIIbh3. 32. Xie D, Yin D, Wang HJ, et al. Levels of expression of fibrogenic in desmoplastic malignant mela- J Biol Chem 1999;274:24321–7. CYR61 and CTGF are prognostic for tumor progression noma. Br J Dermatol 1998;139:192–7. 21. Schober JM, Chen N, Grzeszkiewicz TM, et al. and survival of individuals with gliomas. Clin Cancer a h 8. Shakunaga T, Ozaki T, Ohara N, et al. Expression of Identification of integrin M 2 as an adhesion receptor Res 2004;10:2072–81. connective tissue growth factor in cartilaginous tumors. on peripheral blood monocytes for Cyr61 (CCN1) and 33. Koliopanos A, Friess H, di Mola FF, et al. Connective Cancer 2000;89:1466–73. connective tissue growth factor (CCN2): immediate- tissue growth factor alters tumor 9. Chen PS, Wang MY, Wu SN, et al. CTGF enhances the early gene products expressed in atherosclerotic lesions. progression in esophageal cancer. World J Surg 2002; motility of breast cancer cells via an integrin-avh3-1/2- Blood 2002;99:4457–65. 26:420–7. dependent S100A4-upregulated pathway. J Cell Sci 2007; 22. Weston BS, Wahab NA, Mason RM, et al. CTGF 34. Croci S, Landuzzi L, Astolfi A, et al. Inhibition of 120:2053–65. mediates TGF-h-induced matrix deposition connective tissue growth factor (CTGF/CCN2) expres- a h 10. Kang Y, Siegel PM, Shu W, et al. A multigenic by up-regulating active 5 1 integrin in human sion decreases the survival and myogenic differentiation program mediating breast cancer metastasis to bone. mesangial cells. J Am Soc Nephrol 2003;14:601–10. of human rhabdomyosarcoma cells. Cancer Res 2004;64: Cancer Cell 2003;3:537–49. 23. Guo W, Giancotti FG.. Integrin signalling during 1730–6. 11. Minn AJ, Kang Y, Serganova I, et al. Distinct organ- tumour progression. Nat RevMol Cell Biol 2004;5: 35. Kumar R, Mandal M, Lipton A, Harvey H, Thompson specific metastatic potential of individual breast cancer 816–26. CB. Overexpression of HER2 modulates bcl-2, bcl-XL, cells and primary tumors. J Clin Invest 2005;115:44–55. 24. Bryckaert M, Guillonneau X, Hecque C, Courtois Y, and tamoxifen-induced apoptosis in human MCF-7 12. Greenberg PA, Hortobagyi GN, Smith TL, et al. Long- Mascarelli F. Both FGF1 and bcl-x synthesis are breast cancer cells. Clin Cancer Res 1996;2:1215–9. term follow-up of patients with complete remission necessary for the reduction of apoptosis in retinal 36. Salvesen GS, Duckett CS. IAP proteins: blocking following combination chemotherapy for metastatic pigmented epithelial cells by FGF2: role of the the road to death’s door. Nat RevMol Cell Biol 2002; breast cancer. J Clin Oncol 1996;14:2197–205. extracellular signal-regulated kinase 2. Oncogene 1999; 3:401–10. 13. Slamon DJ, Leyland-Jones B, Shak S, et al. Use of 18:7584–93. 37. Deveraux QL, Reed JC. IAP family proteins-suppres- chemotherapy plus a monoclonal antibody against 25. Deng X, Ruvolo P, Carr B, May WS, Jr. Survival sors of apoptosis. Genes Dev1999;13:239–52. HER2 for metastatic breast cancer that overexpresses function of ERK1/2 as IL-3-activated, staurosporine- 38. Mauri D, Pavlidis N, Ioannidis JPA. Neoadjuvant HER2. N Engl J Med 2001;344:783–92. resistant Bcl2 kinases. Proc Natl Acad Sci U S A 2000;97: versus adjuvant systemic treatment in breast cancer: a 14. Mehlen P, Puisieux A. Metastasis: a question of life or 1578–83. meta-analysis. J Natl Cancer Inst 2005;97:188–94. death. Nat RevCancer 2006;6:449–58. 26. Bonni A, Brunet A, West AE, et al. Cell survival 39. Dornhofer N, Spong S, Bennewith K, et al. Connective 15. Glinsky GV, Glinsky VV, Ivanova AB, Hueser CJ. promoted by the Ras-MAPK signaling pathway by tissue growth factor-specific monoclonal antibody ther- Apoptosis and metastasis: increased apoptosis resis- transcription-dependent and -independent mecha- apy inhibits pancreatic tumor growth and metastasis. tance of metastatic cancer cells is associated with the nisms. Science 1999;286:1358–62. Cancer Res 2006;66:5816–27.

www.aacrjournals.org 3491 Cancer Res 2009; 69: (8). April 15, 2009

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 2009 American Association for Cancer Research. Cancer Editor's Note Research Editor's Note: Connective Tissue Growth Factor Confers Drug Resistance in Breast Cancer through Concomitant Up-regulation of Bcl-xL and cIAP1 Ming-Yang Wang, Pai-Sheng Chen, Ekambaranellore Prakash, Hsing-Chih Hsu, Hsin-Yi Huang, Ming-Tsan Lin, King-Jen Chang, and Min-Liang Kuo

The editors were notified by the authors of this article (1) that the a-tubulin Western blot bands in Fig. 4A and the MDA231/AS ERK Western blot bands in Fig. 4D are incorrect. Because satisfactorily corrected figures could not be provided, the editors are publishing this note to alert readers to these concerns.

Reference 1. Wang MY, Chen PS, Prakash E, Hsu HC, Huang HY, Lin MT, et al. Connective tissue growth factor confers drug resistance in breast cancer through concomitant up-regulation of Bcl-xL and cIAP1. Cancer Res 2009;69:3482–91.

Published online July 15, 2019. Cancer Res 2019;79:3792 doi: 10.1158/0008-5472.CAN-19-1714 2019 American Association for Cancer Research.

3792 Cancer Res; 79(14) July 15, 2019 Published OnlineFirst April 7, 2009; DOI: 10.1158/0008-5472.CAN-08-2524

Connective Tissue Growth Factor Confers Drug Resistance in Breast Cancer through Concomitant Up-regulation of Bcl-xL and cIAP1

Ming-Yang Wang, Pai-Sheng Chen, Ekambaranellore Prakash, et al.

Cancer Res Published OnlineFirst April 7, 2009.

Updated version Access the most recent version of this article at: doi:10.1158/0008-5472.CAN-08-2524

Supplementary Access the most recent supplemental material at: Material http://cancerres.aacrjournals.org/content/suppl/2009/04/06/0008-5472.CAN-08-2524.DC1

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 Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/early/2009/04/07/0008-5472.CAN-08-2524.citation. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 2009 American Association for Cancer Research.