TRAP1 Regulates Proliferation, Mitochondrial Function, and Has Prognostic Signiﬁcance in NSCLC

Published OnlineFirst February 24, 2014; DOI: 10.1158/1541-7786.MCR-13-0481

Molecular Cancer Cell Cycle and Senescence Research

Jackeline Agorreta1,3, Jianting Hu3, Dongxia Liu3,6, Domenico Delia7, Helen Turley3, David JP. Ferguson3, Francisco Iborra2,4, María J. Pajares1, Marta Larrayoz1, Isabel Zudaire1, Ruben Pio1, Luis M. Montuenga1, Adrian L. Harris4,5, Kevin Gatter3, and Francesco Pezzella3

Abstract The TNF receptor-associated protein 1 (TRAP1) is a mitochondrial HSP that has been related to drug resistance and protection from apoptosis in colorectal and prostate cancer. Here, the effect of TRAP1 ablation on cell proliferation, survival, apoptosis, and mitochondrial function was determined in non–small cell lung cancer (NSCLC). In addition, the prognostic value of TRAP1 was evaluated in patients with NSCLC. These results demonstrate that TRAP1 knockdown reduces cell growth and clonogenic cell survival. Moreover, TRAP1 downregulation impairs mitochondrial functions such as ATP production and mitochondrial membrane potential as measured by TMRM (tetramethylrhodamine methylester) uptake, but it does not affect mitochondrial density or mitochondrial morphology. The effect of TRAP1 silencing on apoptosis, analyzed by ﬂow cytometry and immunoblot expression (cleaved PARP, caspase-9, and caspase-3) was cell line and context dependent. Finally, the prognostic potential of TRAP1 expression in NSCLC was ascertained via immunohistochemical analysis which revealed that high TRAP1 expression was associated with increased risk of disease recurrence (univariate analysis, P ¼ 0.008; multivariate analysis, HR: 2.554; 95% conﬁdence interval, 1.085–6.012; P ¼ 0.03). In conclusion, these results demonstrate that TRAP1 impacts the viability of NSCLC cells, and that its expression is prognostic in NSCLC. Implications: TRAP1 controls NSCLC proliferation, apoptosis, and mitochondrial function, and its status has prognostic potential in NSCLC. Mol Cancer Res; 12(5); 660–9. 2014 AACR.

Introduction NSCLC cases, respectively (2). Despite the development of Lung cancer is the leading cause of cancer death worldwide targeted therapies in lung cancer, there has been little (1). Non–small cell lung cancer (NSCLC), the most com- improvement in 5-year survival rates. In this context, mon type of lung cancer, can be subdivided into two main improved knowledge of the molecular biology of lung histologic subtypes: adenocarcinoma and squamous cell cancer, together with biomarkers that predict tumor devel- carcinoma (SCC), accounting for 50% and 30% of all opment and prognosis, is needed. TNF receptor-associated protein 1 (TRAP1) is a mitochondrial protein that belongs to the Hsp90 family, first Authors' Affiliations: 1Oncology Division, Center for Applied Medical identified as interacting with the intracellular domain of the Research (CIMA), University of Navarra, Pamplona; 2Department of Molec- ular and Cellular Biology, Centro Nacional de Biotecnología, Consejo type I TNF receptor (3). Later sequence analysis revealed Superior de Investigaciones Científicas, Madrid, Spain; 3Nuffield Depart- that TRAP1 was identical to Hsp75 (4). TRAP1 is mainly ment of Clinical Laboratory Sciences; 4Weatherall Institute of Molecular localized in mitochondria of normal and tumor cells (4, 5) Medicine, University of Oxford, John Radcliffe Hospital; 5Department of Medical Oncology, University of Oxford, The Churchill Hospital, Oxford, acting as a substrate for the serine/threonine kinase PINK1 United Kingdom; 6Department of Rheumatology and Immunology, Shan- (6). Other localizations include the cytosol, endoplasmic dong Provincial Hospital, Shandong University, Jinan, China; and 7Depart- – ment of Experimental Oncology, Fondazione IRCCS Istituto Nazionale reticulum, and nucleus (7 9). TRAP1 interacts with several Tumori, Milano, Italy proteins such as retinoblastoma (10), the ATPase TBP7, a component of the 19S proteasome regulatory subunit (11), Note: Supplementary data for this article are available at Molecular Cancer þ Research Online (http://mcr.aacrjournals.org/). the Ca2 -binding protein sorcin localized in the mitochon- Corresponding Authors: Jackeline Agorreta, Oncology Division, Center dria (7, 12), the mitochondrial protein cyclophilin D (5), for Applied Medical Research (CIMA), University of Navarra, Pio XII 55, and the tumor suppressors EXT1 and EXT2, proteins 31008 Pamplona, Spain. Phone: 34-948-194-700; Fax: 34-948-194-714; involved in hereditary multiple exostoses (13). Moreover, E-mail: [email protected]; and Francesco Pezzella, Nuffield Department of Clinical Laboratory Sciences, University of Oxford, John Radcliffe Hospital, TRAP1 has been reported to protect against apoptosis Headley Way, OX3 9DU Oxford, United Kingdom. Phone: 441865220497; (5, 14, 15) and oxidative stress (15–17). Interestingly, it Fax: 441865220519; E-mail: [email protected] has been proposed that TRAP1 may be involved in che- doi: 10.1158/1541-7786.MCR-13-0481 moresistance by blocking drug-induced apoptosis in a vari- 2014 American Association for Cancer Research. ety of tumors such as prostate cancer (18), osteosarcoma

660 Mol Cancer Res; 12(5) May 2014

TRAP1 Regulates NSCLC Proliferation and Mitochondrial Function

(15), and colorectal cancer (19). In addition, TRAP1 has Table 1. Clinicopathological characteristics of been reported to be upregulated in some tumors (5, 18, 20) the patients and downregulated in others (21). TRAP1 has been proposed as a candidate biomarker in ovarian and prostate cancer (18, 22) and inhibition of TRAP1 is being explored N ¼ 71 as a novel anticancer target (23). In NSCLC, we have Age, y (median–interquartile range) 63 (54–70) previously demonstrated that TRAP1-positive cells have <70 54 (76%) high levels of cell proliferation promoting genes (21), and 70 17 (24%) ﬁ that in the rst hours following hypoxia, in absence of Sex, n (%) TRAP1, retinoblastoma fails to inhibit proliferation (24). Male 60 (85%) However, the biologic role of this mitochondrial heat shock Female 11 (15%) protein (HSP) in NSCLC and its relation with mitochon- Stage, n (%) drial function has not been evaluated yet. Stage I 44 (62%) The aim of the present study was to determine the role of Stage II 15 (21%) TRAP1 on proliferation, cell survival, apoptosis, and mito- Stage III 10 (14%) chondrial function in lung cancer cell lines and to evaluate Stage IV 2 (3%) the prognostic role of TRAP1 in patients with NSCLC. Our Histology, n (%) results demonstrate that TRAP1 downregulation reduces Adenocarcinoma 26 (37%) cell proliferation and survival, induces apoptosis, and impairs SCC 39 (55%) mitochondrial functions such as ATP production and mito- Other 6 (8%) chondrial membrane potential regulation. However, Smoking status, n (%) TRAP1 knockdown does not affect mitochondrial density Current smoker 39 (55%) or mitochondrial morphology. In addition, overexpression Former smoker 26 (37%) of TRAP1 was associated with shorter recurrence-free sur- Nonsmoker 6 (8%) vival (RFS) in patients with NSCLC.

Materials and Methods and antigen retrieval was carried out by pressure cooking Cells in 10 mmol/L citrate buffer, pH 6. Nonspecificbinding Human NSCLC cell lines NCI-A549 and NCI-H1299 was blocked using 5% normal goat serum in Tris-buffered were obtained from Clare Hall Laboratories and grown in saline for 30 minutes. Sections were incubated with anti- Dulbecco's Modified Eagle Medium supplemented with TRAP1 antibody (1:400; Labvision) overnight at 4C. 10% FBS and penicillin-streptomycin at 100 U/mL. Cell Sections were then incubated with Envision polymer fi cultures were incubated at 37 C in a humidi ed 5% CO2 (Dako) for 30 minutes at room temperature. Peroxidase incubator. activity was developed using diaminobenzidine and coun- terstained with hematoxylin before mounting in DPX Patient samples medium (BDH Chemical). The specificity of TRAP1 A series of 71 patients with a diagnosis of NSCLC who antibody was demonstrated using a variety of controls, underwent surgical resection at Clínica Universidad de including Western blot analysis, inhibition with TRAP1- Navarra (Navarra, Spain) from 2000 through 2008 were siRNA sequences, isotype control, and omission of the included in this study. Clinicopathologic features of the primary antibody. patients are listed in Table 1. Tumor specimens were classified according to the 2004 World Health Organi- Immunostaining evaluation zation criteria (25). The inclusion criteria were NSCLC Two independent, blinded observers (F. Pezzella and histology, no neoadjuvant chemo- or radiotherapy, and J. Agorreta) evaluated the intensity and extensiveness of absence of cancer within the 5 years previous to lung staining in all of the study samples. The evaluation of cancer surgery. The study protocol was approved by the cytoplasmic TRAP1 expression was performed using the institutional medical ethical committee. Written H-score system (26). Briefly, the percentage of positive cells informed consent was obtained from each patient before (0%–100%) and the intensity of staining (1þ, mild; 2þ, participation. RFS was calculated from the date of surgery moderate; and 3þ, intense labeling) were scored. Disagree- to the date of detection of recurrence or the date of the ments were resolved by common reevaluation. last follow-up. The median follow-up time was 42 months. Immunoblotting Protein and total RNA were extracted using Paris kit Immunohistochemistry in clinical specimens from (Ambion-Life Technologies Ltd) according to the manu- patients with NSCLC facturer's instructions. Thirty micrograms of total protein Formalin-fixed paraffin-embedded tissue sections were from each lysate were boiled at 95 C for 5 minutes, separated evaluated. Endogenous peroxidase activity was quenched by SDS/PAGE under reduced conditions (5% 2-

www.aacrjournals.org Mol Cancer Res; 12(5) May 2014 661

Agorreta et al.

mercaptoethanol), and transferred onto a nitrocellulose harvested and fixed overnight in 4% phosphate-buffered membrane. The membranes were subsequently blocked in formalin (pH 7.0), suspended in agar, and embedded in 5% defatted milk-PBS for 1 hour and incubated overnight at paraffin. Antigen retrieval was carried out in 3 mmsections 4C with a primary antibody anti TRAP1 (1:1000, Labvi- by pressure cooking in 10 mmol/L citrate buffer, pH 6, sion) or anti b-actin (1:10000, Sigma). Blots were then and immunohistochemical staining for the human Ki-67 incubated with a horseradish peroxidase-linked secondary protein was performed using the anti-MIB1 antigen anti- antibody (1:5000; Amersham Pharmacia Biotech) and body (Dako) at 1:50 for 30 minutes at room temperature. developed by chemoluminiscence with Lumilight Plus Kit Sections were incubated with the Envision detection (Roche diagnostics). Apoptosis detection by Western blot- system (Dako) and developed with diaminobenzidine. ting was performed as described before (27). Immunohistochemical scoring was performed as previously described (29). RNA interference For inhibition of TRAP1 expression, cells were seeded Cell-cycle and apoptosis analysis (1 106 cells per well) in 10 cm dishes in antibiotic-free Cell-cycle analyses were performed on trypsin-disaggre- medium. At 24 hours, cells were transfected with 40 nmol/L gated cryopreserved cell suspensions containing floating and of siRNA by using Oligofectamine (Invitrogen-Life Tech- attached cells. Following thawing, cells were centrifuged to nologies Ltd) as previously described (21). Two siRNA remove the cryopreservation solution (10% dimethyl sulf- sequences against TRAP1 were designed and synthesized by oxide in FBS), fixed in 70% ethanol on ice, treated with 1 Eurogentec (TRAP1-siRNA1: 50-AUGUUUGGAAGUG- mg/mL RNase, stained with 10 mg/mL propidium iodide, GAACCC-30 and 50-ACCAUCUGAAAGCCACUGG-30; and examined with a FACSCalibur instrument fitted with a TRAP1-siRNA2: 50-TGCTGTTTGGAAGTGGAACCC- Cell Quest software package (BD Biosciences). About TGCACGTTTTGGCCACTGACTGACGTGCAGGGC- 50,000 cells per sample were analyzed. Percentages of cells 0 0 – CACTTCCAAA-3 and 5 -CCTGTTTGGAAGTGGCC- in the Sub-G1,G1, S and G2 M phases were determined. CTGCACGTCAGTCAGTGGCCAAAACGTGCAGGG- For apoptosis analysis, fresh trypsin-disaggregated cell sus- TTCCACTTCCAAAC-30). A scrambled (scr) siRNA (50- pensions containing floating and attached cells were used as AUGUUUGGAAGUGGAACCC-30 and 50-UAGGGU- previously described (30). Briefly, cells were washed and GUACCCGUAAUAG-30) was used as the negative control. stained with 2 mL of Annexin V (BD Biosciences) and 2 mL of 10 mg/mL of propidium iodide (Sigma). Apoptosis was RT-PCR induced by staurosporine treatment (1 mg/mL for 4 h). Retrotranscription was performed using RetroScript Kit Samples were analyzed on a FACSCalibur instrument and (Ambion). TRAP1 and b-actin expression was analyzed by quadrant analysis was performed with FlowJo 9.3 software PCR using TaqMan Gene Expression Assays (Applied (Tree Star). At least three independent experiments per Biosystems). The reaction was performed on a PTC-200 condition were performed. thermal cycler with a Chromo 4 continuous fluorescence C detector (Bio-Rad). The comparative cycle threshold ( t) Mitochondrial function method was used to analyze the data by generating relative The amount of ATP was measured in lysates of 105 cells values of the amount of target cDNA, according to the using the ATP Bioluminescence Assay Kit (Roche) in DDC 2 t method (28) using b-actin as endogenous gene and accordance with the manufacturer's instructions. This meth- scramble (scr) expression as calibrator. od uses the ATP dependency of the light-emitting, lucifer- ase-catalyzed oxidation of luciferin for the measurement of Growth curves ATP concentration. To analyze the mitochondrial mem- Cells were seeded on 6-well dishes at a density of 1 105 brane potential, TMRM (tetramethylrhodamine methyles- cells per well in triplicate and cultured for 1 to 7 days. ter; Invitrogen) staining was used because it is a cell-per- Subsequently, cell number was assessed with a Coulter Z2 meant, cationic, red-orange fluorescent dye that is readily particle count and size analyzer (Beckman Coulter). Auto- sequestered by active mitochondria. MitoTracker Green matically cell count was carried out with a Cell IQ micro- staining (Molecular Probes-Life Technologies Ltd) was also scope (Chipman Technologies). used to measure mitochondrial mass regardless of mitochondrial membrane potential. Moreover, the production of Clonogenic assay reactive oxygen species (ROS) was evaluated by the Mito- Twenty four hours after siRNA transfection, cells were SOX staining (Molecular probes) as previously described harvested, seeded in triplicate (300 cells per well) in 6-well (31). Fluorescence images were collected using a confocal fl plates, and incubated at 37 C in a 5% CO2 atmosphere. microscope (Zeiss LSM 510 META; Carl Zeiss) and uo- After 14 days, colonies were fixed in methanol-acetic acid rescence intensity was measured with ImageJ software (NIH, (1:1), stained with crystal violet, and counted. Bethesda, MD).

Proliferation index determination Electron microscopy siRNA-treated cells were seeded in 10 cm dishes and Cells were ﬁxed in 4% glutaraldehyde in 0.1 mol/L grown for 1, 3, or 5 days. Subsequently, cells were phosphate buffer and processed for routine electron

662 Mol Cancer Res; 12(5) May 2014 Molecular Cancer Research

TRAP1 Regulates NSCLC Proliferation and Mitochondrial Function

microscopy as previously described (32). Mitochondrial examined by testing interactions between the covariates mass was measured with ImageJ software. of the final model and time. P < 0.05 was considered statistically significant. Statistical analysis Statistical analysis was performed using SPSS 15.0. Data Results obtained from cell count, colony formation, MIB1 stain- Expression of TRAP1 is necessary for cell growth ing, cell cycle, and mitochondrial function experiments To examine the effect of TRAP1 inhibition on cell were analyzed by the Student t test or the Mann–Whitney proliferation, we carried out downregulation experiments U test for parametric and nonparametric variables, respec- in lung cancer cell lines. Knockdown was carried out in tively. For survival analysis, Kaplan–Meier survival curves H1299 and A549 cells using two different siRNAs and the and the log-rank test were used to analyze differences in efficacy of TRAP1 siRNA downregulation was verified by RFS (the median was selected as the cutoff value). Mul- Western blotting and real-time PCR (RT-PCR; Fig. 1A and tivariate analysis was carried out using the Cox propor- B). TRAP1 downregulation resulted in a significant reduc- tional hazards model. Only variables of P < 0.1 from the tion in cell growth in both H1299 and A549 cell lines as univariate analysis were entered in the Cox regression confirmed by TRAP1-siRNA1 and 2 sequences (Fig. 1C). analysis. The proportional hazards assumption was Cell growth of TRAP1-siRNA1–treated A549 cells was also

Figure 1. TRAP1 knockdown inhibits cell proliferation and survival on the H1299 and A549 cell lines. Successful knockdown of TRAP1 expression by two independent TRAP1-siRNA sequences was demonstrated by RT-PCR (A) and Western blot analysis (B) in both H1299 and A549 cell lines at day 4. To determine the effect of TRAP1 siRNA knockdown on tumor cell proliferation, cells were transfected with control- (scr) or TRAP1-siRNAs and cell number was determined by a Coulter Z2 particle count and size analyzer (C) or automatically determined by a Cell IQ microscope (D). E, TRAP1 downregulation signiﬁcantly reduced colony formation in the A549 and H1299 cell lines. Data, mean SD from at least three independent experiments.

www.aacrjournals.org Mol Cancer Res; 12(5) May 2014 663

Agorreta et al.

monitored by time-lapse video microscopy for 5 days at a TRAP1 downregulation has a variable effect on 35- minute interval, confirming the reduction in cell apoptosis number after TRAP1 knockdown (Fig. 1D and Supplemen- Quantification of apoptotic cells by Annexin V/PI assay tary Movies S1 and S2). The impairment of cell survival showed that TRAP1 downregulated A549 cells had was further confirmed by clonogenic assay in H1299 and increased apoptotic rates as compared with scr-siRNA A549 cell lines (Fig. 1E). We next investigated the effect of treated (Fig. 2D, top). Those effects were less evident in TRAP1 knockdown on cell proliferation by staining cell H1299 cell line (Supplementary Fig. S1). The induction pellets of scr- and TRAP1-siRNA A549-treated cells at of apoptosis in A549 cells was confirmed by the increase different time points for ki-67 protein (MIB1 antigen). of activated (cleaved) caspase-3, caspase-9, and PARP We found that from day 3, there was a significant reduction (Fig. 2E, left). It should be noted that when apoptosis of MIB1-positive cells when TRAP1 was inhibited (Fig. 2A). was induced by treating cells with staurosporine, there was Cell-cycle analysis by flow cytometry showed a significant a dramatic induction of apoptosis in TRAP1-siRNA– – reduction in the percentage of cells in G2 M phase after treated cells but not in scr-control cells (Fig. 2D, bottom TRAP1 knockdown (Fig. 2B and C), confirming the andFig.2E,right).ThecelllineH1299didnotshow results from the immunohistochemical analysis of ki-67 clear evidence of apoptosis as silencing of TRAP produced expression. amildincreaseofPARPbutasimilarlymilddecreaseof

Figure 2. Downregulation of TRAP1 arrests cell proliferation and induces apoptosis in the A549 lung cancer cell line. A, proliferative fraction given by the percentage of ki-67–positive cells was signiﬁcantly reduced in TRAP1-siRNA1–treated cells. B, cell-cycle distribution of A549 cells at different days after TRAP1-siRNA1 transfection. C, differences in the percentage of cells in S and G2–M phases at day 3. D, Annexin V/propidium iodide staining was performed on A549 cells transfected with scr, TRAP1-siRNA1, and TRAP1-siRNA2 sequences (top) or transfected cells treated with 1 mg/mL þ staurosporine (bottom) and analyzed by ﬂow cytometry. Percentages of intact cells (Annexin V PI ), early apoptotic cells (Annexin V PI )andlate þ þ apoptotic or necrotic cells (Annexin V PI ) are shown in the plot. One representative experiment is shown from three independent repetitions. E, apoptosis was also demonstrated by Western blot analysis of cleaved PARP and cleaved caspase-3 and -9 in A549 cells. Data, mean SD from at least three independent experiments.

664 Mol Cancer Res; 12(5) May 2014 Molecular Cancer Research

TRAP1 Regulates NSCLC Proliferation and Mitochondrial Function

Figure 3. TRAP1 downregulation impairs mitochondrial function in A549 cells. A, measurement of ATP levels by a bioluminescence assay demonstrates that TRAP1 inhibition reduces ATP levels. B, representative images of TMRM uptake in scr- and TRAP1-siRNA1–treated cells. C, quantification of TMRM uptake by image analysis shows a reduction in mitochondrial membrane potential in TRAP1-siRNA1–treated A549 cells as compared with control cells. D, mitochondrial mass measured by MitoTracker staining was similar in scr and TRAP1-siRNA1–treated cells. E, no differences in ROS production, as determined by MitoSOX staining, were found. F, the ultrastructure of the mitochondria was visualized using transmission electron microscopy, andno differences in the mitochondrial morphology were observed between scr- and TRAP1-siRNA1–treated cells. Moreover, no significant differences in mitochondrial mass were found. Scale bar, 500 nm. Data, mean SD from three independent experiments. cleaved caspase-9 while there was no evidence of caspase-3 reduction on membrane potential was shown in TRAP1- activation (Supplementary Fig. S1). siRNA1–treated cells as compared with scr-siRNA (P < No morphologic evidence of accumulation of apoptotic 0.001; Fig. 3B and C). However, no differences were found bodies could be seen in the cell culture movies (Supplemen- in the mitochondrial mass measured by MitoTracker stain- tary Movies). ing (P ¼ 0.809; Fig. 3D) or in ROS production measured by MitoSOX staining (P ¼ 0.078; Fig. 3E). Electron micros- TRAP1 downregulation impairs mitochondrial function copy also did not demonstrate changes in mitochondrial We hypothesized that the effects of TRAP1 inhibition morphology or mitochondrial mass (Fig. 3F). could be caused by mitochondrial dysfunction because TRAP1 is known to be mainly expressed in the mitochon- High TRAP1 expression is associated with worse dria. Therefore, we analyzed a variety of mitochondrial prognosis in patients with NSCLC functions after TRAP1 inhibition. First, we analyzed ATP TRAP1 expression was analyzed by immunohistochem- production in scr- and TRAP1-siRNA1–treated cells, show- istry in a series of 71 patients with NSCLC. The main clinical ing that TRAP1 inhibition was associated with a 30% and pathologic characteristics of these patients are summa- reduction of ATP (P ¼ 0.002; Fig. 3A). Next, we examined rized in Table 1. Moderate TRAP1 staining was found in the effect of TRAP1 expression on mitochondrial membrane normal bronchial mucosa adjacent to the tumor (Fig. 4A), potential as measured by TMRM uptake. A significant although no immunoreactivity was found in alveoli (Fig.

www.aacrjournals.org Mol Cancer Res; 12(5) May 2014 665

Agorreta et al.

A B

C D Figure 4. High cytoplasmic TRAP1 staining is associated with adverse prognosis in NSCLC. Representative TRAP1 immunostaining in bronchial epithelium (A), lung parenchyma (B), lung adenocarcinoma (C), and SCC of the lung (D). E, Kaplan– Meier RFS curves for TRAP1 expression and log-rank test. Shorter RFS time was found in tumors bearing high TRAP1 E expression. Scale bar, 50 mm. 100

80 Low expression

20 High expression P = 0.008 % Recurrence-free survival 0 0 20 40 60 80 100 120 Follow-up time (months)

4B). In tumor samples, TRAP1 staining was observed model. The Cox regression analysis revealed that high predominantly in the cytoplasm of all tumors analyzed (Fig. cytoplasmic TRAP1 expression was an independent predic- 4C). In some cases, TRAP1 staining was both in the nucleus tor of shorter RFS when patients were adjusted by stage [HR and in the cytoplasm (Fig. 4D). ¼ 2.554; conﬁdence interval (CI), 1.085–6.012; Table 2]. We next sought to evaluate the prognostic role of TRAP1 expression in our cohort of NSCLC. Nuclear expression of P ¼ TRAP1 did not correlate with prognosis (log-rank test; Table 2. Multivariate Cox regression analysis of 0.486). However, when cytoplasmic TRAP1 expression was RFS in patients with NSCLC analyzed, those patients with high TRAP1 expression (using the median as the cutoff value) showed shorter RFS than patients with low levels of TRAP1 (P ¼ 0.008; Fig. 4E). HR (95% CI) P Multivariate analysis using Cox regression model was per- Cytoplasmic TRAP1 expression formed to determine the independent prognostic factors. Low —— Relevant clinicopathological variables such as age, gender, High 2.554 (1.085–6.012) P ¼ 0.032 stage, histology, and smoking history, as well as cytoplasmic Stage TRAP1 expression, were analyzed in the Cox univariate I, II —— model, and only those variables with a P value < 0.1 (stage III, IV 1.720 (0.725–4.077) P ¼ 0.218 and TRAP1 expression) were included in the multivariate

666 Mol Cancer Res; 12(5) May 2014 Molecular Cancer Research

TRAP1 Regulates NSCLC Proliferation and Mitochondrial Function

These results demonstrate that TRAP1 expression correlates lines (e.g., A549) but an intermediate in others (e.g., with poor outcome in patients with NSCLC. H1299). Moreover, we have demonstrated herein that TRAP1 Discussion inhibition leads to a reduction of ATP production and In the present study, we have demonstrated that TRAP1 mitochondrial membrane potential. These findings are con- has important effects on mitochondrial function and plays a sistent with previous results indicating that TRAP1 in cancer key role in the regulation of proliferation, survival, and prevents mitochondrial damage (33) and after its inhibition, apoptosis of NSCLC cells. Moreover, we have shown that the misfolding of Peptidylprolyl isomerases D results in high cytoplasmic TRAP1 expression is associated with worse loss of ATP production (5) and are also consistent with its prognosis in patients with NSCLC. protective role from apoptosis. The role of TRAP1 in During the last 10 years, an increasing number of studies maintaining ATP levels has also been reported in rat brain have demonstrated the versatility of TRAP1 protein and its (43). In contrast, two recent papers (40, 44) report that loss involvement in a number of pathways. Originally cloned of TRAP1 results in an increase of ATP production follow- because of its interaction with TNF receptor, and therefore ing a switch from glycolysis to oxidative phosphorylation. likely to be involved in cell signaling (3, 4), it was then found We cannot provide an explanation for this discrepancy as that TRAP1 also acts as a chaperon to retinoblastoma, far as the role of TRAP1 in ATP production is concerned, maintaining retinoblastoma protein in its active conforma- but it is likely to be with the different types of final effects tion (10). The importance of the role of TRAP1 as chaperon that TRAP1 can have according to the type of cell being has quickly outgrown its original role in retinoblastoma investigated. as its pivotal role in cytoprotection has emerged. TRAP1 We have previously demonstrated that after a short has been described to protect mitochondria from oxidative hypoxic shock, TRAP1 translocates to the nucleus and is stress and ROS (reviewed in refs. 33–35). In this sense, essential to maintain retinoblastoma suppressor gene func- TRAP1 blocks ROS activity (15), ROS production (36), and tion: in its absence, the cells fail to slow down proliferation regulates the mitochondrial permeability transition pores (24). However, this is a short-term effect and, in agreement (5, 37). It has been recently demonstrated that TRAP1 has with our previous results and as demonstrated here, in an important role controlling central metabolic networks normoxia and in the longer term TRAP1 promotes the cell in the mitochondria of tumor cells (38, 39). All these cycle. Therefore, it seems to have two opposite functions: in functions are believed to be important for its role in protect- normoxia, TRAP1 promotes cell proliferation and protects ing from apoptosis and inducing chemoresistance (11, 33). from apoptosis; however, following a hypoxic shock it moves In the present study, we investigated the role of TRAP1 in to the nucleus where it is essential for retinoblastoma to NSCLC cell lines growth by looking at its effects on cell induce a rapid, short-term slowing down of proliferation. If proliferation, apoptosis, and mitochondrial function. hypoxia persists, TRAP1 cytoplasmic levels will decrease Downregulation of TRAP1 expression in NSCLC cell (unpublished results) alongside a diminution of the prolif- lines produced a significant reduction of cell proliferation eration rate. In this respect, it can be considered to have and survival as assessed by cell count, clonogenic assays, ki- oncogenic properties as suggested by Sciacovelli and collea- 67 expression, and cell-cycle analysis. In agreement with gues (40) although the potential suppressor effect in malig- these results, proliferation had been previously linked to nancies, due to its chaperone role to retinoblastoma (10), TRAP1 by our group (24) and others (40). However, there is remains to be elucidated. no general agreement in literature as some authors have failed On the other hand, TRAP1 expression has also been to notice any effect on cell growth (38). Moreover, we have correlated with chemoresistance in breast, colorectal, and determined that TRAP1 knockdown is associated with a ovarian carcinoma (7, 19, 43, 45). Although there is con- subtle and variable effect on induction of apoptosis. This sensus that TRAP1 expression is a predictive biomarker for induction becomes more evident when apoptosis is induced drug resistance, because of the broad range of cellular func- pharmacologically with staurosporine. Our results have tions influenced by TRAP1 (33), it is perhaps not surprising demonstrated a variable role in antiapoptotic functions of that its role as immunohistochemical prognostic biomarker TRAP1 in NSCLC cell lines with clear signs of apoptosis is controversial. As a matter of fact, a role as oncogene (40), detected in one of the cell lines studied. These results are still together with a role as tumor suppressor gene (44), has been in agreement with previous reports showing a protective proposed. Our data obtained from lung primary tumors effect of TRAP1 against apoptosis-inducing anticancer drugs showed both nuclear and cytoplasmic staining of TRAP1 in or ROS (12, 14, 15, 18, 19). Having taken all these data into tumor cells, as it was previously reported in NSCLC, breast account, we suggest that in many types of cells, loss of carcinoma, ovarian cancer, and other malignancies (18, TRAP1 has a mitochondrial priming effect which tips the 24, 45). Only a few studies have been performed looking at cells toward apoptosis making them more sensitive to the TRAP1 expression on tumor samples and correlations with effects of cytotoxic drugs (41, 42). The observation that prognosis (24, 45, 46), and the results are controversial. In apoptosis is obvious in one of the two cell lines analyzed, colorectal carcinoma, high TRAP1 expression has been cor- independently from the pharmacologic stimulation, further related with shorter RFS and overall survival (OS; ref. 46). support the protective role of TRAP1 on apoptosis as its However, in a study on ovarian carcinoma, cytoplasmic downregulation can be the last step of the priming in some expression of TRAP1 was associated with better OS, whereas

www.aacrjournals.org Mol Cancer Res; 12(5) May 2014 667

Agorreta et al.

no association was found with RFS (45). Finally, in a series of Disclosure of Potential Conflicts of Interest breast carcinoma, we found no correlation between cyto- No potential conflicts of interest were disclosed. plasmic TRAP1 and OS or RFS, although nuclear TRAP1 expression was instead associated with both retinoblastoma Authors' Contributions positivity and longer RFS but no association was found with Conception and design: J. Agorreta, J. Hu, A.L. Harris, K. Gatter, F. Pezzella OS (24). In the present study, we demonstrated that cyto- Development of methodology: J. Agorreta, J. Hu, H. Turley, F. Iborra, F. Pezzella Acquisition of data (provided animals, acquired and managed patients, provided plasmic expression of TRAP1 is an independent predictor of facilities, etc.): J. Agorreta, J. Hu, D. Delia, D.J.P. Ferguson, F. Iborra, M.J. Pajares, the shortest relapse-free survival in surgically resected L.M. Montuenga, A.L. Harris, F. Pezzella NSCLC. To our knowledge, this is the first study that Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): J. Agorreta, J. Hu, D.J.P. Ferguson, I. Zudaire, R. Pio, analyzes the role of TRAP1 in the prognosis of lung cancer. L.M. Montuenga, A.L. Harris, F. Pezzella Although the causes underlying diverse results in different Writing, review, and/or revision of the manuscript: J. Agorreta, J. Hu, D. Liu, tumor types are still unknown, we may argue that TRAP1 D. Delia, H. Turley, M.J. Pajares, M. Larrayoz, I. Zudaire, R. Pio, L.M. Montuenga, fi A.L. Harris, K. Gatter, F. Pezzella plays organ-speci c roles in each tumor type. Indeed, it Administrative, technical, or material support (i.e., reporting or organizing data, should be noted that the association between cytoplasmic constructing databases): J. Agorreta, M. Larrayoz, K. Gatter, F. Pezzella staining and worse prognosis demonstrated in NSCLC in Study supervision: J. Agorreta, J. Hu, A.L. Harris, F. Pezzella the present study and in colorectal carcinoma (46) is fully consistent with the suggested role of TRAP1 in cell prolif- Acknowledgments eration and protection to apoptosis. However, more extensive The authors thank G. Steers, Drs. R. Leek, K. Giaslakiotis, H. Mellor and studies exploring the role of TRAP1 as a potential target or S. Wigfield for technical support. predictor of response in NSCLC are warranted. In conclusion, the complexity of the role played by TRAP1 in the cancer cell biology is the most likely expla- Grant Support This work was supported by Cancer Research UK grants and "UTE project nation for the discrepant results observed in literature; CIMA". J. Agorreta was funded by the Sara Borrell Program of ISCIII, Spanish however, our data further support the role played by TRAP1 Government. The costs of publication of this article were defrayed in part by the payment in mitochondrial function and regulation of apoptosis, of page charges. This article must therefore be hereby marked advertisement but also demonstrate a role in cell growth through the in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. regulation of the cell cycle. Furthermore, we have demonstrated that high TRAP1 expression is an adverse prognostic Received September 10, 2013; revised January 27, 2014; accepted January 30, factor for patients with NSCLC. 2014; published OnlineFirst February 24, 2014.

References 1. Siegel R, Naishadham D, Jemal A. Cancer statistics, 2013. CA Cancer blastoma protein during mitosis and after heat shock. Mol Cell Biol J Clin 2013;63:11–30. 1996;16:4691–9. 2. Perez-Moreno P, Brambilla E, Thomas R, Soria JC. Squamous cell 11. Amoroso MR, Matassa DS, Laudiero G, Egorova AV, Polishchuk RS, carcinoma of the lung: molecular subtypes and therapeutic opportu- Maddalena F, et al. TRAP1 and the proteasome regulatory particle nities. Clin Cancer Res 2012;18:2443–51. TBP7/Rpt3 interact in the endoplasmic reticulum and control cellular 3. Song HY, Dunbar JD, Zhang YX, Guo D, Donner DB. Identification of a ubiquitination of specific mitochondrial proteins. Cell Death Differ protein with homology to hsp90 that binds the type 1 tumor necrosis 2012;19:592–604. factor receptor. J Biol Chem 1995;270:3574–81. 12. Landriscina M, Laudiero G, Maddalena F, Amoroso MR, Piscazzi A, 4. Felts SJ, Owen BA, Nguyen P, Trepel J, Donner DB, Toft DO. The Cozzolino F, et al. Mitochondrial chaperone Trap1 and the calcium hsp90-related protein TRAP1 is a mitochondrial protein with distinct binding protein Sorcin interact and protect cells against apoptosis functional properties. J Biol Chem 2000;275:3305–12. induced by antiblastic agents. Cancer Res 2010;70:6577–86. 5. Kang BH, Plescia J, Dohi T, Rosa J, Doxsey SJ, Altieri DC. Regulation 13. Simmons AD, Musy MM, Lopes CS, Hwang LY, Yang YP, Lovett M. A of tumor cell mitochondrial homeostasis by an organelle-specific direct interaction between EXT proteins and glycosyltransferases is Hsp90 chaperone network. Cell 2007;131:257–70. defective in hereditary multiple exostoses. Hum Mol Genet 1999; 6. Pridgeon JW, Olzmann JA, Chin LS, Li L. PINK1 protects against 8:2155–64. oxidative stress by phosphorylating mitochondrial chaperone TRAP1. 14. Masuda Y, Shima G, Aiuchi T, Horie M, Hori K, Nakajo S, et al. PLoS Biol 2007;5:e172. Involvement of tumor necrosis factor receptor-associated protein 1 7. Maddalena F, Sisinni L, Lettini G, Condelli V, Matassa DS, Piscazzi A, (TRAP1) in apoptosis induced by beta-hydroxyisovalerylshikonin. et al. Resistance to paclitxel in breast carcinoma cells requires a quality J Biol Chem 2004;279:42503–15. control of mitochondrial antiapoptotic proteins by TRAP1. Mol Oncol 15. MontesanoGesualdiN,ChiricoG,PirozziG,CostantinoE,LandriscinaM, 2013;7:895–906. Esposito F. Tumor necrosis factor-associated protein 1 (TRAP-1) pro- 8. Cechetto JD, Gupta RS. Immunoelectron microscopy provides evi- tects cells from oxidative stress and apoptosis. Stress 2007;10:342–50. dence that tumor necrosis factor receptor-associated protein 1 (TRAP- 16. Im CN, Lee JS, Zheng Y, Seo JS. Iron chelation study in a normal 1) is a mitochondrial protein which also localizes at specific extra- human hepatocyte cell line suggests that tumor necrosis factor recep- mitochondrial sites. Exp Cell Res 2000;260:30–9. tor-associated protein 1 (TRAP1) regulates production of reactive 9. Takemoto K, Miyata S, Takamura H, Katayama T, Tohyama M. Mito- oxygen species. J Cell Biochem 2007;100:474–86. chondrial TRAP1 regulates the unfolded protein response in the 17. Voloboueva LA, Duan M, Ouyang Y, Emery JF, Stoy C, Giffard RG. endoplasmic reticulum. Neurochem Int 2011;58:880–7. Overexpression of mitochondrial Hsp70/Hsp75 protects astrocytes 10. Chen CF, Chen Y, Dai K, Chen PL, Riley DJ, Lee WH. A new member of against ischemic injury in vitro. J Cereb Blood Flow Metab 2008;28: the hsp90 family of molecular chaperones interacts with the retino- 1009–16.

668 Mol Cancer Res; 12(5) May 2014 Molecular Cancer Research

TRAP1 Regulates NSCLC Proliferation and Mitochondrial Function

18. Leav I, Plescia J, Goel HL, Li J, Jiang Z, Cohen RJ, et al. Cytoprotective 32. Sivridis E, Koukourakis MI, Zois CE, Ledaki I, Ferguson DJ, Harris AL, mitochondrial chaperone TRAP-1 as a novel molecular target in et al. LC3A-positive light microscopy detected patterns of autophagy localized and metastatic prostate cancer. Am J Pathol 2010;176: and prognosis in operable breast carcinomas. Am J Pathol 2010; 393–401. 176:2477–89. 19. Costantino E, Maddalena F, Calise S, Piscazzi A, Tirino V, Fersini A, 33. Altieri DC, Stein GS, Lian JB, Languino LR. TRAP-1, the mitochondrial et al. TRAP1, a novel mitochondrial chaperone responsible for multi- Hsp90. Biochim Biophys Acta 2012;1823:767–73. drug resistance and protection from apoptotis in human colorectal 34. Kang BH. TRAP1 regulation of mitochondrial life or death decision in carcinoma cells. Cancer Lett 2009;279:39–46. cancer cells and mitochondria-targeted TRAP1 inhibitors. BMB Rep 20. Fang W, Li X, Jiang Q, Liu Z, Yang H, Wang S, et al. Transcriptional 2012;45:1–6. patterns, biomarkers and pathways characterizing nasopharyngeal 35. Matassa DS, Amoroso MR, Maddalena F, Landriscina M, Esposito F. carcinoma of Southern China. J Transl Med 2008;6:32. New insights into TRAP1 pathway. Am J Cancer Res 2012;2:235–48. 21. Liu D, Hu J, Agorreta J, Cesario A, Zhang Y, Harris AL, et al. Tumor 36. Hua G, Zhang Q, Fan Z. Heat shock protein 75 (TRAP1) antagonizes necrosis factor receptor-associated protein 1(TRAP1) regulates genes reactive oxygen species generation and protects cells from granzyme involved in cell cycle and metastases. Cancer Lett 2010;296:194–205. M-mediated apoptosis. J Biol Chem 2007;282:20553–60. 22. Landriscina M, Amoroso MR, Piscazzi A, Esposito F. Heat shock 37. Xiang F, Huang YS, Shi XH, Zhang Q. Mitochondrial chaperone tumour proteins, cell survival and drug resistance: the mitochondrial chaper- necrosis factor receptor-associated protein 1 protects cardiomyo- one TRAP1, a potential novel target for ovarian cancer therapy. cytes from hypoxic injury by regulating mitochondrial permeability Gynecol Oncol 2010;117:177–82. transition pore opening. FEBS J 2010;277:1929–38. 23. Siegelin MD. Inhibition of the mitochondrial Hsp90 chaperone 38. Caino MC, Chae YC, Vaira V, Ferrero S, Nosotti M, Martin NM, et al. network: a novel, efﬁcient treatment strategy for cancer? Cancer Metabolic stress regulates cytoskeletal dynamics and metastasis of Lett 2013;333:133–46. cancer cells. J Clin Invest 2013;123:2907–20. 24. Hu J, Tan EY, Campo L, Leek R, Seman Z, Turley H, et al. TRAP1 is 39. Chae YC, Angelin A, Lisanti S, Kossenkov AV, Speicher KD, Wang H, involved in cell cycle regulated by retinoblastoma susceptibility gene et al. Landscape of the mitochondrial Hsp90 metabolome in tumours. (RB1) in early hypoxia and has variable expression patterns in human Nat Commun 2013;4:2139. tumors. Journal of Cancer Research Updates 2013;2:194–210. 40. Sciacovelli M, Guzzo G, Morello V, Frezza C, Zheng L, Nannini N, et al. 25. Travis WD, Brambilla E, Muller-Hermelink€ HK, Harris CC. Tumours The mitochondrial chaperone TRAP1 promotes neoplastic growth by of the Lung, Pleura, Thymus and Heart. Lyon, France: IARC Press; inhibiting succinate dehydrogenase. Cell Metab 2013;17:988–99. 2004. 41. Ni Chonghaile T, Sarosiek KA, Vo TT, Ryan JA, Tammareddi A, Moore 26. Pajares MJ, Agorreta J, Larrayoz M, Vesin A, Ezponda T, Zudaire I, et al. Vdel G, et al. Pretreatment mitochondrial priming correlates with Expression of tumor-derived vascular endothelial growth factor and its clinical response to cytotoxic chemotherapy. Science 2011;334: receptors is associated with outcome in early squamous cell carcino- 1129–33. ma of the lung. J Clin Oncol 2012;30:1129–36. 42. Sarosiek KA, Ni Chonghaile T, Letai A. Mitochondria: gatekeepers of 27. Cossu F, Milani M, Vachette P, Malvezzi F, Grassi S, Lecis D, et al. response to chemotherapy. Trends Cell Biol 2013;23:612–9. Structural insight into inhibitor of apoptosis proteins recognition by a 43. Chien WL, Lee TR, Hung SY, Kang KH, Lee MJ, Fu WM. Impairment of potent divalent smac-mimetic. PLoS ONE 2012;7:e49527. oxidative stress-induced heme oxygenase-1 expression by the defect 28. Livak KJ, Schmittgen TD. Analysis of relative gene expression data of Parkinson-related gene of PINK1. J Neurochem 2011;117:643–53. using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. 44. Yoshida S, Tsutsumi S, Muhlebach G, Sourbier C, Lee MJ, Lee S, et al. Methods 2001;25:402–8. Molecular chaperone TRAP1 regulates a metabolic switch between 29. Sington J, Giatromanolaki A, Campo L, Turley H, Pezzella F, Gatter KC. mitochondrial respiration and aerobic glycolysis. Proc Natl Acad Sci BNIP3 expression in follicular lymphoma. Histopathology 2007;50: U S A 2013;110:E1604–12. 555–60. 45. Aust S, Bachmayr-Heyda A, Pateisky P, Tong D, Darb-Esfahani S, 30. de Miguel FJ, Sharma RD, Pajares MJ, Montuenga LM, Rubio A, Pio R. Denkert C, et al. Role of TRAP1 and estrogen receptor alpha in patients Identiﬁcation of alternative splicing events regulated by the oncogenic with ovarian cancer: A study of the OVCAD consortium. Mol Cancer factor SRSF1 in lung cancer. Cancer Res 2014;74:1105–15. 2012;11:69. 31. das Neves RP, Jones NS, Andreu L, Gupta R, Enver T, Iborra FJ. 46. Gao JY, Song BR, Peng JJ, Lu YM. Correlation between mitochondrial Connecting variability in global transcription rate to mitochondrial TRAP-1 expression and lymph node metastasis in colorectal cancer. variability. PLoS Biol 2010;8:e1000560. World J Gastroenterol 2012;18:5965–71.

www.aacrjournals.org Mol Cancer Res; 12(5) May 2014 669

TRAP1 Regulates Proliferation, Mitochondrial Function, and Has Prognostic Significance in NSCLC

Jackeline Agorreta, Jianting Hu, Dongxia Liu, et al.

Mol Cancer Res 2014;12:660-669. Published OnlineFirst February 24, 2014.

Updated version Access the most recent version of this article at: doi:10.1158/1541-7786.MCR-13-0481

Supplementary Access the most recent supplemental material at: Material http://mcr.aacrjournals.org/content/suppl/2014/02/27/1541-7786.MCR-13-0481.DC1

Cited articles This article cites 45 articles, 11 of which you can access for free at: http://mcr.aacrjournals.org/content/12/5/660.full#ref-list-1

Citing articles This article has been cited by 5 HighWire-hosted articles. Access the articles at: http://mcr.aacrjournals.org/content/12/5/660.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/12/5/660. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.