ANTICANCER RESEARCH 34: 753-758 (2014)

The Significance of HSP90AA1, HSP90AB1 and HSP90B1 Polymorphisms in a Turkish Population with Non-small Cell Lung

ENDER COSKUNPINAR1, NERGIZ AKKAYA1, PINAR YILDIZ2, YASEMIN MUSTERI OLTULU1, ENGIN AYNACI3, TURGAY ISBIR4 and ILHAN YAYLIM1

1Department of Molecular Medicine, Institute for Experimental Medicine (DETAE), Faculty of Pharmacy, Istanbul University, Capa, Istanbul, Turkey; 2Third Clinic, Yedikule Teaching Hospital for Chest Diseases and Thoracic Surgery, Istanbul, Turkey; 3Department of Chest Diseases, Faculty of Medicine, Medipol University, Istanbul, Turkey; 4Department of Medical Biology, School of Medicine, Yeditepe University, Istanbul, Turkey

Abstract. Background/Aim: Heat-shock (HSPs) are of these cases will be locally advanced or metastatic at the molecular chaperones which modify the structures and time of presentation (2). The early detection of lung cancer interactions of other proteins. The aim of our study was to leads to a better prognosis for reduced mortality and investigate HSP90AA1, HSP90AB1 and HSP90B1 gene morbidity (3-7). polymorphisms in patients with non-small cell lung cancer Although platinum-based doublets are standard-of-care for (NSCLC). Materials and Methods: Ninety-seven patients advanced NSCLC, treatment is very difficult and is a with NSCLC and 97 healthy controls were included in the controversial area mainly because of the large heterogeneity study. Real-time polymerase chain reaction was used for of different pathological conditions seen in advanced genotyping. Results: The frequency of mutant CC genotype NSCLC (6, 7). Despite all improvements in the current for HSP90AA1 (rs4947C/T), mutant AA genotype for treatment modality, including surgical resection, HSP90AB1 (rs13296A/G) and mutant CC genotype for chemotherapy and radiation therapy alone or in combination, HSP90B1 (rs2070908 C/G) was significantly higher in the prognosis remains poor in patients with locally advanced or patient group than in controls (p=0.019, p=0.004 and metastatic disease (6-8). p=0.036, respectively). The frequency of patients with Heat-shock proteins (HSPs) are characterized as homozygote mutant allele was also significantly higher than molecular chaperones, proteins which modify the that of controls and possessing of the mutant genotype structures and interactions of other proteins, and, are increased the risk for disease by approximately 2.9, 4.8, 1.9 induced by heat shock, and other chemical and physical for HSP90AA1, HSP90AB1 and HSP90B1, respectively. The stresses (9). The levels of HSPs are elevated in many types present study appears to be the first of its kind to report data of cancer, including prostate and breast cancer. HSP on these gene polymorphisms in patients with NSCLC in the overexpression was shown to be related to poor prognosis Turkish population. in terms of survival and response to therapy in specific cancer types (10-12). HSP expression may have a Lung cancer is one of the leading causes of deaths globally protective effect on spontaneous apoptosis of malignant (1). Smoking causes 80% to 90% of cases of lung cancer. cells, as well as apoptosis induced by therapy, which can Non-small cell lung carcinoma (NSCLC) accounts for be associated with tumor progression and resistance to approximately 80-85% of all lung cancer cases, and most treatment (13-15). The human family includes four confirmed members–the cytosolic HSP90α and HSP90β, the glucose-regulated -74 (GRP74), and the mitochondrial TNF receptor-associated Correspondence to: Associate Professor B. Pinar Yildiz, Yedikule protein-1 (TRAP1), which play a versatile role in cell Chest Disease and Surgery Training and Research Hospital, Third regulation, forming complexes with a large number of clinic, Zeytinburnu, Istanbul, Turkey. Tel: +90 05333585708, e-mail: cellular , transcription factors, and other molecules. [email protected] HSP90 seems to be required for cell survival and might be Key Words: NSCLC, HSP90AA1, HSP90B1, HSP90AB1, SNPs, a target for new treatment strategies for various tumor Turkish population. types (16).

0250-7005/2014 $2.00+.40 753 ANTICANCER RESEARCH 34: 753-758 (2014)

Materials and Methods Results

Study groups. Ninety-seven patients diagnosed as having NSCLC The frequency of males was significantly higher for (92 men; 5 women) and 97 healthy individuals (68 men; 29 women) patients (94.8%) in comparison to the control group were admitted to the Istanbul Yedikule Chest Diseases and Thoracic (70.1%), with a 7,8-fold increased risk for lung cancer Surgery Training Hospital and consecutively included in the (p<0.0001, Chi-square=20.541, OR=7.847, 95% CI=2.888- case–control study. Patients who were referred to our clinic and 21.320). From a total of 97 patients with NSCLC included diagnosed with NSCLC were informed of the study. Ninety-seven healthy persons without any malignancy were selected for the in our study, the histological tumor type was defined as control group that comprised only individuals with a negative family squamous cell in 32 (33%) cases, adenocarcinoma in 20 history of cancer. The patient and control groups were matched for (21%), malignant carcinoid tumor in 1 (1.%), large cell age. All participants signed an informed consent form before carcinoma in 6 (6%), while 38 (39%) were described as enrollment and Institutional Ethical committee (Istanbul University, unclassified NSCLC. School of Medicine Ethical Committee, November 14, 2011; no. The distribution of the HSP90AA1 (rs4947C/T) genotypes 1825) approval was obtained for the study. Blood samples from all in the patient group was significantly different when study participants were collected in EDTA-containing tubes. Genomic DNA was extracted from peripheral whole blood compared to that of the controls (p=0.019) (Table I). The according to the kit protocol (Roche Diagnostics GmbH, Mannheim, prevalence of TC heterozygosity was 25.8% (25/97) in Germany). patients and 17.5% (17/97) in the control group. The frequency of homozygous CC genotype (14.4%) was Genotyping by real-time-PCR. Three single nucleotide significantly higher in patients than in controls (5.02%) polymorphisms (rs4947C/T, rs13296 A/G and rs2070908 C/G) were (p=0.019). Moreover, the frequency of the C allele was analyzed in three different : HSP90AA1 [TTTGAAAACAG higher in controls when compared with NSCLC (27.3% and AAAGAAAAAGAACAA(C/T)ATCAAATTGTATGTACGCAGAGT TT], HSP90AB1 [TCTGAACCCACATCTTCGA TCTTGGG (A/G) 13.9% respectively; p=0.001, Chi-square=10.64) (Table I). TTTTCTTCATCATCTTTATCTTCCT], HSP90B1 [GGGTGAAAA The distribution of genotype and alleles and comparisons GCGGCCCGACCTGCTTG(C/G)GGTGTAGTGGGCGGACCGCG between study groups is shown in Table I. CGGCT] by using quantitative real-time PCR (RT-PCR LightCycler; Genotypic and allelic frequencies for HSP90AB1 Roche Diagnostics). Blood samples from all study participants (rs13296A/G) in patients with NSCLC and controls are listed were collected in EDTA-containing tubes. Genomic DNA was in Table II. Statistically significant differences in both isolated from blood with a spin column kit (Roche Diagnostics) distribution of genotypes and alleles of HSP90AB1 according to the manufacturer’s instructions. Quantitative RT-PCR for genotyping was performed on the LightCycler 1.5 system using (rs13296A/G) were found. The frequency of the homozygous 1 μl of hybridization probe pair (Light Cycler Fast Start DNA AA genotype was almost five-fold higher in NSCLC Master HybProbe) labeled with 3’-fluorescein and 5’-LightCycler compared to controls (p=0.004), and that of individuals with Red. The following protocol was used for amplification; initial A allele, similarly, 1.7-fold higher (Chi-square=7.2, denaturation at 95˚C for 10 min, followed by 40 cycles with p=0.007) (Table II). denaturation at 95˚C for 10 s, annealing at 60˚C for 10 s, and The distribution of HSP90B1 (rs2070908 C/G) genotype extension at 72˚C for 10 s. An additional melting curve analysis and allelic frequency are shown in Table III. The frequency was performed at 95˚C for 30 s, 40˚C for 2 min and 75˚C for 0 s in order to detect non-specific amplifications. of the homozygous CC genotype was almost two-fold higher in patients when compared to controls (p=0.036). In Statistical analysis. All statistical analyses were carried out using addition, the frequency of C allele carriers was higher in the SPSS version 17.0 (IBM, Armonk, NY, USA) statistical NSCLC than controls (Chi-square=6.5, p=0.01). package for Windows. The Chi-square test was used to assess The risk of was related to GA heterozygocity both the prevalence of the genotypic distribution, and allelic for the HSP90AB1 genotype (p=0.026). When we compared differences between groups. The associations between expression patients with advanced-stage (stage 3 and 4) and early-stage status and clinicopathological parameters were analyzed using Chi-square and Fisher’s exact test. The relative associations (stage 1 and 2) disease, individuals carrying A alleles of the between patients and controls were assessed by calculating crude HSP90AB1 gene had a 1.79-fold increased odds for Garth’s odds ratios (ORs) and 95% confidence intervals (95% advanced-stage disease (p=0.018, OR=1.786, 95% CIs). The threshold for significance was p<0.05. Linkage CI=1.022-3.120). disequilibrium between different polymorphisms was assessed No linkage disequilibrium was found between the three using D0 and r2 values obtained with the Haploview program genes. Haplotypes were evaluated for association with (http://www.broadinstitute.org/scientific- NSCLC (Table IV). The frequencies of individuals with community/science/programs/medical-and-population-genetics/ haploview/haploview). A multivariate logistic regression model haplotype-1 (CGA), haplotype-3 (TCA), and haplotype-5 was used to investigate the effects of genotypes and alleles after (TCG) were significantly higher in NSCLC when compared adjustment for age. Values of p<0.05 were considered to be to controls (p-values 0.0286, 0.0357 and 0.001, respectively) statistically significant. (Table IV).

754 Coskunpinar et al: HSP90AA1, HSP90AB1 and HSP90B1 Genes in NSCLC

Discussion Table I. Distribution of HSP90AA1 genotype and allele in NSCLC and control groups. In the current study, real-time quantitative PCR was used to NSCLC Controls OR (95% CI) p-Value investigate the possible association between polymorphisms n (%) n (%) in HSP90AA1 (rs4947C/T), HSP90AB1 (rs13296A/G), HSP90B1 (rs2070908C/G) and NSCLC. To our knowledge, Genotype 0.019 this is the first work to evaluate the collective impact of TT 58 (59.8) 75 (77.3) 0.436 (0.234-0.815) 0.009 these single-nucleotide polymorphisms on NSCLC in our TC 25 (25.8) 17 (17.5) 0.612 (0.306-1.224) 0.163 CC 14 (14.4) 5 (5.02) 3.1 (1.072-8.988) 0.027* ethnic population. According to our results, the frequency Allele 0.001 of CC genotype for HSP90AA1 (rs4947C/T), AA genotype T 141 (72.7) 167 (86.1) for HSP90AB1 (rs13296A/G) and CC genotype for C 53 (27.3) 27 (13.9) HSP90B1 (rs2070908) was significantly higher in NSCLC when compared to controls (p-value of 0.019, 0.004 and OR: Odds ratio; CI: confidence ınterval. *Fisher’s exact test. 0.036, respectively). HSP molecular chaperones guide the normal folding, intracellular disposition and proteolytic turnover of many of Table II. Distribution of HSP90AB1 genotype and allele in NSCLC and the key regulators of , differentiation and survival. control groups. Recent studies have reported that HSP90 and , in AB1 NSCLC Controls OR (95%CI) p-Value particular, might have an important role in the process of n (%) n (%) carcinogenesis (17-19). HSP90 was localized not only in cancer cells, but also in cells adjacent to cancer cells. It was Genotype 0.004 also shown that both HSP90α and -β mRNAs were GG 37 (38.1) 47 (48.5) 0.656 (0.371-1.162) 0.147 overexpressed in the cytoplasm around the nucleus of cancer GA 41 (42.3) 46 (47.4) 1.232 (0.699-2.171) 0.47 AA 19 (19.6) 4 (4.1) 5.66 (1.849-17.348) 0.001* cells (20). Overexpression of HSP90α has been reported in Allele 0.007 various types of cancer, such as human leukemia cells (18) G 115 (59.3) 140 (72.2) and pancreatic carcinomas (19). In addition, HSP90β has A 79 (40.7) 54 (27.8) been reported to inhibit apoptosis and cell differentiation (21). Currently, HSP90 inhibitors are being developed as OR: Odds Ratio; CI: confidence Interval *Fisher’s Exact Test. anticancer agents. Early studies have shown promising results in a subgroup of solid tumors such as ALK- rearranged NSCLC (22). Table III. Distribution of HSP90B1 genotype and allele in NSCLC and Although lung cancer has complex and various genetic control groups. abnormalities, the molecular nature of carcinogenesis is not B1 NSCLC Controls OR p-Value fully-understood. Understanding the complex networks that n (%) n (%) (95%CI) lead to cancer development could lead to development of molecular target-based anticancer treatment modalities which Genotype 0.036 might allow for better prognosis of this deadly disease. GG 23 (23.7) 36 (37.1) 0.527 (0.282-0.982) 0.042 Increased expression of HSP90 proteins has been shown in GC 49 (50.5) 48 (49.5) 0.96 (0.547-1.685) 0.886 CC 25 (25.8) 13 (13.4) 2.24 (1.07-4.704) 0.03 different types of tumors (23). Up-regulation of both Allele 0.01 cytosolic HSPα and HSP90β have been demonstrated in G 95 (49) 120 (61.9) various tumors, including lung cancer, with data for HSP90β C 99 (51) 74 (38.1) being considerably limited (24, 25). Prognostic implication and relationship with metastatic potential has been OR: Odds Ratio; CI: confidence Interval. *Fisher’s Exact Test. implicated by the same authors (25). Inhibition of HSP90 can lead to degradation of multiple oncogenic signaling proteins which are involved in tumor progression (26). HSP90α has been shown to be associated with poor increased risk for advanced-stage disease (p=0.018). Our prognosis in human breast cancer (27). In our study, we results suggest that HSP90AB1 (rs13296A/G) polymorphism showed that the risk of metastasis was related to the GA might have a prognostic impact in NSCLC. heterozygocity for HSP90AB1 genotype (p=0.026). Then we In a study investigatirg the role of HSP90 Gln488His compared patients with advanced-stage (stage 3 and 4) and (C>G) polymorphism as a potential risk factor for breast early-stage (stage 1 and 2) disease, are found individuals cancer, a significant association between HSP90 G allele and carrying A alleles of HSP90AB1 gene had a 1.78-fold breast cancer risk was detected (28). Urban et al. (29) aimed

755 ANTICANCER RESEARCH 34: 753-758 (2014)

Table IV. Results of the haplotype analysis in patients and controls (HSP90AA1, HSP90AB1 and HSP90B1 genes). The frequencies of haplotype-1 (CGA), haplotype-3 (TCA), haplotype-5 (TCG) were significantly higher in patients with NSCLC when compared with controls.

Haplotype no. Haplotype Frequency χ2 p-Value

Total NSCLC Controls

1 CGA 0.385 0.439 0.330 4.793 0.0286 2 CCA 0.162 0.135 0.190 2.179 0.1399 3 TCA 0.150 0.188 0.112 4.41 0.0357 4 TGA 0.097 0.074 0.120 2.416 0.1201 5 TCG 0.090 0.137 0.042 10.78 0.001 6 CGG 0.066 0.077 0.055 0.74 0.3895 7 CCG 0.044 0.051 0.038 0.377 0.5391

to characterize single-nucleotide polymorphism of human References cytosolic HSP90 genes (HSP90AA1 and HSP90AB1). It was concluded that all HSP90 single-nucleotide polymorphisms 1 Gao W, Liu L, Lu X and Shu Y: Circulating microRNAs: would have a limited effect on predicting functional Possible prediction biomarkers for personalized therapy of non- small cell lung carcinoma. Clin Lung Cancer 12: 14-17, 2011. sequence in key domains of the human HSP90 proteins. 2 Feng JF, Wu JZ, Hu SN, Gao CM, Shi MQ, Lu ZH, Sun XC, Their DNA samples originated from 26 Caucasians, 27 Zhou JR and Chen BA: Polymorphisms of the ribonucleotide African Americans, 18 Chinese, 13 Japanese, 10 Mexicans reductase M1 gene and sensitivity to platin-based chemotherapy and seven Southeast Asians, and a total of 15 exonic single- in non-small cell lung cancer. Lung Cancer 66: 344-349, 2009. nucleotide polymorphisms were identified in their study. To 3 Paesmans M: Stage IV NSCLC. Prognostic factors. Rev Mal date, there have been limited data examining the presence of Respir 25: 99-106, 2008. cytosolic HSP90 SNPs in the human population (29-31). 4 Lang-Lazdunski L: Surgery for non-small cell lung cancer. Eur Respir Rev 22: 382-404, 2013. This study is the first, to our knowledge, which analyzed 5 Jemal A, Siegel R, Ward E, Hao Y, Xu J and Thun MJ: Cancer HSP90 gene variants in a Turkish population. We showed statistics. CA Cancer J Clin 59: 225-249, 2009. that genotype distribution for each gene and allelic 6 Rudd RM, Gower NH, Spiro SG, Eisen TG, Harper PG, Littler predominance is significantly different in NSCLC compared JA, Hatton M, Johnson PW, Martin WM, Rankin EM, James LE, to healthy controls. We clearly showed that the frequency of Gregory WM, Qian W and Lee SM: Gemcitabine plus the patients homozygous for the mutant allele of all genes carboplatin versus mitomycin, ifosfamide, and cisplatin in (HSP90AA1, HSP90AB1, HSP90B1) was also significantly patients with stage IIIB or IV non-small cell lung cancer: A phase III randomized study of the London Lung Cancer Group. higher than in controls. Possessing a mutant genotype J Clin Oncol 23: 142-153, 2005. increases the odds of disease by approximately 2.9-, 4.8-, 7 Martoni A, Marino A, Sperandi F, Giaquinta S, Di Fabio F, and 1.9-fold, respectively. It can be concluded that Melotti B, Guaraldi M, Palomba G, Preti P, Petralia A, Artioli F, HSP90AA1 (rs4947C/T), HSP90AB1 (rs13296A/G) and Picece V, Farris A and Mantovani L: Multicentre randomised HSP90B1 (rs2070908) polymorphisms might be related to an phase III study comparing the same dose and schedule of cisplatin increased risk of lung cancer. Such a hypothesis needs to be plus the same schedule of vinorelbine or gemcitabine in advanced analyzed with large population-based multicenter studies. non-small cell lung cancer. Eur J Cancer 41: 81-92, 2005. 8 Sun S, Schiller JH, Spinola M and Minna JD: New molecularly To our knowledge, these are the first data to evaluate the targeted therapies for lung cancer. J Clin Invest 117: 2740-2750, collective impact of the single-nucleotide polymorphisms of 2007. these three genes on NSCLC in our ethnic population. 9 Ciocca DR and Calderwood SK: Heat-shock proteins in cancer: Mutation of HSP90AA1 (rs4947C/T), HSP90AB1 Diagnostic, prognostic, predictive, and treatment implications. (rs13296A/G), HSP90B1 appears to increase the risk for lung Cell Stress Chaperones 10: 86-103, 2005. cancer. 10 van de Vijver MJ, He YD, van’t Veer LJ, Dai H, Hart AA, Voskuil DW, Schreiber GJ, Peterse JL, Roberts C, Marton MJ, Parrish M, Atsma D, Witteveen A, Glas A, Delahaye L, van der Acknowledgements Velde T, Bartelink H, Rodenhuis S, Rutgers ET, Friend SH and Bernards R: A gene-expression signature as a predictor of The present work was supported by a grant from the Scientific survival in breast cancer. N Engl J Med 347: 1999-2009, 2002. Research Projects Coordination Unit of Istanbul University (Project 11 Jolly C and Morimoto RI: Role of the heat-shock response and No: 21163). The Authors would like to thank Mrs. Aylin Çayirli for molecular chaperones in oncogenesis and cell death. J Natl her help in data collection. Cancer Inst 92: 1564-1572, 2000.

756 Coskunpinar et al: HSP90AA1, HSP90AB1 and HSP90B1 Genes in NSCLC

12 Cornfold PA, Dodson AR, Parsons KF, Desmond AD, 24 Yano M, Naito Z, Tanaka S and Asano G: Expression and roles Woolfenden A, Fordham M, Neoptolemos JP, Ke Y and Foster of heat-shock proteins in human breast cancer. Jpn J Cancer Res JS: Heat-shock protein expression independently predicts clinical 87(9): 908-915, 1996. outcome in prostate cancer. Cancer Res 60: 7099-7105, 2000. 25 Biaoxue R, Xiling J, Shuanying Y, Wei Z, Xiguang C, Jinsui W 13 Ciocca DR, Rozados VR, Cuello-Carrio´n FD, Gervasoni SI, and Min Z: Up-regulation of HSP90-β and A1 correlates Matar P and Scharovsky OG: Heat-shock proteins 25 and 70 in with poor survival and lymphatic metastasis in lung cancer rodent tumors treated with doxorubicin and lovastatin. Cell patients. J Exp Clin Cancer Res 31: 1-14, 2012 Stress Chaperones 8: 26-36, 2003. 26 Ono N, Yamazaki T, Tsukaguchi T, Fujii T, Sakata K, Suda A, 14 Nylandsted J, Brand K and Jaattela M: Heat-shock protein 70 is Tsukuda T, Mio T, Ishii N, Kondoh O and Aoki Y: Enhanced required for the survival of cancer cells. Ann NY Acad Sci 926: antitumor activity of erlotinib in combination with the HSP90 122-125, 2000. inhibitor CH5164840 against non-small cell lung cancer. Cancer 15 Volloch VZ and Sherman MY: Oncogenic potential of HSP72. Sci 104: 1346-1352, 2013. Oncogene 18: 36 48-51, 1999. 27 Jameel A, Law M, Coombes R and Luqmani Y: Significance of 16 Neckers L and Ivy SP: Heat-shock protein 90. Curr Opin Oncol heat-shock protein-90 as a prognostic indicator in breast cancer. 15: 419-424, 2003. Int J Oncol 2: 1075-1080, 1993. 17 Jameel A, Skilton RA, Campbell TA, Chander SK, Coombes RC 28 Zagouri F, Sergentanis TN, Gazouli M, Tsigginou A, and Lugmani YA: Clinical and biological significance of Dimitrakakis C, Papaspyrou I, Eleutherakis-Papaiakovou E, HSP89α in human breast cancer. Int J Cancer 50: 409-415, Chrysikos D, Theodoropoulos G, Zografos GC, Antsaklis A, 1992. Dimopoulos AM and Papadimitriou CA: HSP90, HSPA8, HIF-1 18 Yufu Y, Nishimura J and Nawata H: High constitutive expression alpha and HSP70-2 polymorphisms in breast cancer: A of heat-shock protein 90 alpha in human acute leukemia cells. case−control study. Mol Biol Rep 39(12): 10873-10879, 2012. Leuk Res 16: 597-605, 1992. 29 Urban JD, Budinsky RA and Rowlands JC: An evaluation of 19 Gress TM, Müller-Pillasch F, Weber C, Lerch MM, Friess H, single nucleotide polymorphisms in the human heat-shock Büchler M, Beger HG and Adler G: Differential expression of protein 90 kDa α and β isoforms. Drug Metab Pharmacokinet heat-shock proteins in pancreatic carcinoma. Cancer Res 54: 27: 268-278, 2012. 547-551, 1994. 30 Passarino G, Cavalleri GL, Stecconi R, Franceschi C, Altomare 20 Ogata M, Naito Z, Tanaka S, Moriyama Y and Asano G: K, Dato S, Greco V, Luca Cavalli Sforza L, Underhill PA and de Overexpression and localization of heat-shock proteins mRNA Benedictis G: Molecular variation of human HSP90α and in pancreatic carcinoma. J Nippon Med Sch 67(3): 177-185, HSP90β genes in Caucasians. Hum Mutat 21(5): 554-555, 2003. 2000. 31 Hassun Filho PA, Cedenho AP, Lima SB, Ortiz V and Srougi M: 21 Didelot C, Lanneau D, Brunet M, Bouchot A, Cartier J, Jacquel Single nucleotide polymorphisms of the heat-shock protein 90 A, Ducoroy P, Cathelin S, Decologne N, Chiosis G, Dubrez- gene in varicocele-associated infertility. Int Braz J Urol 31: 236- Daloz L, Solary E and Garrido C: Interaction of heat-shock 242, 2005. protein 90 β isoform (HSP90β) with cellular inhibitor of apoptosis 1 (c-IAP1) is required for cell differentiation. Cell Death Differ 15(5): 859-866, 2008. 22 Garcia-Carbonero R, Carnero A and Paz-Ares L: Inhibition of HSP90 molecular chaperones: Moving into the clinic. Lancet Oncol 14(9): 358-369, 2013. 23 McDowell CL, Bryan Sutton R and Obermann WM: Expression Received November 4, 2013 of HSP90 [corrected] proteins in human tumor tissue. Revised December 25, 2013 Int J Biol Macromol 45(3): 310-314, 2009. Accepted December 31, 2013

757