Abstract
Predicting bladder cancer progression is important in selecting the optimal treatment for bladder cancer. Because current diagnostic factors regarding progression are lacking, new factors are needed to further stratify the curative potential of bladder cancer. Glycoprotein-130 (GP130), a transmembrane protein, is central to a number of signal transduction pathways involved in tumor aggressiveness, making it an attractive target. We hypothesize that if GP130 is found in an aggressive population of bladder tumors, then blocking GP130 expression may inhibit bladder cancer growth. Herein, we quantitatively show, using 11 patient samples and four bladder cancer cell lines, that GP130 is expressed in the aggressive human bladder tumors and in high-grade bladder cancer cell lines. Moreover, GP130 is significantly correlated with tumor grade, node category, tumor category, and patient outcome. We demonstrated a tumor-specific GP130 effect by blocking GP130 expression in bladder tumor cells, which resulted in decreased cell viability and reduced cell migration. Furthermore, we reduced tumor volume by approximately 70% compared with controls by downregulating GP130 expression using chitosan-functionalized nanoparticles encapsulating GP130 siRNA in an in vivo bladder cancer xenograft mouse model. Our results indicate that GP130 expression is linked to the aggressiveness of bladder tumors, and blocking GP130 has therapeutic potential in controlling tumor growth.
Introduction
Bladder cancer is the sixth most common cancer in the United States with an estimated 81,490 newly diagnosed cases expected this year, and is a common cancer worldwide with the highest rates in Europe, North America, and Northern Africa (1, 2). Seventy percent of bladder cancers are superficial (3), and among patients who present with superficial disease, recurrence rates range from 67% to 73%, with disease progression rates ranging from 20% to 30% (4). Although many patients present with non–muscle-invasive disease, approximately 35% of patients will present with or develop muscle-invasive disease (5). Major challenges in treating bladder cancer are the high recurrence rates and the need for long-term follow-up consisting of urine cytology and cystoscopy, thus making the cost-per-case for bladder cancer one of the highest among all cancers (6–8). Better diagnostic markers are required to gain insight into the progression of bladder cancer, which ultimately may be used to determine the most effective therapeutic interventions.
Glycoprotein-130 (GP130, CD130, or IL6ST), a transmembrane protein, is located at a central point for a number of oncogenic signaling cascades including the JAK/STAT, PI3K/AKT, and MAPK/ERK pathways (9). These downstream pathways have been shown to be involved in multiple hallmarks of cancer development including invasion, metastasis, resistance to apoptosis, and proliferation (9–11), which makes GP130 an attractive therapeutic target.
We hypothesize that if GP130 is found in an aggressive population of bladder tumors, then blocking GP130 expression may lead to inhibition of bladder cancer growth. To address this hypothesis, we performed molecular analyses on clinically annotated bladder specimens using our novel GP130 bladder cancer biomarker as well as survival and metastatic biomarkers that have been correlated with advanced bladder cancer including survivin, B-cell lymphoma-extra large (Bcl-xL), and VEGFR-2 (VEGFR2). To functionally test the role of GP130 in human bladder cancer cells, we blocked the expression of GP130 using GP130 siRNA (siGP130) and investigated cell migration, viability, and growth. We also used a GP130 inhibitor, SC144, to confirm cell migration, viability, and growth upon the downregulation of GP130 expression. Then, we implemented our previously described chitosan (CH) nanoparticle (NP) delivery system (12) to decrease GP130 protein expression in an in vivo xenograft bladder cancer mouse model. For the first time, we show that blocking GP130 using CH-functionalized nanoparticles encapsulating siRNA GP130 has therapeutic potential in controlling tumor growth.
Materials and Methods
Patient tissue specimens
Eleven deidentified specimens were collected from 9 patients by transurethral resection of bladder tumor (TURBT) or cystectomy. The specimens were comprised of nine urothelial cell carcinomas and two normal control bladder tissues. All specimens were collected between 2014 and 2016. From each patient, we recorded tumor histology type, node category, tumor category, tumor grade, sex, treatment type, tumor relapse, and patient survival (Table 1). Bladder tumors were defined as aggressive if they met the following criteria: tumor was of high grade, patient had poor survival outcome, and tumor category was T3 or T4. All patients were provided informed consent and offered enrollment into a biospecimen repository approved by the Yale University Institutional Review Board.
Clinical and histopathologic data
Cell culture
RT-4, T24, UM-UC-3, and TCCsup bladder cancer cells were obtained directly from the American Type Culture Collection from 2007 to 2011. T24 and RT-4 cells were maintained in McCoy medium, whereas UM-UC-3 and TCCsup cells were maintained in Eagle minimum essential medium. All cells were maintained at 37°C in 5% CO2 atmosphere, and supplemented with 10% FBS and 1% glutamine. At the start of each experiment, cells were used in the exponential growth phase. Also, a fresh vial of cells was thawed after ten passages. Six months prior to initiating experiments, cells were reauthenticated for quality control using short tandem repeat DNA profiles (The Yale University DNA Analysis Facility, New Haven, CT). In addition, cells were routinely screened for Mycoplasma using MycoAlert (Lonza Biologics Inc.) every 6 months in our laboratory.
siRNA treatment
Bladder cancer cells were transfected with: siGP130 to a final concentration of 100 nmol/L with Lipofectamine RNAiMAX (Invitrogen); scrambled siRNA (siSC) to a final concentration of 100 nmol/L with Lipofectamine RNAiMAX; Lipofectamine RNAiMAX alone (vehicle); or were left untreated. The GP130 target sequence was as follows: 5′-CAGUAAAUCUCACAAAUGA-3′ (sense), with the scramble control: 5′-AACGUACGCGGAAUACUUCGA-3′ (Dharmacon). After 72 hours, the cells were assessed for cell viability, cell migration, cell growth, or GP130 protein expression.
Cell viability of bladder cancer cells
Cell viability assays were conducted according to the manufacturer's instructions (Clontech Laboratories). In brief, siGP130, siSC, or vehicle were administered to 5 × 103 human bladder cancer cells per well in a 96-well plate or the cells were left untreated, and cell viability was measured using the tetrazolium (WST-1) reagent. The percent inhibition of cell viability relative to siSC control was measured using Origin Lab Data Analysis Software.
Migration assay
Bladder cancer cells were treated with siGP130, siSC, or vehicle and plated at 6 × 104 cells per well of a 24-well plate and grown for 24 hours to confluency. A scratch was created using a P200 tip in a confluent monolayer of cells with reduced serum. The distance the cells migrated to cover the scratch was calculated over 24 hours.
Cell growth assay and crystal violet staining
Bladder cancer cells were treated with siGP130, siSC, or vehicle and plated at 1.5 × 104 cells per well (6-well plate). After 48 hours, the cells were fixed with ice-cold methanol for 10 minutes and stained with 0.5% crystal violet dye for 10 minutes (13). The bladder cancer cells were washed three times with deionized water for 5 minutes before being dried and photographed. Cell growth (or survival) rate (%) = (Cell number of treatment group)/(Cell number of control group) × 100%.
Western blot analysis
Protein expression including phosphorylation was determined by Western blot analysis. In brief, protein lysates were prepared from human bladder cancer cells and from snap-frozen human bladder tumors. Radioimmunoprecipitation assay buffer (Cell Signaling Technology) supplemented with a cOmplete, mini, EDTA-free protease inhibitor cocktail (Roche Applied Science), 1 mmol/L phenylmethylsulfonyl fluoride, 2 μg/mL aprotinin (protease inhibitor), and 1 mmol/L sodium fluoride was used to lyse the tissues and cells prior to quantification with the Bradford assay (14). Samples were separated on 4%–15% Mini-PROTEAN Gels (Bio-Rad lab. Inc.) and proteins were transferred to polyvinylidene fluoride membranes. Membranes were blocked with 5% nonfat dry milk in Tris-buffered saline (TBS) with 0.1% Tween 20 (TBST) for 1 hour at room temperature and incubated with primary antibodies in 5% nonfat dry milk in TBST at 4°C overnight. The membranes were subsequently washed with TBST and incubated with HRP-conjugated donkey anti-rabbit or donkey anti-mouse IgG secondary antibodies (Cell Signaling Technology). A chemiluminescence system (Thermo Fisher Scientific) was used to detect protein signal. Samples were normalized on the basis of GAPDH quantification, and band density was determined using ImageJ software (NIH).
NP fabrication and characterization with siGP130
The double emulsion solvent evaporation technique was used to synthesize NPs containing siRNA (15). In brief, polylactic-co-glycolic acid (PLGA) was dissolved overnight in methylene chloride (DCM), prior to siRNA and spermidine complex formation using an 8:1 molar ratio of the polyamine nitrogen to the nucleotide phosphate. One hundred nanomoles of siGP130 per 100 mg polymer in Tris-EDTA (10 mmol/L Tris-HCl, 1 mmol/L EDTA) buffer was added dropwise to the PLGA solution while vortexing. This solution was sonicated and subsequently added to a 2.5% polyvinyl alcohol (PVA), 5 mg/mL avidin palmitate solution for the second emulsion. NPs were hardened during solvent evaporation in 0.3% PVA for 3 hours. To generate the CH surface modification, the hardened NPs were reacted with 10 times molar excess biotin-CH2.5 to avidin (NP-CH2.5) in PBS for 30 minutes. To synthesize unmodified NPs (NP-Unmod), the second emulsion contained only 2.5% PVA, and NPs were incubated posthardening in PBS without ligand for 30 minutes. All NPs were washed twice in deionized water to remove residual solvent, centrifuged at 4°C, lyophilized, and stored at −20°C. As previously reported (12, 15), 3–5 mg of siRNA NPs were dissolved in 0.5 mL of DCM for 30 minutes, and siGP130 was extracted twice into Tris-EDTA buffer. The quantity of extracted double-stranded siRNA was determined using the QuantIT PicoGreen assay (Invitrogen). Fluorescence was compared with a known siRNA standard. Encapsulation efficiency was determined by comparing the amount of siRNA loaded into the NPs with theoretical loading (1 nmol siRNA/mg polymer). For the GP130 NPs (NP-siGP130-CH2.5), the loading was 514 pmol siRNA/mg NP. Scanning electron microscopy images were analyzed using ImageJ software to determine NP morphology, diameter, and size distribution (Table 2; ref. 12).
NP characterization
Bladder cancer mouse model
Foxn1 nu/nu mice were subcutaneously injected in the flank with 5 × 106 UM-UC-3 bladder cancer cells. Tumor volumes were measured two times per week using calipers and estimated using the formula (tumor length × tumor width2) × π/6. When tumor volumes reached approximately 100 ± 25 mm3 (referred to as day 0), the mice were randomly divided into three groups and intratumorally injected with: PBS (untreated), 1 mg/100 μL NP-Bk-CH2.5 (CH-functionalized blank NPs) diluted in PBS, or 1 mg/100 μL NP-siGP130-CH2.5 (514 pmol siGP130; CH-functionalized NPs encapsulating siGP130) diluted in PBS. As we have previously reported, CH functionalization was used to facilitate the delivery of the siRNA (12). Mice were treated on days 0, 4, 7, and 11. After 14 days, the xenograft mice were sacrificed and mouse tumors were split; half was fixed in 10% neutral buffer formalin (VWR international) for IHC, and half was snap frozen and stored at −80°C for Western blot analysis. All animal studies were approved by the Institutional Animal Care and Use Committee of Yale University (New Haven, CT).
IHC
IHC was performed as described previously (16). In brief, the paraffin-embedded mouse tumors were deparaffinized in xylene, rehydrated with graded ethanol, and postfixed in 4% paraformaldehyde before undergoing antigen retrieval. The slides were incubated with a keratin 20 (C9Z1Z) antibody (Cell Signaling Technology) and developed using a vector red/DAB substrate kit, which previously had been shown to significantly correlate with tumor grade and stage (17).
Statistical analysis
Data are presented as mean ± SD from 3 to 6 independent experiments unless otherwise noted. Statistical significance was determined by Student t test (P < 0.05) to assess significant differences. Results are presented as mean ± SD in which * denotes P < 0.05, ** denotes P < 0.01, *** denotes P < 0.001, and **** denotes P < 0.0001 unless indicated differently. Significance was assessed between clinical and histopathologic data with respect to biomarker expression; between NP-siGP130-CH2.5 and controls (PBS or NP-Bk-CH2.5) with respect to tumor growth; and between siGP130 and controls with respect to cell migration and cell viability.
Results
GP130 expression in bladder tumors
Western blot analysis demonstrated that GP130 was highly expressed in aggressive bladder cancer patient specimens compared with the nonaggressive or normal control bladder specimens (Fig. 1A). Also, we demonstrated that GP130 expression correlated significantly with tumor grade (P = 0.027), node category (P = 0.001), and patient survival (P = 0.0001; Fig. 1B). Similarly, Survivin expression was correlated significantly with patient survival (P = 0.048), and Bcl-xL expression correlated significantly with node category (P = 0.040), tumor category (P = 0.040), and patient survival (P = 0.040). Conversely, VEGFR2 expression did not correlate with tumor grade, node category, tumor category, or patient survival (Fig. 1B). When focusing only on the matched specimens (n = 4), there was 12, 4, 12, and 7 times higher expression of GP130, survivin, Bcl-xL, and VEGFR2, respectively, in the aggressive cancer specimens compared with the matched adjacent normal control tissues (Fig. 1C). In addition to the patient specimens, aggressive human bladder cancer cell lines (TCCsup, T24, UM-UC-3) demonstrated higher GP130 expression compared with the more differentiated bladder cancer RT-4 cells (Fig. 1D).
Biomarker expression in human bladder specimens and cancer cells. A, Western blot analysis was performed on aggressive human bladder cancer specimens (2,3,7), nonaggressive human bladder cancer specimens (1,5,6), and normal human bladder (4,8). Representative blots are shown. B, Biomarkers (GP130, survivin, Bcl-xL, and VEGFR2) were related to the node category, tumor category, tumor grade, and survival in human bladder specimens. The error bars represent SD. C, A bar graph was generated comparing the biomarkers within the patient-matched tumors and adjacent normal tissues. D, Western blot analysis was performed on human bladder cancer cells (RT-4, UM-UC-3, T24, and TCCsup). A representative blot is shown. GAPDH was used for loading equivalency and protein integrity. Results are presented as mean ± SD in which *, P < 0.05; **, P < 0.01; ***, P < 0.001; and ****, P < 0.0001.
Blocking GP130 expression in in vitro human bladder cancer cells decreased cell viability, growth, and migration
Greater than 95% reduction in GP130 expression was achieved for UM-UC-3 and T24 human bladder cancer cells treated with siGP130 as demonstrated by Western blot analysis (Fig. 2A and B). UM-UC-3, T24, and RT-4 bladder cancer cells were treated with 100 nmol/L siGP130, resulting in a 45%, 35%, and 20% decrease in cell viability, respectively, as assessed by a WST-1 colorimetric assay (Fig. 2C–E).
Effect of GP130 siRNA knockdown on GP130 expression and cell viability in in vitro human bladder cancer cell lines. Western blot analysis was performed on human high-grade UM-UC-3 (A) and T24 (B) bladder cancer cells untreated or treated with siSC or siGP130. Actin Western blot shows loading equivalency and protein integrity. Viability assays were performed on UM-UC-3 (C), T24 (D), and RT-4 (E) bladder cancer cells untreated or treated with siSC or siGP130 using a WST-1 colorimetric assay. Results are presented as mean ± SD in which *, P < 0.05; ****, P < 0.0001.
We determined the cell growth (or survival) rate of siGP130 treated UM-UC-3 bladder cancer cells with crystal violet staining (Fig. 3A) and confirmed that siGP130 inhibited cell growth by 60%. A migration (or wound) assay was performed on untreated UM-UC-3 bladder cancer cells or cells treated with vehicle control, siSC control, or siGP130. Twelve hours postscratch, the untreated, vehicle-treated, and siSC-treated UM-UC-3 human bladder cancer cells migrated to cover 71%, 71%, and 67% of the scratch, respectively, whereas the siGP130-treated bladder cancer cells only migrated to cover 21% of the scratch. After 18 hours, the untreated, vehicle-treated, and siSC-treated controls reached 100% confluency, whereas the siGP130-treated cells migrated to cover only 50% of the scratch, confirming the role of GP130 in inhibiting cell migration (Fig. 3B).
Cell growth and migration of bladder cancer cells upon siGP130 knockdown. UM-UC-3 bladder cancer cells were treated with siGP130 to assess cell growth as visualized with 0.5% crystal violet (A), or cell migration as measured by scratch closure (B).
Tumor response to GP130 siRNA in a xenograft mouse model
Mouse bladder tumors were treated with PBS, NP-Bk-CH2.5, or NP-siGP130-CH2.5 for 14 days. On average, the NP diameters were 152 ± 16 nm (Table 2). Mouse bladder tumors treated with NP-siGP130-CH2.5 were significantly smaller than both the PBS or NP-Bk-CH2.5 controls at day 7 (41%, 38%), 11 (65%, 60%), and 14 (75%, 60%), respectively (Fig. 4A).
In vivo treatment of xenograft bladder tumors. A, Mice were intratumorally injected on day 0, 4, 7, and 11 with 100 μL of PBS, NP-BK-CH2.5, or NP-siGP130-CH2.5 diluted in PBS and tumor volume determined. Results are presented as mean ± SD in which **, represents P < 0.01 and ***, represents P < 0.001. B, Western blot analysis of biomarkers within the tumors treated with PBS, NP-Bk-CH2.5, or NP-siGP130-CH2.5. C, IHC analysis of the keratin-20 marker in tumors, treated with PBS, NP-Bk-CH2.5, or NP-siGP130-CH2.5.
Western blot analysis results confirmed that GP130 expression was decreased in the mouse bladder tumors treated with siGP130. Lower levels of GP130 expression were present in the NP-siGP130-CH2.5–treated tumors compared with the PBS or NP-Bk-CH2.5 controls. Upon GP130 inhibition, we observed a decrease in the expression of pAKT, pmTOR, VEGFR2, survivin, and Bcl-xL. However, the level of pSTAT3 expression in PBS or NP-Bk-CH2.5 controls was below our level of detection. Interestingly, we observed an increase in pERK expression upon GP130 knockdown (Fig. 4B).
In addition, IHC staining of keratin-20, a marker of high-grade urothelial carcinomas, showed elevated levels in the PBS and NP-Bk-CH2.5 controls compared with the NP-siGP130-CH2.5–treated mouse tumors (Fig. 4C).
Discussion
To date, the role of GP130 in bladder cancer has not been elucidated. In this study, we assessed GP130, survivin, Bcl-xL, and VEGFR2 expression on 11 snap-frozen bladder specimens collected from TURBT or cystectomy in which the clinical and histopathologic data were available (Table 1; Fig. 1B). We demonstrated that GP130 expression significantly correlates with an aggressive population of bladder cancers. In particular, GP130 expression significantly correlates with tumor grade, node category, and patient survival. In contrast to our findings, Xu and Neamati reported that GP130 was significantly decreased in bladder cancer (9). In our study, low GP130 expression was mainly found in nonaggressive papillary bladder tumors, whereas elevated GP130 expression levels correlated with malignant tumors with poor outcome, which may add value in the diagnosis of aggressive bladder tumors. Nevertheless, GP130 may have a role in cancer growth as it was reported by Xu and Neamati that GP130 was coexpressed with cancer-promoting genes including BPTF, a cancer growth and proliferation marker, as well as with ARHGAP18, a metastasis marker using a lymphoma microarray (9).
To functionally assess the role of GP130 in bladder cancer development, we used siGP130 to decrease GP130 expression in in vitro bladder tumor models. We observed a decrease in bladder cell growth, viability, and migration, corresponding with diminished GP130 expression thereby suggesting a tumor-specific GP130 effect. To further confirm the role of GP130 in tumor growth, we utilized our previously described NP delivery system in vivo (12). On the basis of our previous studies, we used our NP delivery system to prolong siRNA protection in our animal model. Also, we reported that encapsulating siRNAs in NPs increased their half-life from hours to weeks, thereby enabling siRNA release from the NPs for an extended duration (15, 18). The released siRNA decreased target mRNA expression, and resulted in diminished protein expression and tumor growth inhibition (12). To confirm our findings with respect to prosurvival and cell migration markers such as survivin, Bcl-xL, and VEGFR2 from our human bladder cancer specimens, we evaluated these markers in our bladder cancer mouse model. We found the aforementioned markers to be downregulated upon intratumoral injection of NP-siGP130 in our mouse tumors, suggesting a possible therapeutic role of GP130 in bladder cancer growth. However, future experiments will be required to assess the potential off-target effects on normal tissue expression of GP130 in the liver, kidney, colon, and urothelium.
In addition to GP130, survivin expression also correlated with survival in our human bladder cancer specimens. Previously, our group showed that survivin was able to predict aggressive tumor behavior in vivo (19) and to predict bladder cancer recurrence, using human bladder cancer specimens (20). Although VEGFR2 expression has been reported to be linked to the invasiveness of human bladder cancers (21, 22), in our study VEGFR2 was not significantly correlated with our clinical and histopathologic data, although it appeared to trend toward significance. In contrast, when we examined prosurvival and cell migration factors in our NP-siGP130-CH2.5–treated mouse tumors, we found survivin, Bxl-xL, and VEGFR2 to be downregulated, suggesting a role for GP130 in bladder cancer growth. A previous report by Nakanishi and colleagues suggested that inhibiting VEGFR expression resulted in a decrease in bladder cancer growth and invasion (23).
We demonstrated that GP130, located at a central point for a number of oncogenic signaling cascades, has a role in bladder cancer growth. Using our xenograft tumor model (Fig. 4B), we demonstrated an inhibition of the downstream pathway AKT (and mTOR) as was noted by a decrease in the phosphorylation levels, correlating with a decrease in bladder cancer growth. This phenomenon of deactivating AKT (and mTOR) has been well established in playing an important role in regulating cancer growth (24). Although STAT3, an antiapoptosis-related protein, is one of the major downstream pathways of GP130/IL6 and is upregulated in many tumors including bladder cancer (25, 26), it did not appear to be active in our bladder cancer model, suggesting a STAT3-independent mechanism. Godoy-Tundidor and colleagues, previously demonstrated that proliferation could be stimulated in prostate cancer cells through the PI3K/AKT pathway without activating STAT3, which also may be occurring in our bladder cancer model (27). Currently, there are no small-molecule GP130 inhibitors in clinical trials. However, there are reports of small-molecule GP130 inhibitors including SC144 (11) and LMT-28 (28) being tested preclinically. We have tested the SC144 inhibitor using in vitro bladder cancer cells and observed a significant knockdown of GP130, survivin, and Bcl-xL expression. In addition, a dose–response curve assessing cell viability was performed with SC144 (Supplementary Fig. S1). We found a decrease in cell growth and cell migration when SC144 was used at 3.5 μmol/L (Supplementary Figs. S2 and S3). Similarly, Xu and colleagues demonstrated a decrease in GP130, survivin, and Bcl-xL in OVCAR-8 cells when treated with 2 μmol/L SC144 (11).
Knocking down GP130 alone in our in vivo model did not completely eradicate the tumor, suggesting that additional feedback mechanisms are involved. We examined the ERK pathway and found that pERK was upregulated in our NP-siGP130–treated bladder tumors, indicating that this downstream pathway was still active and may be responsible in preventing complete abrogation of the bladder tumor. Hepburn and colleagues demonstrated that activation of the ERK pathway supported a role for cancer stem cell self-renewal and tumorigenicity (29).
For the first time, we report that GP130 is linked to the aggressiveness of human bladder tumors and that blocking GP130 using surface-modified NPs encapsulating siRNA GP130 may have a therapeutic role in controlling tumor growth. Therefore, future studies may involve dual-loaded NPs (siGP130 and siERK) that will have the ability to target both GP130 and ERK. This may be a more effective strategy for completely eradicating aggressive bladder tumors.
Disclosure of Potential Conflicts of Interest
No potential conflicts of interest were disclosed.
Authors' Contributions
Conception and design: D.T. Martin, J.M. Steinbach-Rankins, W.M. Saltzman, R.M. Weiss
Development of methodology: D.T. Martin, J.M. Steinbach-Rankins, W.M. Saltzman
Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): D.T. Martin, H. Shen, J.M. Steinbach-Rankins, K.K. Johnson, J. Syed
Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): D.T. Martin, J.M. Steinbach-Rankins, X. Zhu, J. Syed, R.M. Weiss
Writing, review, and/or revision of the manuscript: D.T. Martin, H. Shen, J.M. Steinbach-Rankins, K.K. Johnson, J. Syed, W.M. Saltzman, R.M. Weiss
Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): D.T. Martin, X. Zhu
Study supervision: D.T. Martin, R.M. Weiss
Acknowledgments
The authors thank Marcia Wheeler for helpful discussions with this article. We also thank our Genitourinary Data and Biospecimen Repository team [Drs. Martha Boeke and Brian Shuch (principal investigator)] for the management of the Biospecimen data and providing the GU specimens.
This study was supported in part by the NIH (grant no. 5RC1DK087015, to R.M. Weiss) from the National Institute of Diabetes and Digestive and Kidney Diseases, NIH (grant no. EB000487, to W.M. Saltzman), and a pilot grant from the Yale Comprehensive Cancer Center (to D.T. Martin and R.M. Weiss).
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Footnotes
Note: Supplementary data for this article are available at Molecular Cancer Therapeutics Online (http://mct.aacrjournals.org/).
- Received November 14, 2017.
- Revision received March 1, 2018.
- Accepted October 9, 2018.
- Published first October 31, 2018.
- ©2018 American Association for Cancer Research.