HMGB1 Release by Urothelial Carcinoma Cells is Required for the In Vivo Antitumor Response to Bacillus Calmette-Guérin

HMGB1 Release by Urothelial Carcinoma Cells is Required for the In Vivo Antitumor Response to Bacillus Calmette-Guérin

Investigative Urology HMGB1 Release by Urothelial Carcinoma Cells is Required for the In Vivo Antitumor Response to Bacillus Calmette-Guérin Guangjia...

630KB Sizes 0 Downloads 65 Views

Investigative Urology

HMGB1 Release by Urothelial Carcinoma Cells is Required for the In Vivo Antitumor Response to Bacillus Calmette-Guérin Guangjian Zhang, Fanghong Chen, Yanli Cao, Bryon Johnson and William A. See* From the Departments of Urology and Pediatrics (BJ), Medical College of Wisconsin, Milwaukee, Wisconsin

Purpose: Prior series showed that a portion of urothelial carcinoma cells exposed to bacillus Calmette-Guérin undergoes nonapoptotic cell death and release of the chemokine HMGB1. We evaluated the role of tumor cell derived HMGB1 in mediating the in vivo antitumor effect of bacillus Calmette-Guérin. Materials and Methods: The murine urothelial carcinoma cell line MB49 was engineered to express a shRNA construct targeting HMGB1. The shRNA expressing cell line underwent characterization to ensure its comparability to the parental MB49 cell line. An orthotopic tumor model was used to compare the in vivo antitumor efficacy of bacillus Calmette-Guérin in the parental cell line (24 control and 24 bacillus Calmette-Guérin treated) vs the HMGB1 knockdown line (23 control and 21 treated). Results: Expression of the shRNA construct decreased HMGB1 expression and its release in response to bacillus Calmette-Guérin. The parental and shRNA cell lines showed similar in vitro doubling time and cytotoxicity in response to bacillus Calmette-Guérin. Treatment significantly decreased tumor volume vs controls in parental MB49 tumor bearing mice (p ⫽ 0.036). Tumor volume in treated mice inoculated with the shRNA cell line was higher than that in sham treated shRNA controls (p ⫽ 0.12). Of the bacillus Calmette-Guérin treated mice tumor volume was significantly lower in parental tumor bearing mice vs the shRNA group (p ⬍0.00001). ANOVA revealed a significant interaction between the cell line (shRNA vs parental) and the bacillus Calmette-Guérin effect (p ⫽ 0.0076). Conclusions: The direct tumor response to bacillus Calmette-Guérin, culminating in HMGB1 release, may be an important contributor to the clinical efficacy of bacillus Calmette-Guérin.

Abbreviations and Acronyms BCG ⫽ bacillus Calmette-Guérin ELISA ⫽ enzyme-linked immunosorbent assay HMGB1 ⫽ high molecular group box protein 1 LDH ⫽ lactate dehydrogenase PBS ⫽ phosphate buffered saline UC ⫽ urothelial carcinoma Accepted for publication September 26, 2012. Study received institutional animal care committee approval. Supported by a Department of Veterans Affairs grant and the Milwaukee Veterans Affairs Medical Center. * Correspondence: Department of Urology, Medical College of Wisconsin, 9200 West Wisconsin Ave., Milwaukee, Wisconsin 53226 (telephone: 414-805-0787; FAX: 414-805-0771).

Key Words: urinary bladder, urothelium, carcinoma, BCG vaccine, HMGB1 protein INTRAVESICAL administration of BCG is the standard of care for high risk, nonmuscle invasive UC of the bladder.1–3 Through the combination of a direct effect on tumor cell biology and the induction of a cellular host immune response, BCG results in a significant antitumor response. From the bladder perspective the antitumor response includes efferent and afferent pathways. The efferent pathway results from direct BCG ef-

fects on UC cells, including signaling pathway activation, gene transactivation, cytokine expression and phenotypic changes.4 – 8 BCG exposure also results in nonapoptotic cell death in a subpopulation of cells and release of the potent chemokine HMGB1.9,10 These elements of the direct tumor cell response to BCG combine to form an efferent pathway, signaling to the host the need to mount an afferent immune response.

0022-5347/13/1894-1541/0 THE JOURNAL OF UROLOGY® © 2013 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION

http://dx.doi.org/10.1016/j.juro.2012.09.123 Vol. 189, 1541-1546, April 2013 RESEARCH, INC. Printed in U.S.A.

AND

www.jurology.com

1541

1542

HMGB1 RELEASE BY UROTHELIAL CARCINOMA CELLS AND BACILLUS CALMETTE-GUÉRIN

To determine the importance of tumor derived/ BCG induced factors, we focused on the role of HMGB1 in mediating the antitumor effect of BCG. We noted that knockdown of HMGB1 release by UC cells is associated with a loss of the BCG antitumor effect. These findings suggest a vital role of nonapoptotic cell death and associated HMGB1 release in mediating the in vivo antitumor response to BCG.

MATERIALS AND METHODS Cell Lines and BCG We used the murine UC cell line MB49. Cells were maintained at 37C, 5% CO2 in RPMI 1640 (Gibco®) supplemented with 10% fetal bovine serum, penicillin and streptomycin. In these experiments TICE BCG (Organon, West Orange, New Jersey) was used. Freeze-dried BCG was reconstituted at an estimated concentration of 2.5 ⫻ 107 viable organisms per ml.

HMGB1 Knockdown Using shRNA/HMGB1 Construct HMGB1 expression was knocked down by transfecting parental MB49 cells with the shRNA/HMGB1 lentiviral plasmid pLKO.1-puro (Sigma®) using Lipofectamine™. The shRNA plasmid contains the sequence CCGGGCACAGCACAAATTAGTTATACTCGAGTAT-AACTAATTTGTGCTGTGCTTTTTG targeting the 3= untranslated region of mouse HMGB1. Puromycin resistant clones were selected and expanded. HMGB1 expression of each clone was evaluated by Western blot using Anti-HMGB1/HMG-1 Antibody 07-584 (EMD Millipore Bioscience, Billerica, Massachusetts), as previously described.10 Positive clones, defined as those with a greater than 75% decrease in protein vs the parental line, were selected to create the HMGB1 knockdown cell line.

Luciferase Expressing Cell Line Creation Parental MB49 and shRNA/HMGB1 knockdown cells were transfected with pGL4.51[luc2/CMG/Neo] Vector (Promega®). Positive clones were selected, expanded and used in all experiments.

Assays HMGB1 ELISA. HMGB1 levels in cell culture supernatant 24 hours after BCG treatment (1:50 cell-to-BCG ratio) were measured by a commercially available HMGB1 ELISA kit (Shino-Test, Sagamihara, Japan). Cell proliferation. Cell proliferation was measured using the MTT assay, as previously described.11 Dye exclusion for cell viability. At 24 and 72 hours after adding BCG (1:50 cell-to-BCG ratio) we evaluated cell viability using trypan blue dye. Cells were incubated with 0.4% trypan blue for 3 minutes before counting. Trypan blue uptake indicated loss of membrane integrity. LDH release. LDH is a stable cytosolic enzyme that is released upon cell lysis. LDH in culture supernatants was measured 24 hours after BCG exposure (1:50 cell-to-BCG ratio) using the commercially available CytoTox 96® Assay Kit.

In Vivo Animal Model All animal experiments were approved by the institutional animal care committee. Tumor was implanted in C57BL/6 (albino) female mice at ages 7 to 9 weeks on experimental day 0, as described by Ninalga et al.12 Mice underwent intravesical treatment with BCG (5 ⫻ 106 organisms in 50 ␮l) or PBS (50 ␮l) on days 1, 4, 8 and 12. Biophotonic imaging was performed on days 7 and 14 to assess tumor take and interval growth. Mice were sacrificed on day 15. The bladders were excised and fixed in formalin. At 48 hours the bladders were bisected, blotted dry and weighed. Bladder weight served as the primary end point for the treatment effect. For the parental cell line group 24 mice were treated with PBS and 24 were treated with BCG. For the shRNA cell line group 23 mice were treated with PBS and 21 were treated with BCG. This experiment was repeated twice. Results were consistent between experiments and combined for statistical analysis.

Statistical Analysis Except as noted, all experiments were done in triplicate. Differences between treatment groups were evaluated using ANOVA for repeat measures with results considered significant at p ⬍0.05. For in vivo experiments tumor weight measurement data were analyzed using 2-way ANOVA, adjusting for the experimental run. Based on published data and the initial analysis of the data from this experiment, log transformation was used to stabilize measurement variability. Interaction testing in the ANOVA model answered the primary question of comparing the effect of treatment on control and HMBG1 silenced animals. The Mann-Whitney U test was used to compare tumor weight between treatment groups.

RESULTS shRNA expression targeting HMGB1 mRNA resulted in decreased intracellular HMGB1. Relative to the parental line, MB49 cells with constitutive expression of the shRNA construct targeting HMGB1 demonstrated markedly less HMGB1 protein in cell lysates. Figure 1 shows a representative Western blot of HMGB1 from parental and shRNA expressing cells. Actin served as the lane loading control. HMGB1 release in response to BCG was decreased in shRNA expressing cell lines. Figure 2 shows the results of ELISA for HMGB1 in cell culture supernatant 24 hours after exposure to BCG. In the parental line BCG exposure resulted in a statistically significant nineteenfold increase in HMGB1 concentration (p ⬍0.01). Relative to the parental line, HMGB1 release by shRNA expressing cell lines was decreased from that in controls and BCG exposed cells. HMGB1 release by shRNA clones in response to BCG was decreased sixteenfold vs the parental cell line. Notably, relative to untreated controls the shRNA expressing cell lines showed significantly increased HMGB1 release in response to BCG (average 2.8-fold), indicating persistent non-

Relave Number of Metabolically Acve Cells

HMGB1 RELEASE BY UROTHELIAL CARCINOMA CELLS AND BACILLUS CALMETTE-GUÉRIN

1543

40 shrna

35

parental

30 25 20 15 10 5 0 Day 0

Day 1

Day 2

Day 3

day 4

Figure 3. In vitro calculated doubling times of parental MB49 clone and shRNA knockdown clone were virtually identical at 30.9 and 29.6 hours, respectively. Number of metabolically active cells did not differ as function of time in 2 cell lines (p ⫽ 0.44).

Figure 1. Western blot shows HMGB1 in cell lysates from shRNA clone and parental MB49 line. Constitutive expression of shRNA construct targeting mouse HMGB1 markedly decreased HMGB1 protein. Actin served as lane loading control.

apoptotic cell death in these lines in response to BCG (ANOVA p ⬍0.05). shRNA expressing clones and the parental cell line had similar in vitro doubling time. On MTT assay the calculated doubling time of the parental and knockdown cell lines was 30.9 and 29.6 hours, respectively (fig. 3). The number of metabolically

3

Control BCG

Relave HMGB1 Concentraon

2.5

2

1.5

1

0.5

0

Parental

shRNA Clone 92

shRNA Clone 96

Figure 2. HMGB1 release from MB49 cells in response to BCG was decreased in cells constitutively expressing shRNA construct targeting HMGB1. Parental clone 7 was compared to shRNA expressing clones 92 and 96. Only parental clone showed statistically significant increase in HMGB1 vs controls in response to BCG (p ⬍0.05). Optical density values represent relative HMGB1 concentration in cell culture medium 24 hours after exposure to BCG or control. Values represent mean absorbance ⫹ SE.

active cells present as a function of time did not significantly differ between the 2 cell lines (ANOVA p ⫽ 0.44). Cellular cytotoxicity in response to BCG was not decreased by shRNA knockdown of HMGB1. The shRNA expressing cell line underwent extensive analysis of known intermediate end points of BCG cytotoxicity to ensure that the loss of HMGB1 release was a consequence of decreased expression and not of resistance to BCG (fig. 4). On vital dye exclusion assay the number of parental cells with loss of membrane integrity in response to BCG was increased twofold and 4.8-fold over that of control cells on posttreatment days 1 and 3, respectively (fig. 4, A). In contrast, in the shRNA expressing group there were 2.4 and 4.5-fold increases vs controls on days 1 and 3, respectively. BCG treatment significantly increased the number of cells with loss of membrane integrity in each group at each time point (ANOVA p ⬍0.001). Relative to the parental MB49 cell line, shRNA expressing cells had increased sensitivity to BCG on day 1 (ANOVA p ⬍0.001). The BCG effect on membrane integrity of the cell lines did not differ on day 3. BCG increased LDH release on day 1 posttreatment by 2.6 and 7.5-fold in parental and HMGB1shRNA cells, respectively, vs controls. On day 3 these relative values were 1.7 and 1.9-fold, respectively (fig. 4, B). BCG treatment significantly increased the magnitude of LDH release in each group at each time point (ANOVA p ⬍0.05). The BCG effect on the cell lines did not differ on day 1 or 3. On MTT assay adding BCG decreased the number of metabolically active cells in the parental and shRNA lines to 79% and 80% that of controls on day 1, and to 59% and 61%, respectively, on posttreatment day 3 (fig. 4, C). BCG treatment significantly decreased the number of metabolically active cells in each group at each time point (ANOVA p ⬍0.05).

1544

HMGB1 RELEASE BY UROTHELIAL CARCINOMA CELLS AND BACILLUS CALMETTE-GUÉRIN

0.12

0.1

PBS

Bladder Weight (gms)

BCG 0.08

0.06

0.04

0.02

0 Parental

shRNA Cell Line

Figure 5. shRNA knockdown of HMGB1 resulted in loss of antitumor response to BCG. Mice inoculated with parental MB49 cell line had significant decrease in intravesical tumor volume in response to BCG vs PBS treated controls (p ⫽ 0.036). Mice inoculated with shRNA construct targeting HMGB1 showed nonsignificant increase in intravesical tumor volume after BCG vs PBS treated controls (p ⫽ 0.12). Tumor volume did not significantly differ in parental vs shRNA mice treated with PBS. Tumor volume in shRNA mice treated with BCG was significantly greater than in BCG treated parental mice (p ⫽ 0.00001). Values are shown as mean ⫹ SE bladder weight. Interaction between shRNA vs parental tumor cell line and BCG effect was statistically significant (ANOVA p ⫽ 0.0076).

Figure 4. BCG exposure resulted in comparable in vitro toxicity in parental and shRNA producing MB49 cell lines. A, number of cells with membrane integrity loss in response to BCG on vital dye exclusion was increased 2.8 and 4.8-fold in controls on days 1 and 3 after treatment vs 2.4 and 4.5-fold, respectively, in shRNA expressing group. BCG significantly increased number of cells with membrane integrity loss in each group at each time point (p ⬍0.001). shRNA expressing cells had increased sensitivity to BCG on day 1 vs parental MB49 cell line (p ⬍0.001). BCG effect did not differ in cell lines on day 3. B, BCG increased LDH release on day 1 by 2.6 and 7.5-fold in parental and HMGB1shRNA cells, and on day 3 by 1.7 and 1.9-fold, respectively. BCG significantly increased LDH release in each group at each time point (p ⬍0.05). BCG effect did not differ in cell lines on day 1 or 3. C, on MTT assay adding BCG decreased number of metabolically active cells in parental and shRNA to 79% and 80% that of controls on day 1, and 59% and 61%, respectively, on day 3. BCG significantly decreased number of metabolically active cells in each group at each time point (p ⬍0.05). BCG effect on cell lines did not differ on day 1 or 3.

The BCG effect on the cell lines did not differ on day 1 or 3. HMGB1 knockdown was associated with loss of antitumor responsiveness to BCG in vivo. The incidence of overall tumor take in the orthotopic model was 95.7%. Mice inoculated with the parental MB49 cell line showed a significant decrease in intravesical tumor volume in response to BCG vs controls (p ⫽ 0.036, fig. 5). Mice inoculated with shRNA cells showed a nonsignificant increase in intravesical tu-

mor volume after BCG treatment vs controls (p ⫽ 0.12). Tumor volume did not significantly differ between parental and shRNA treated mice that received PBS (p ⫽ 0.12). Tumor volume was significantly greater in shRNA treated mice that received BCG than in parental treated mice that received BCG (p ⫽ 0.00001, see table). ANOVA revealed significant interaction between tumor cell line (shRNA vs parental) and the BCG treatment effect, indicating that the effect of BCG depended on whether the parental line or the shRNA line was used (p ⫽ 0.0076).

DISCUSSION Today the predominant paradigm for cancer therapy is to view the cancer cell as a passive entity to be targeted by the antitumor agent. Treatment leading to nonapoptotic cell death (necrosis) and Statistical comparison of bladder weight by treatment group Cell Line/Treatment

p Value

Parental/control vs: Parental/BCG shRNA/control shRNA/control vs shRNA/BCG Parental/BCG vs shRNA BCG

0.036 00.12 0.12 0.00001

HMGB1 RELEASE BY UROTHELIAL CARCINOMA CELLS AND BACILLUS CALMETTE-GUÉRIN

HMGB1 release may represent a fundamental change in this model. Nonapoptotic cell death can influence local tumor biology and the systemic host response. Paracrine release of HMGB1 in concert with the expression of tumor cell surface receptors for HMGB1 translates into local effects.13 Systemically, HMGB1 activates macrophages, induces neutrophil chemotaxis and alters vascular endothelial cell biology.14,15 Whether these local/systemic influences of HMGB1 contribute positively or negatively to tumor biology and treatment outcome is a subject of debate.16 HMGB1 stimulates tumor growth and metastasis in some systems while appearing to be growth inhibitory in others.17–20 In the clinical setting intravesical BCG administration increases urinary HMGB1. In vitro studies demonstrated HMGB1 release by human UC cells in response to BCG exposure. These observations, coupled with the ambiguity surrounding the friend or foe role of HMGB1,21 prompted the current study. The extensively characterized MB49 line served as the basis for developing the clonal cell lines used in these experiments. In vitro comparisons of the parental and HMGB1 knockdown clones revealed comparable doubling times. In vitro sensitivity of the parental and knockdown cell lines to BCG was similar, as measured by vital dye exclusion, LDH release and metabolically active cell number on day 3 after BCG exposure. Day 1 results suggest early, increased in vitro sensitivity of the knockdown clone to BCG with statistically increased loss of membrane integrity associated with numerically increased LDH release. While to our knowledge the etiology of this early sensitivity is unknown, it may relate to the HMGB1 role of facilitating gene transcription, an important component of the direct cellular response to BCG.22 Predictably, in vitro HMGB1 release in response to BCG was decreased in HMGB1 knockdown cells relative to the parental line. However, HMGB1 knockdown cells continued to manifest a necrosis response to BCG, as evidenced by a significant increase in HMGB1 release vs controls. In our orthotopic model our results reveal that HMGB1 knockdown cells/tumors failed to respond in vivo to intravesical BCG. Treatment was begun on postimplantation day 1 to maximize BCG access

1545

to implanted tumor, analogous to clinical carcinoma in situ. This strategy is widely used in studies using orthotopic models.23–25 Possibly the early initiation of treatment limits clinical relevance. However, the experimental design of our study, with a positive treatment effect in the parental control arm compared to treatment failure in the HMGB1 knockdown line, provides strong evidence of an essential role for HMGB1 release in this model. These findings are biologically consistent with the immune dependence of BCG activity. The in vivo antitumor efficacy of BCG requires an intact host immune system and T-cell response.26 Acute HMGB1 release in response to BCG could serve as the impetus for an afferent host immune response. Acute HMGB1 release combined with known BCG induced tumor cell secretion of other cytokines is distinct from the chronic tumor necrosis/HMGB1 release that occurs without simultaneous cytokine/chemokine secretion. This may explain the apparently positive role of HMGB1 in BCG treatment for bladder cancer. Notably, these statements represent hypothesis. Additional studies are required to obtain insight into the mechanism by which HMGB1 contributes to the BCG antitumor effect. When considering the clinical relevance of our results, several possibilities come to mind. HMGB1 could be used to monitor the tumor response to BCG or as a target for clinical manipulation. Most importantly, nonapoptotic cell death and its associated HMGB1 release may serve as in vitro surrogates for treatment efficacy. While prior studies of the BCG mechanism of action identified elements needed for a tumor response, to our knowledge this is the first report to identify a tumor specific variable on which the BCG treatment effect depends.27,28 These findings provide new insight into how the tumor cell could be used as part of the therapeutic armamentarium. In conclusion, HMGB1 release by tumor cells appears to be an important element of the direct BCG effect on UC cells and one that is required for an antitumor effect.

ACKNOWLEDGMENTS Dr. Timothy Ratliff, Purdue University Center for Cancer Research, provided the MB49 cell line.

REFERENCES 1. Babjuk M, Oosterlinck W, Sylvester R et al: EAU guidelines on non-muscle-invasive urothelial carcinoma of the bladder, the 2011 update. Eur Urol 2011; 59: 997. 2. Kassouf W, Kamat AM and Zlotta A: Canadian guidelines for treatment of non-muscle invasive

bladder cancer: a focus on intravesical therapy. Can J Urol 2010; 4: 168.

Update 2007. Linthicum: American Urological Association Education and Research 2007.

3. Hall MC, Chang SS, Dalbagni G et al: Guideline for the Management of Non-muscle Invasive Bladder Cancer: (Stages Ta, T1, and Tis): AUA

4. Zhang G, Chen F and See WA: Bacillus CalmetteGuérin induces p21 expression in human transitional carcinoma cell lines via an immediate

1546

HMGB1 RELEASE BY UROTHELIAL CARCINOMA CELLS AND BACILLUS CALMETTE-GUÉRIN

early, p53 independent pathway. Urol Oncol 2007; 25: 221. 5. Chen FH, Crist SA and Zhang GJ: Interleukin-6 production by human bladder tumor cell lines is up-regulated by bacillus Calmette-Guérin through nuclear factor-kappaB and Ap-1 via an immediate early pathway. J Urol 2002; 168: 786. 6. Chen F, Zhang G, See WA et al: MB49 murine urothelial carcinoma: molecular and phenotypic comparison to human cell lines as a model of the direct tumor response to bacillus CalmetteGuerin. J Urol 2009; 182: 2932. 7. Chen F, Zhang G and See WA: BCG directly induces cell cycle arrest in human transitional carcinoma cell lines as a consequence of integrin cross-linking. BMC Urol 2005; 5: 8. 8. Chen F, Zhang G and See WA: Bacillus CalmetteGuerin inhibits apoptosis in human urothelial carcinoma cell lines in response to cytotoxic injury. J Urol 2007; 178: 2166.

12. Ninalga C, Loskog A and Klevenfeldt M: CpG oligonucleotide therapy cures subcutaneous and orthotopic tumors and evokes protective immunity in murine bladder cancer. J Immunol 2005; 28: 20. 13. Ellerman JE, Brown CK and de Vera M: Masquerader: high mobility group box-1 and cancer. Clin Cancer Res 2007; 13: 2836. 14. Andersson U, Wang H and Palmblad K: High mobility group 1 protein (HMG-1) stimulates proinflammatory cytokine synthesis in human monocytes. J Exp Med 2000; 192: 565. 15. Treutiger CJ and Mullins GE: High mobility group 1 B-box mediates activation of human endothelium. J Immunol Methods 2003; 254: 375. 16. Zeh HJ 3rd and Lotze MT: Addicted to death: invasive cancer and the immune response to unscheduled cell death. J Immunol 2005; 28: 1.

9. Raucci A, Palumbo R and Bianchi ME: HMGB1: a signal of necrosis. Autoimmunity 2007; 40: 285.

17. Kuniyasu H, Chihara Y and Kondo H: Amphotericin induction in prostatic stromal cells by androgen deprivation is associated with metastatic prostate cancer. Oncol Rep 2003; 10: 1863.

10. See WA, Zhang G and Chen F: Bacille-Calmette Guerin induces caspase-independent cell death in urothelial carcinoma cells together with release of the necrosis-associated chemokine high molecular group box protein 1. BJU Int 2009; 103: 1714.

18. Huttunen HJ, Fages C, Kuja-Panula J et al: Receptor for advanced glycation end products-binding COOH-terminal motif of amphotericin inhibits invasive migration and metastasis. J Can Res 2002; 62: 4805.

11. Mossman T: Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays. J Immunol Methods 1983; 65: 55.

19. He Q, Liang CH and Lippard SJ: Steroid hormones induce HMG1 overexpression and sensitize breast cancer cells to cisplatin and carboplatin. Proc Natl Acad Sci U S A 2000; 97: 5768.

20. Tang D, Kang R and Lotze MT: High-mobility group box 1 and cancer. Biochim Biophys Acta 2010; 1799: 131. 21. Ulloa L and Messmer D: High-mobility group box 1 (HMGB1) protein: friend and foe. Cytokine Growth Factor Rev 2006; 17: 189. 22. Andersson U, Erlandsson-Harris H, Yang H et al: HMGB1 as a DNA-binding cytokine. J Leukoc Biol 2002; 72: 1084. 23. Mossberg AK, Hou Y and Svensson M: HAMLET treatment delays bladder cancer development. J Urol 2010; 183: 1590. 24. Bockholt NA, Knudson MJ, Henning JR et al: Anti-interleukin-10R1 monoclonal antibody enhances bacillus Calmette-Guérin induced T-helper type 1 immune responses and antitumor immunity in a mouse orthotopic model of bladder cancer. J Urol 2012; 187: 2228. 25. Leite KR, Chade DC and Sanudo A: Effects of curcumin in an orthotopic murine bladder tumor model. Int Braz J Urol 35: 599. 26. Ratliff TL: Role of the immune response in BCG for bladder cancer. Eur Urol, suppl., 1992; 21: 17. 27. Brandau S, Riemensberger J and Böhle A: Role of IFN gamma and IL-10 in immunotherapy of bladder cancer with Mycobacterium bovis BCG. Immunobiology, 2000; 203: 350. 28. Böhle A and Brandau S: Immune mechanisms in bacillus Calmette-Guerin immunotherapy for superficial bladder cancer. J Urol 2003; 170: 964.