Bacillus Calmette-Guérin Induces Cellular Reactive Oxygen Species and Lipid Peroxidation in Cancer Cells

Basic and Translational Science Bacillus Calmette-Guérin Induces Cellular Reactive Oxygen Species and Lipid Peroxidation in Cancer Cells Juwita N. Rahmat, Kesavan Esuvaranathan, and Ratha Mahendran OBJECTIVE

METHODS

RESULTS

CONCLUSION

To determine whether Bacillus Calmette-Guérin (BCG) and/or BCG-soluble factors could modulate cellular reactive oxygen species (ROS) in human bladder cancer cells and the impact this could have on response to therapy. The expression of ␣5␤1 integrins on human bladder cancer cell lines and their ability to internalize BCG were determined. The effect of live and lyophilized BCG on cellular ROS, lipid peroxidation, and DNA damage was determined using H2DCF-DA, TBARS, and comet assays. The cytotoxic effects of live and lyophilized BCG on cancer cells were determined after 24 hours. ROS modulation by Antigen 85B and mycobacterial protein tyrosine phosphatases was monitored. Live and lyophilized BCG were internalized to a similar extent, but live BCG increased cellular ROS, whereas lyophilized BCG reduced ROS. High ROS levels correlated with increased lipid peroxidation. The cytotoxic effect of BCG was independent of cellular ROS but dependent on internalization. Lyophilized BCG was more cytotoxic to bladder cancer cells than live BCG. BCG soluble factors such as Antigen85B could increase cellular ROS. Internalization of lyophilized BCG abrogated the ROS, and lipid peroxidation increase induced by BCG soluble factors. Both live and lyophilized BCG induced DNA damage but to different extents. The end products of ROS, such as lipid peroxides and superoxide, could induce DNA damage, which could lead to mutations in cancer cells that select for their survival. Reducing BCG instillations may reduce the risk of mutational changes occurring in remnant cancer cells. UROLOGY 79: 1411.e15–1411.e20, 2012. © 2012 Elsevier Inc.

B

ladder cancer is characterized by frequent recurrences that can lead to disease progression. Mycobacterium bovis, Bacillus Calmette-Guérin (BCG), is superior to chemotherapy for the therapy of nonmuscle invasive superficial bladder cancer following removal of the tumor.1-3 It induces a nonspecific inflammatory response that may remove remnant tumor cells, one possible cause of disease recurrence.1,3 Patients who fail BCG immunotherapy are at higher risk for disease progression,3 but the reason for this is unknown. This study evaluates whether direct/indirect interactions of BCG and bladder cancer cells could be a cause of failure. Direct interaction of BCG with the bladder wall is necessary for successful immunotherapy,4 and this re-

Supported by a grant from the BMRC (BMRC/04/1/21/19/311) and the Urology Fund, Department of Surgery. Financial Disclosure: The authors declare that they have no relevant financial interests. From the Department of Surgery, Yong Loo Lin School of Medicine, National University of Singapore; and Department of Urology, National University Health System, Singapore Reprint requests: Ratha Mahendran, Ph.D., Department of Urology, National University Health System, Singapore, Kent Ridge Road, NUHS Tower Block, Level 8, Singapore 119 228. E-mail: [email protected] Submitted: June 14, 2011, accepted (with revisions): January 11, 2012

© 2012 Elsevier Inc. All Rights Reserved

quires fibronectin5 and the ␣5␤1 integrins.6 However, the role of BCG interactions with cancer cells during patient therapy is not clear. In vitro studies have shown that by crosslinking the ␣5␤1 integrins BCG causes cell cycle arrest.7 The ␣5␤1 integrins are also necessary for BCG internalization6 and cell death.8 Lyophilized BCG induced a decrease in cellular reactive oxygen species (ROS)9 in human bladder cancer cell lines. In contrast, it was recently reported that live BCG increased cellular ROS in a lung epithelial cancer cell line.10 High levels of ROS have been implicated in cell signaling events, leading to cell migration as well as the induction of DNA damage which could result in tumor metastasis and progression or cancer cell death. Lyophilized BCG preparations are used clinically for ease of quality control, storage, and transport of the microorganism. The process of lyophilization results in a decrease in bacterial viability and in the disruption of the microbial cell wall such that intracellular components may be exposed. Akaza et al reported that live, lyophilized, and killed BCG preparations could all reduce tumor growth.11 However, the contrasting results obtained in vitro on lung and bladder cancer cell lines may indicate either differences in the response of cancer cell lines from 0090-4295/12/$36.001411.e15 doi:10.1016/j.urology.2012.01.017

different tissue sources or differences in the response induced by live and lyophilized BCG. This study evaluates the effect of live and lyophilized BCG on cancer cells with respect to microbe internalization, cellular ROS, lipid peroxidation (LPO), cytotoxicity, and DNA damage, to determine how the interaction of BCG with cancer cells could influence the outcome of therapy. The effect of BCG-soluble proteins, namely Antigen 85B (Ag85B) and mycobacterial protein-tyrosine-phosphatase A (MPtpA), on cellular ROS were determined. During bladder cancer therapy, BCG is instilled in the bladder for 2 hours, but BCG has been found in the urine after 24 hours and even several days post-therapy.12 Therefore, cells were exposed to BCG for 2 and 24 hours.

MATERIALS AND METHODS Growth and Maintenance of Bladder Cancer Cell Lines Human transitional cell carcinoma cell lines MGH, RT4, J82 and SW-780 (ATCC, Manassas, VA) were routinely cultured in RPMI (BioWest, Nuaillé, France) supplemented with 10% heat-inactivated fetal bovine serum (BioWest), 2 mmol/L L-glutamine, 50 U/mL penicillin G, and 50 ␮g/mL streptomycin (Invitrogen, Carlsbad, CA), whereas UM-UC-3 cells (ATCC) were cultured in Dulbecco’s modified Eagle’s medium (DMEM) (Sigma, St. Louis, MO). Cells were maintained at 37°C in a 5% CO2 atmosphere and routinely passaged when 85% to 90% confluent.

Preparation of BCG and the BCG Internalization Assay Lyophilized BCG, Connaught strain, was purchased from Aventis Pasteur, Canada and used to prepare live BCG by plating on 7H9 Middlebrook medium (BD, Franklin Lakes, NJ) supplemented with 10% ADC supplement, 0.05% Tween 80, and 0.2% glycerol and agar and selecting colonies. For lyophilized BGC, OD 440 nm of 0.2 is equivalent to 3 ⫻ 106 cfu (cfu)/mL. For live BCG, OD 600 nm of 0.1 ⫽ 2.6 ⫻ 106 cfu/mL, and the bacteria were harvested in the exponential phase. BCG was labeled with FITC (Merck, Darmstadt, Germany) as described by de Boer et al.13 Cells were plated at a density of 1 ⫻ 105 cells per well in a 24-well plate. After 6 hours, the media was removed and replaced with blank media overnight. The following day, cells were treated for 2 hours and 24 hours with FITC-labeled live BCG at a 1:100 cells-to-bacteria ratio. Cells were harvested and surface BCG detected with anti-BCG antibodies13 and TRITC conjugated swine antirabbit antibodies (1:50 dilution) (Dako, Glostrup, Denmark). Cells were washed and fixed in 0.1% formalin before analysis by flow cytometry.

␣5␤1 Analysis and Cell Proliferation Assays Cells were harvested and collected by centrifugation and resuspended in 100 ␮L of PBA (1 ⫻ PBS with 1% BSA and 0.01% sodium azide) with 0.2 ␮g of antibodies to either ␣5, ␤1 integrins, or isotype controls and incubated for 20 minutes at 4°C. Cells were washed to remove unbound antibodies, followed by centrifugation at 7500 g for 1 minute. The pellet was re-suspended in 50 ␮L of PBA and incubated with 2 ␮L of 1411.e16

secondary antibody (rabbit antimouse IgG, RPE conjugated) at 4°C for 20 minutes. Cells were washed and fixed with 0.1% formalin before flow cytometry. Cells were labeled with bromodeoxyuridine (BrdU; Sigma) and the assay was performed as previously described.8 In short, labeled cells were incubated with BCG at a 1:100 cells-tobacteria ratio for 24 hours, and anti-BrdU antibodies (BD) and flow cytometry were used to enumerate live cells.8 Necrotic and apoptotic cells would have reduced BrdU signals.

Measuring ROS Cells were plated at a density of 3 ⫻ 105 cells per well in a six-well plate and incubated overnight at 37°C. The next day, 1 ⫻ 107 cfu/mL BCG washed twice with PBS and re-suspended in fresh medium was added to the cells and incubated for 2 or 24 hours. To block direct interaction between BCG and the cells, a cell culture insert (0.4 ␮m, BD), which prevents the migration of BCG, was placed in the culture dish. ROS was measured as previously described using H2DCF-DA (Invitrogen).9 The fluorescence signal was detectable at an excitation wavelength 488 nm and emission wavelength 535 nm with FACS Canto flow cytometer (BD). For treatment with mycobacterial secreted factors, cells were incubated with Antigen 85B (Sigma) at 1 ␮g/mL or MPtpA (0.5 ␮g/mL or 1 ␮g/mL) for 2 hours.

LPO ASSAY MGH cells (2 ⫻ 106 cells) were plated in a 10-mm culture dish, and 24 hours later the cells were exposed to 2 ⫻ 108 cfu of BCG for 2 hours. Cells were washed 3 times with 1 ⫻ PBS and collected by centrifugation (450 g) for 5 minutes, and the pellet was re-suspended in 0.4 mL of 20 mmol/L Tris, pH 7.4, with 5 mmol/L butylated hydroxytoluene (dissolved in ethanol; Sigma). The cells were lysed by sonication on ice using an Ultrasonic cell disrupter (Misonix, Farmingdale, NY) at a power output of 5 W for 3 rounds of 5 seconds. The homogenate was centrifuged at 1250 g for 10 minutes, and the supernatant was collected for protein determination and LPO assay. For transwell blocking of BCG interaction, 4 ⫻ 105 MGH cells were plated in 6-well plates, and 24 hours later a cell culture insert was placed in the well and BCG (4 ⫻ 107 cfu) was placed in the top chamber. LPO levels were measured by the detection of thiobarbituric acid (TBA)—malondialdehyde (MDA) adduct formed. To each sample, 200 ␮L of 8.1% SDS, 1.5 mL of 20% glacial acetic acid, pH 3.5, 1.5 mL of 0.8% thiobarbituric acid (dissolved in 50 mmol/L NaOH; Sigma), and 0.6 mL of water were added, and mixture was heated at 95°C for 1 hour then cooled to room temperature with tap water. The chromogen was extracted with 1 ml of water, then 5 ml n-butanol:pyridine (15:1 ratio) mixture. The samples are centrifuged at 1250 g for 10 minutes, and the organic top layer was read at OD 532 nm with a glass cuvette using an ultraviolet (UV) spectrophotometer. The concentration of MDA (expressed in nmol per mg protein) was calculated based on the following formula: Absorbance (OD 532 nm) ⫽ ␧ ⫻ concentration ⫻ 1(path UROLOGY 79 (6), 2012

Table 1. Percentage of cells with internalized BCG and high ROS after BCG treatment

Cell Line MGH J82 UMUC-3 RT4 SW780

% of Cells With High ROS Lyo BCG Control

Time (h)

% Cells With BCG

Control

2 24 2 24 2 24 2 24 2 24

21.0 ⫾ 1.9 43.1 ⫾ 7.7* 16.3 ⫾ 3.1 27.2 ⫾ 3.4* 60.2 ⫾ 6.4 82.1 ⫾ 4.4* 10.1 ⫾ 0.5 11.1 ⫾ 0.8 5.7 ⫾ 0.8 17.9 ⫾ 1.1**

60.2 ⫾ 2.4 62.5 ⫾ 0.7 47.2 ⫾ 2.0 50.4 ⫾ 4.7 76.3 ⫾ 1.3 72.1 ⫾ 2.2 13.9 ⫾ 1.9 14.0 ⫾ 0.5 13.3 ⫾ 2.2 22.8 ⫾ 1.9

36.9 ⫾ 4.5* 46.9 ⫾ 0.7* 39.4 ⫾ 0.9** 33.8 ⫾ 4.5** 54.3 ⫾ 3.8** 48.4 ⫾ 2.7** 15.2 ⫾ 0.8 16.5 ⫾ 2.7 13.5 ⫾ 1.8 25.2 ⫾ 2.12

59.6 ⫾ 2.1 59.2 ⫾ 1.8 46.5 ⫾ 2.0 39.6 ⫾ 1.8 57.7 ⫾ 1.0 67.4 ⫾ 1.4 16.4 ⫾ 2.6 13.7 ⫾ 1.8 29.3 ⫾ 2.1 21.0 ⫾ 2.3

Live BCG 73.7 ⫾ 1.1* 78.2 ⫾ 2.0* 71.2 ⫾ 4.8** 58.6 ⫾ 0.8** 51.6 ⫾ 3.3 53.5 ⫾ 2.3** 23.7 ⫾ 1.4* 21.6 ⫾ 1.7* 29.4 ⫾ 2.1 21.7 ⫾ 0.7

Data are percentage mean ⫾ SEM. The experiments were performed twice in triplicate (n ⫽ 6). * Significant difference with respect to 2 hours (*P ⬍ .05, ** P ⬍ .001).

length), where ␧ is the characteristic absorptivity of MDA; ␧ ⫽ 1.56 ⫻ 105 mol/L⫺1cm⫺1.14 Comet Assay This assay was performed as described by Bajpayee et al15. Tail moment was calculated using the CometScore software (Tritek CometScore Freeware v1.5). Three microscope fields were examined at ⫻4 magnification, and 100 comets were scored per sample. The experiment was performed twice. Purification of MPtpA The plasmid construct pGEX mptpA was provided by Axel Ullrich (Martinsreid, Munich) from the Department of Molecular Biology, Max-Planck-Institut für Biochemie, and MPtpA was purified as described by Koul et al16. Pure MPtpA without GST was obtained by treatment with FactorXa and passing the protein through an HiTrap benzamidine column (GE Healthcare, Buckinghamshire, England) and eluting MPtpA with the High Salt elution buffer (20 mM sodium phosphate, 1 mol/L NaCl). Its purity was confirmed by SDS-PAGE and Coomassie staining. MPtpA was incubated with 10 mmol/L dithiothreitol (DTT) (Bio-Rad, Hercules, CA) for 30 minutes at room temperature to reverse oxidation in the cysteine active site of the enzyme and then dialyzed against 20 mmol/L sodium phosphate buffer, pH 7.5, to remove excess DTT. The protein concentration was determined with the Micro BCA Protein Assay Kit (Thermo Fisher Scientific, Rockford, IL). Statistical Analysis Comparisons among 3 or more groups were performed with descriptive analysis of variance with Bonferroni analysis using SPSS 17.0 software (SPSS Inc., Chicago, IL). For comparisons between 2 groups, an independent t test was used. A P value less than 0.05 was considered statistically significant.

RESULTS Lyo/Live BCG Internalization Most of the cells expressed ␤1 integrins (ranging from 86.8% ⫾ 2.9% in SW780 to 57.1% ⫾ 9.7% in J82 cells). UROLOGY 79 (6), 2012

However, ␣5 expression was variable, with 57.3% ⫾ 3.3% of MGH cells, 47.4% ⫾ 2.7% of J82 cells, 76% ⫾ 2.4% of UMUC-3 cells, 1.5% ⫾ 0.5% of RT4, and 6.4% ⫾ 2.7% SW780 cells being positive. Most cells showed increased BCG internalization with time, except RT4 which had a very low percentage of cells with ␣5 expression (Table 1). Both live and lyophilized BCG were internalized within 2 hours; however, by 24 hours, more lyophilized BCG and/or BCG debris was bound to the surface of the cells.

Lyophilized/Live BCG Induce Differential Effects on Cellular ROS Lyophilized and live BCG induced changes in cellular ROS are shown in Figure 1. Lyophilized BCG induced a significant decrease in the number of cells with high cellular ROS at both 2 and 24 hours (Table 1). However, live BCG caused an increase in cellular ROS above that in control cells, in all cell lines except UMUC-3 (Table 1). UMUC-3 cells responded to live BCG by reducing cellular ROS.

Blocking Direct Contact Between BCG and Cells Modulates ROS In the presence of the blocking membrane, lyophilized BCG resulted in an increase in cellular ROS rather than a return to control levels (Fig. 2A). This increase in ROS was similar to that obtained with live BCG with and without membrane blocking. This indicates that soluble factor(s) associated with BCG (lyophilized or live) caused the increase in ROS, but that internalization of lyophilized BCG blocked the effect of the soluble factor(s) and triggered a decrease in ROS. Lipid peroxidation, a downstream by product of H2O2 interaction with membrane lipids, was measured using the TBARS assay to confirm that the ROS increase was due to increased H2O2 levels (Figure 2B). As expected, lyophilized BCG reduced lipid peroxidation and live BCG increased it, and blocking direct contact between lyophilized BCG and the cells increased lipid peroxidation, Figure 2B. 1411.e17

Figure 1. Typical histogram showing cellular ROS. MGH cells were treated with H2DCF-DA and analyzed by flow cytometry. The low and high ROS marker settings were based on the profile of control cells. The percentage of cells in the high ROS fraction was monitored after treatment with lyophilized and live BCG. A drop in the percentage of cells in this fraction resulted in an increase in cells in the low ROS fraction. Thus, only the percentage of cells in the high ROS fraction was reported.

The ability of MPtpA and Ag85B, 2 BCG-secreted proteins, to modulate cellular ROS was assessed. Ag85B by itself increased cellular ROS, but MPtpA had no effect on ROS levels (Fig. 2C). Thus, Ag85B could perhaps be one of the factors that contribute to the increase in cellular ROS. Blocking Direct Contact Protects Cells From the Cytotoxic Effect of BCG Although both live and lyophilized BCG induced different cellular levels of ROS, both induced cell death (Fig. 3A). Lyophilized BCG was more cytotoxic than live BCG. Blocking BCG internalization abrogated the cytotoxic effect. These data indicate that the ROS changes may not be responsible for cell death. BCG Induces DNA Damage in Bladder Cancer Cells MGH cells exposed to BCG for 24 hours showed increased DNA damage (Fig. 3B). Live BCG induced more DNA damage than lyophilized BCG, and blocking BCG internalization reduced the amount of DNA damage that was observed with both BCG preparations. However, if cells were exposed to BCG for only 2 hours and allowed to recover for 22 hours, there was minimal DNA damage (data not shown), indicating that the DNA damage was repaired.

Figure 2. BCG-soluble factors induce ROS and lipid peroxidation in cancer cells. MGH cells were exposed to live and lyophilized BCG for 2 hours in the presence and absence of a blocking membrane, and the effects that this has on (A) cellular ROS and (B) lipid peroxidation were determined. These results indicate that soluble mycobacterial factors are responsible for the observed ROS increase. To test this hypothesis, the effect of (C) Ag85B and MPtpA on cellular ROS was determined. In A and B, *P ⬍ .05 and **P ⬍ .001 denote significance with respect to control. In the presence of the transwell device, there was a significant difference with respect to samples in direct contact with lyophilized BCG. In C, double asterisk (**) denotes significance with respect to control of P ⬍ .01. lyo BCG ⫽ lyophilized BCG.

COMMENT Both lyophilized and live BCG are internalized to a similar extent, but induce different ROS levels in the cells. Thus, the differences in ROS reports previously published are possibly due to the bacterial preparation and not the cell lines evaluated. In the absence of contact between BCG and cells, both BCG preparations induced an increase in the proportion of cells with high 1411.e18

ROS, indicating that BCG soluble factors may increase ROS. Known BCG-secreted molecules include mycobacterial protein tyrosine phosphatases,16 superoxide dismutase,17 Lipoarabinomannan (LAM),18 and the Antigen 85 complex.19 LAM activates cellular signal transduction pathways but not ROS.18 Superoxide dismutase generates UROLOGY 79 (6), 2012

Figure 3. BCG induces cytotoxic effects and DNA damage in cancer cells. Bladder cancer cells were treated with live and lyophilized BCG for 24 hours, with or without the presence of cell culture inserts to prevent physical contact. (A) Cytotoxicity assays were performed using BrDU-labeled cells. Cells were harvested, and BrDU-labeled DNA was detected by flow cytometry using an antibody to BrDU. Data are presented as the percentage of cells over the control of 1 set of the experiments (mean ⫾ SEM). *Significantly different with respect to control (*P ⬍ .05, **P ⬍ .005). Experiments were performed twice in triplicate (n ⫽ 6 for MGH cells) or duplicate (n ⫽ 4 for UMUC3 and SW780 cells). (B) DNA damage was assessed using the Comet assay with the alkaline single cell gel electrophoresis method. Comets were scored using the CometScore Software v1.5 from Tritek Corporation. Three fields from each sample slide were chosen for scoring, and at least 100 comets were scored from each slide. Tail moment was taken as a measure of DNA damage. Cells were treated with H2O2 for 10 minutes as a positive control and untreated cells served as a control for basal levels of DNA damage. Asterisk (*) denotes P ⬍ 0.001.

oxygen and hydrogen peroxide from superoxide, and could have increased ROS in cells that internalized BCG9; however, without BCG, superoxide dismutase could not have contributed to increased ROS. In this study Ag85B induced ROS. It was previously reported that Ag85B induced nitric oxide (NO), which induces ROS in alveolar macrophages.20 Given the rapid (2 hours) drop in ROS, it is likely that either cellular proteins are modulated by BCG membrane proteins8 (as most gene expression changes take longer than 2 hours to induce new protein synthesis) or that BCG membrane proteins reduce cellular ROS. Support for the latter possibility comes from studies in macrophages, in which phagocytosed BCG has been shown to release the fiUROLOGY 79 (6), 2012

bronectin attachment protein and proteins from the antigen 85 complex into other subcellular compartments in the cell.21 The ROS marker used, H2DCF-DA, measured H2O2 and peroxyl radicals, and the measurement of lipid peroxides confirmed the generation of H2O2. In a previous study, other ROS species, such as superoxide and nitric oxide (NO) were found to be induced by BCG.9 Thus, is it likely that related radical species as well as hydroxyl radicals could be generated in the cells by BCG, but these would not have been detected by H2DCF-DA. LPO produces aldehydes, such as acrolein, malondialdehyde (MDA), and 4-hydroxy-2-nonenal (HNE). Accumulation of these end-products in large quantities can be carcinogenic and mutagenic because of DNA and protein damage.22 However, in small quantities, these molecules can act as signaling messengers.23 Superoxide can also induce DNA damage.24 In most cells, ROS are rapidly removed; thus a short, 2-hour exposure to BCG, although it increased lipid peroxidation and superoxide levels, induced only minimal DNA damage. This was similar to findings in a previous report that DNA damage could be repaired after a short exposure to H2O2 in rat urothelial cells.25 However, with a longer exposure of 24 hours, the damage induced by both live and lyophilized BCG in bladder cancer cells were not repaired. This was probably because the DNA repair system was overwhelmed. When DNA damage cannot be repaired, cells generally undergo apoptosis or some form of cell death. This might be one of the causes of the cytotoxic effect of internalized BCG on bladder cancer cells. When contact between BCG and the cells was blocked, there was less DNA damage, and so the cells survived. It was previously reported that the p21 expression is necessary for BCG’s direct cytotoxic effect on bladder cancer cells.26 The p21 protein induces cell cycle arrest until DNA damage is repaired; if it cannot be repaired, cells die. Cells exposed to BCG for 24 hours may have died because of the presence of extensive DNA damage. Besides BCG, activated immune cells can also generate ROS. Immune cells attracted to the bladder and activated by BCG are in the bladder for a longer duration than the BCG. Thus it is possible that, during therapy, the remnant cancer cells are exposed to high levels of ROS for a long time. As cancer cells often have mutations in the DNA damage response and repair pathways and/or apoptosis pathways, ROS-induced DNA damage may not be repaired efficiently. This could induce DNA mutations that may increase resistance to therapy or even enable metastasis.27 It was recently reported that, in bladder cancer patients who received BCG immunotherapy, a polymorphism in an antioxidant enzyme gene, glutathione peroxidase 1 (GPX1), correlated with earlier recurrences.28 This enzyme prevents oxidative DNA damage by reducing hydroperoxides and reducing cellular ROS. This does support a role for ROS, generated by either BCG or immune cells in the bladder environment, 1411.e19

inducing disease progression. Multiple weekly BCG instillations,3 causing repeated increases in ROS, could lead to accumulated DNA damage and increased mutations in remnant cancer cells. This may explain why some patients who fail to respond to BCG immunotherapy are at increased risk for disease progression.3,27,29

13.

14. 15.

CONCLUSIONS Reducing either the number of instillations of BCG that patients receive or the dose of BCG may reduce the amount of ROS and DNA damage and could lead to reduced disease progression. De Boer et al30 showed, in a murine model, that reducing BCG instillation to week 1 and week 6 still ensured sufficient TH1 cytokine expression and reduced TH2 cytokines. This should be explored in a tumor model. References 1. Herr HW, Morales A. History of Bacillus Calmette-Guerin and bladder cancer: an immunotherapy success story. J Urol. 2008;179: 53-56. 2. Lamm DL. Efficacy and safety of Bacille Calmette-Guerin immunotherapy in superficial bladder cancer. Clin Infect Dis. 2000; 31(Suppl 3):S86-S90. 3. O’Donnell MA. Optimizing BCG therapy. Urol Oncol. 2009;27: 325-328. 4. Kavoussi LR, Brown EJ, Ritchey JK, et al. Fibronectin-mediated Calmette-Guerin bacillus attachment to murine bladder mucosa. Requirement for the expression of an antitumor response. J Clin Invest. 1990;85:62-67. 5. Ratliff TL, Palmer JO, McGarr JA, et al. Intravesical Bacillus Calmette-Guerin therapy for murine bladder tumors: initiation of the response by fibronectin-mediated attachment of Bacillus Calmette-Guerin. Cancer Res. 1987;47:1762-1766. 6. Kuroda K, Brown EJ, Telle WB, et al. Characterization of the internalization of Bacillus Calmette-Guerin by human bladder tumor cells. J Clin Invest. 1993;91:69-76. 7. Chen F, Zhang G, Iwamoto Y, et al. BCG directly induces cell cycle arrest in human transitional carcinoma cell lines as a consequence of integrin cross-linking. BMC Urol. 2005;5:1-8. 8. Pook SH, Rahmat JN, Esuvaranathan K, et al. Internalization of Mycobacterium bovis, Bacillus Calmette-Guerin, by bladder cancer cells is cytotoxic. Oncol Rep. 2007;18:1315-1320. 9. Pook SH, Esuvaranathan K, Mahendran R. N-acetylcysteine augments the cellular redox changes and cytotoxic activity of internalized Mycobacterium bovis in human bladder cancer cells. J Urol. 2002;168:780-785. 10. Méndez-Samperio P, Pérez A, Torres L. Role of reactive oxygen species (ROS) in Mycobacterium bovis Bacillus Calmette-Guerinmediated up-regulation of the human cathelicidin LL-37 in A549 cells. Microb Pathog. 2009;47:252-257. 11. Akaza H, Iwasaki A, Ohtani M, et al. Expression of antitumor response. Role of attachment and viability of Bacillus CalmetteGuerin to bladder cancer cells. Cancer. 1993;72:558-563. 12. Siatelis A, Houhoula DP, Papaparaskevas J, et al. Detection of Bacillus Galmette-Guerin (Mycobacterium bovis BCG) DNA in

1411.e20

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

urine and blood specimens after intravesical immunotherapy for bladder carcinoma. J Clin Microbiol. 2011;49:1206-1208. de Boer EC, Bevers RF, Kurth KH, et al. Double fluorescent flow cytometric assessment of bacterial internalization and binding by epithelial cells. Cytometry. 1996;25:381-387. Buege JA, Aust SD. Microsomal lipid peroxidation. Methods Enzymol. 1978;52:302-310. Bajpayee M, Dhawan A, Parmar D, et al. Gender-related differences in basal DNA damage in lymphocytes of a healthy Indian population using the alkaline Comet assay. Mutat Res. 2002;520: 83-91. Koul A, Choidas A, Treder M, et al. Cloning and characterization of secretory tyrosine phosphatases of Mycobacterium tuberculosis. J Bacteriol. 2000;182:5425-5432. Piddington DL, Fang FC, Laessig T, et al. Cu,Zn superoxide dismutase of Mycobacterium tuberculosis contributes to survival in activated macrophages that are generating an oxidative burst. Infect Immun. 2001;69:4980-4987. Sirkar M, Majumdar S. Lipoarabinomannan-induced cell signaling involves ceramide and mitogen-activated protein kinase. Clin Diagn Lab Immunol. 2002;9:1175-1182. Wiker HG, Harboe M. The antigen 85 complex: a major secretion product of Mycobacterium tuberculosis. Microbiol Rev. 1992;56: 648-661. Sable SB, Goyal D, Verma I, et al. Lung and blood mononuclear cell responses of tuberculosis patients to mycobacterial proteins. Eur Respir J. 2007;29:337-346. Beatty WL, Russell DG. Identification of mycobacterial surface proteins released into subcellular compartments of infected macrophages. Infect Immun. 2000;68:6997-7002. Esterbauer H, Schaur RJ, Zollner H. Chemistry and biochemistry of 4-hydroxynonenal, malonaldehyde and related aldehydes. Free Radic Biol Med. 1991;11:81-128. Forman HJ. Reactive oxygen species and alpha,beta-unsaturated aldehydes as second messengers in signal transduction. Ann N Y Acad Sci. 2010;1203:35-44. Thomas S, Lowe JE, Hadjivassiliou V, et al. Use of the Comet assay to investigate the role of superoxide in glutathione-induced DNA damage. Biochem Biophys Res Commun. 1998;243:241-245. Wang A, Robertson JL, Holladay SD, et al. Measurement of DNA damage in rat urinary bladder transitional cells: improved selective harvest of transitional cells and detailed Comet assay protocols. Mutat Res. 2007;634:51-59. See WA, Zhang G, Chen F, et alWA, Zhang G, Chen F et al: p21 expression by human urothelial carcinoma cells modulates the phenotypic response to BCG. Urol Oncol 2010;28:526533. Carretero R, Cabrera T, Gil H, et al. BCG immunotherapy of bladder cancer induces selection of HLA class I-deficient tumor cells. Int J Cancer. 2011;129:839-846. Chiong E, Kesavan A, Mahendran R, et al. NRAMP1 and hGPX1 gene polymorphism and response to Bacillus Calmette-Guerin therapy for bladder cancer. Eur Urol. 2011;59:430-437. Lerner SP, Tangen CM, Sucharew H, et al. Failure to achieve a complete response to induction BCG therapy is associated with increased risk of disease worsening and death in patients with high risk non-muscle invasive bladder cancer. Urol Oncol. 2009;27:155159. de Boer EC, Rooyakkers SJ, Schamhart DH, et al. BCG dose reduction by decreasing the instillation frequency: effects on local Th1/Th2 cytokine responses in a mouse model. Eur Urol. 2005;48: 333-338.

UROLOGY 79 (6), 2012