Glutathione-S-Transferase Theta 2 (GSTT2) Modulates the Response to Bacillus Calmette-Guérin Immunotherapy in Bladder Cancer Patients

Abstract

Glutathione-S-transferases (GST) enzymes detoxify xenobiotics and are implicated in response to anticancer therapy. This study evaluated the association of GST theta 1 (GSTT1), GSTT2, and GSTT2B with Mycobacterium bovis Bacillus Calmette-Guérin (BCG) response in non-muscle-invasive bladder cancer treatment. In vitro assessments of GSTT2 knockout (KO) effects were performed using cell lines and dendritic cells (DCs) from GSTT2KO mice. Deletion of GSTT2B, GSTT1, and single-nucleotide polymorphisms in the promoter region of GSTT2 was analysed in patients (n = 205) and healthy controls (n = 150). Silencing GSTT2 expression in MGH cells (GSTT2B^FL/FL) resulted in increased BCG survival (p < 0.05) and decreased cellular reactive oxygen species. In our population, there are 24.2% with GSTT2B^Del/Del and 24.5% with GSTT2B^FL/FL. With ≤ 8 instillations of BCG therapy (n = 51), 12.5% of GSTT2B^Del/Del and 53.8% of GSTT2B^FL/FL patients had a recurrence (p = 0.041). With ≥9 instillations (n = 153), the disease recurred in 45.5% of GSTT2B^Del/Del and 50% of GSTT2B^FL/FL. GSTT2^FL/FL patients had an increased likelihood of recurrence post-BCG therapy (HR 5.5 [1.87-16.69] p < 0.002). DCs from GSTT2KO mice produced three-fold more IL6 than wild-type DCs, indicating a robust inflammatory response. To summarise, GSTT2B^Del/Del patients respond better to less BCG therapy and could be candidates for a reduced surveillance regimen.

Keywords: BCG vaccine; bladder cancer; genetic; glutathione S-transferase theta 2; immunotherapy; polymorphism; urinary bladder neoplasms.

Figure 1

GSTT2 expression in bladder cancer cell lines and its impact on cellular ROS, cytotoxicity, BCG survival, and DNA damage after exposure to BCG. (A) Real-time analysis of GSTT2 gene expression after silencing with siGSTT2 and siControl in MGH cells. (B) RT-PCR detection of GSTT2 gene expression in UMUC3 cells after transfection with a plasmid carrying the GSTT2 gene, pBudGSTT2, and pBudCE4.1 (empty vector). (C) The impact of GSTT2 downregulation on the percentage of MGH cells with high ROS (measured with H₂DCFDA) after exposure to lyophilised BCG. (D) The impact of GSTT2 expression on the percentage of UMUC-3 cells with high ROS levels after stimulation with lyophilised BCG. A p-value < 0.05 is designated as * and < 0.001 as **. (E,F) The cytotoxic effect of BCG on MGH and UMUC3 cells in the presence and absence of GSTT2 expression. ** represents a value < 0.001 and * p-value < 0.05. (G–J) Intracellular survival of live BCG in (G) MGH, (H) UMUC3, and (I) U937 cells after blocking or overexpression of GSTT2. The cells were GSTT2 silenced or induced to overexpress GSTT2, followed by a 2 h exposure to BCG, removing excess BCG for the 24 h analysis. The number of BCG colony-forming units that survived was determined by lysing the cells and plating the BCG on agar plates and (J) performing a live/dead bacteria assay on MGH cells. * represents a p-value < 0.05. (K) Photos of DNA comets observed in MGH cells with and without GSTT2 silencing. The green arrows indicate the nucleus head and the yellow arrows indicate the comet trail comprising damaged DNA. (L) Graph showing the calculated tail moments using CometScore v1.5. Experiments were performed twice in triplicate (n = 6). A minimum of 20 comets were analysed per sample. * represents a p-value < 0.05 and ** a p-value < 0.01.

Figure 2

Clinical impact of GSTT2B and GSTT2 in bladder cancer. (A–E) Kaplan–Meier plots showing cumulative survival with time. (A) Time to recurrence segregated by GSTT2B genotypes. (B) Time to recurrence segregated by the number of BCG instillations received (i.e., ≤8 versus ≥9), (C) time to recurrence, (D) progression, and (E) survival for subjects with the GSTT2B^Del/Del segregated by the number of BCG instillations received. The Log-rank p-value is noted on each graph, and the number of samples (N) is indicated. The dotted line on graph (A) indicates that most recurrences in the cohort of patients with the GSTT2B^Del/Del genotype occurred within 2 years of starting therapy. (F) Data extracted from the TCGA database for GSTT2 expression levels in tumour and normal bladder tissues, (G) disease-free survival with respect to high and low GSTT2 mRNA expression in tumour tissue, and (H) disease-free survival in papillary and (I) non-papillary tumours with respect to GSTT2 expression levels.

Figure 3

Schematic of the impact of GSTT2 expression on BCG immunotherapy of bladder cancer. (1) The internalisation of BCG particles (2) induces lipid peroxidation and (3) GSTT2 expression upregulation to catalyse the breakdown of lipid peroxidation by-products. (4) When GSTT2 was transiently overexpressed, lyophilised BCG treatment, which usually induced ROS reduction, resulted in a higher net ROS level than normal cells without GSTT2 expression treated similarly with BCG. (5) BCG survival was impaired in GSTT2 overexpressed bladder cancer cells, possibly due to the generation of lipid peroxides (from bacteria or host cells) and increased ROS. The enhanced BCG clearance in these cells reduces subsequent inflammatory immune response. (6) GSTT2 overexpression led to a baseline increase in DNA damage without BCG treatment. (7) Lyophilised BCG treatment of GSTT2-silenced cells led to a lower net ROS level than normal GSTT2-expressing cells treated similarly with BCG. (8) There was significantly more BCG surviving intracellularly in GSTT2-silenced cells, which enhanced inflammatory response and cytokine production. (9) GSTT2 silencing led to a baseline decrease in DNA damage without BCG treatment. This image was created with BioRender.com.