Bladder cancer is the fourth most commonly diagnosed cancer in the United States with over 60,000 new cases per year [1
]. Fortunately, the majority of these cancers are superficial and successfully treated surgically. Unfortunately, these patients require intense follow-up due to high recurrence rates and the potential for progression to invasive cancer. Cystoscopy is recommended at regular intervals and even the lowest risk patients have a 30% recurrence rate at 5
]. This constant need for surveillance imposes economic and life style hardship. An effective therapy to decrease bladder cancer recurrence could have significant impact on both quality of life and survival for over 500,000 patients with a history of bladder cancer in the United States alone.
Post-translational histone modifications such as acetylation are associated with transcriptionally active regions of the genome. Histone deacetylation appears to be a mechanism whereby cancers decrease expression of genes involved in cell cycle control and apoptosis. Histone deacetylase inhibitors (HDACi) are an emerging class of cancer drugs that might be useful in preventing bladder cancer recurrence. Valproic acid (sodium valproate) is a relatively weak HDACi but has demonstrated potential in the treatment of glioblastomas [3
], thyroid cancer [4
], and leukemia [5
]. There are several on-going clinical trials of valproate for the treatment of other cancers registered on ClinicalTrials.gov. Extensve clinical experience with valproate as a seizure medication demonstrates that it is generally a well-tolerated drug that can be administered for long periods. For these reasons valproate is an attractive candidate for the prevention of bladder cancer recurrence.
Anti-neoplastic properties of valproate in bladder cancer models have recently been reported by several groups. Valproate decreased proliferation (or viability) of TCC-SUP, T24, RT4, and HT1376 cell lines; increased histone H3 acetylation and p21 expression and activated caspase 2 and caspase 3 in T24 cells [6
]. In addition, in vitro invasiveness was decreased in valproate treated T24, TCC-SUP, and HT1376 cells. This is not restricted to in vitro studies: T24 xenografts had reduced growth with chronic administration of valproate (0.4% in drinking water for 35
days) in male athymic nu/nu mice. Similar results were reported by Byun et al. [7
] for TCC-SUP and 5637 cell lines (identified in the publication by their ATCC Numbers as HTB5 and HTB9 respectively). Histone deacetylase 1 is expressed at higher levels in human bladder cancer compared to normal urothelium and its expression is also increased in the BBN mouse bladder cancer model [8
]. These authors also reported delayed BBN-induced bladder tumors in mice. Valproate decreased proliferation in UMUC3, RT112, TCCSUP, and RT4 bladder cancer cell lines and, increased the percentage of cells in the G1 phase of the cell cycle with concomitant changes in cell cycle regulatory proteins [9
Thrombospondin-1 (TSP1) is a well known natural inhibitor of angiogenesis. TSP1 anti-angiogenesis activity is mediated at least in part through the CD36 receptor, which initiates a cascade of events culminating in death of endothelial cells [10
]. TSP1 expression in the urinary bladder is altered in bladder cancer and associated with low nuclear p53, increased tumor recurrence, and decreased survival [11
]. Cultured bladder cancer cell lines stimulated to migrate and neovascularization showed lower TSP1 expression compared to normal urothelial cells, suggesting that bladder tumors may selectively down regulate TSP1 thus promoting angiogenesis [12
We have previously shown that TSP1 expression is reduced in the bladders of UPII-SV40T transgenic mice relative to wildtype littermates [13
]. UPII-SV40T mice develop bladder cancer due to urothelium-specific expression of the simian virus 40
T antigen protein [14
]. Tumor growth was reduced and TSP1 expression increased by castration. One of us (CS) investigating the teratogenic properties of valproate noted that TSP1 expression was enhanced in embryos carried by dams treated with valproate (unpublished results). We speculated that the anti-angiogenic action of valproate might be due to increases in TSP1 expression in addition to a direct effect on cancer cell proliferation.
Here we report that valproate does induce TSP1 expression in bladder cancer cell lines and that this is likely mediated through HDAC inhibition. The latter was evidenced by increased TSP1 expression in response to another HDAC inhibitor vorinostat (SAHA).