DNA repair genes and prognosis in sporadic forms of urothelial carcinoma of the upper urinary tract

INTRODUCTION

Lynch syndrome or hereditary nonpolyposis colorectal cancer is caused by mutations in DNA repair genes, known as mismatch repair (MMR) genes, and is associated with microsatellite instability. Urothelial carcinoma of the renal pelvis is also associated with this syndrome. These genetic abnormalities have been described in sporadic forms of upper tract urothelial carcinoma (UTUC).

MATERIAL AND METHOD

This was a descriptive study and survival analysis of a series of 80 patients with sporadic UTUC with no metastases at diagnosis (N0/Nx M0) treated exclusively with nephroureterectomy. We evaluated the expression of MMR genes (hMLH1, hPMS2, hMSH2 and hMSH6) in sections performed with tissue microarray (TMA) and their association with clinical-pathological parameters. We analyzed the prognostic value of the loss of expression of these genes in UTUC.

RESULTS

We detected no loss of MSH2 or of MSH6, but there was a loss of MLH1 in 11 cases (13.8%) and of PMS2 in 21 cases (26.3%). The expression of hMLH1 and hPMS2 were strongly associated (P<.0001), and this phenotype expression entails significant clinical implications. The loss of MLH1 was associated with a low grade (P=.02). Loss of PMS2 was associated with a lower stage (P=.05), a pushing pattern with no invasive edges (P=.008) and less angiogenesis (P=.008). The inactivation of hPMS2 or hMLH1 is an independent protective factor (HR, 0.309) and, along with the histologic grade (HR, 5.561), defines the patients' prognosis.

CONCLUSION

In our experience, the inactivation of hPMS2 or hMLH1 is an independent marker of good prognosis and occurs in a quarter of sporadic UTUC cases. The immunohistochemical study of these patients can be used to assess the screening of hidden forms of Lynch syndrome.

The study of DNA methylation in urological cancer: present and future

OBJECTIVES

We have synthesized the principal advances in the field of the study of epigenetics and specifically DNA methylation regarding the diagnosis of urological neoplasms.

ACQUISITION OF EVIDENCE

Review of the literature (PubMed, MEDLINE and Cochrane) on the study of DNA methylation in urological neoplasms (prostate cancer, bladder cancer, renal cancer and testicular cancer), considering all the studies published up to January 2013.

SYNTHESIS OF EVIDENCE

It was possible to determine the state of methylation of many genes in our tumor samples. When these were compared with healthy tissue samples, it was possible to define the specific aberrant methylation patterns for each type of tumor. The study and definition of specific abnormal methylation patterns of each type of tumor is a tool having potential utility for diagnosis, evaluation, prediction of prognosis and treatment of the different forms of genitourinary cancer. The analysis of gene methylation in urine after micturition or post-prostatic massage urine, semen, in the wash plasma or fluid from prostatic biopsies may allow early detection of bladder, prostate, renal and testicular cancer. In each one of the neoplasms, an epigenetic signature that may be detected in the DNA has been identified, obtained from very scarce or not at all invasive specimens, with potential in the diagnosis and evaluation of prognosis. Validation of these studies will confirm the accuracy, effectiveness and reproducibility of the results available up to now. Criteria have still not been developed that determine if a gene panel provides sufficient information in the health care practice to guide an unequivocal diagnosis or therapeutic conduct. More studies are needed to compare sensitivity, specificity, positive and negative predictive value of the test in each case. Multicenter studies analyzing the real reproducibility of these results in a clinical setting also do not exist.

CONCLUSIONS

The study of aberrant DNA methylation in biological specimens of patients has an enormous potential for the early diagnosis and screening of genitourinary neoplasms. A larger number of studies is needed to be able to define the series of genes that would mean unequivocal signatures of malignancy. This methodology also has potential when defining prognostic groups and potential of response to different therapies.

Abstract S4-8: Promoter CpG Methylation of BRCA1 Predicts Sensitivity to PARP Inhibitors in Breast Cancer

Background: Poly(ADP)-ribose polymerase (PARP) inhibitors were shown to selectively kill BRCA1/2 mutated cancer cells due to conferring synthetic lethality, leading to first clinical trials in BRCA1 germline mutated breast and ovarian cancer with promising results. However, inherited breast and ovarian cancers are relatively rare. In sporadic breast cancer aberrant promoter methylation of BRCA1 is a more frequent event, intriguingly contributing to a common “BRCA phenotype”, as determined by high similarity of gene expression patterns between BRCA1 inherited and sporadic breast tumors. Currently, it is unknown whether BRCA1 methylated breast cancer cells are comparably sensitive to PARP inhibition like BRCA1 mutated breast cancer cells.

Methods: We screened a panel of 7 breast cancer cell lines for BRCA1/2 promoter methylation by bisulfite genomic sequencing, none of which harbored methylation in BRCA2. One cell line (UACC3199) revealed dense methylation in the BRCA1 gene promoter. Diminished BRCA1 protein expression in these cells was re-established after treatment with 1 µM of the DNA demethylating agent 5-aza-2'-deoxycytidin. For further analysis, UACC3199 cells were compared with MDA-MB-231 cells (BRCA1 wildtype) and MDA-MB-436 cells (BRCA1 homozygous mutant).

Results: In XTT assays the PARP inhibitors 3-ABA, DPQ and NU1025 revealed a similar toxicity in BRCA1 deficient UACC3199 and MDA-MB-436 cells, whereas BRCA1 proficient MDA-MB-231 cells were more resistant (IC50 values for Mda-MB-231, MDA-MB-436, and UACC3199 cells for 3-ABA: 8775 µM, 37 µM, 35 µM; for DPQ: 26 µM, 12 µM, 17 µM; for NU1025: 746 µM, 162 µM, 301 µM, respectively). Confocal immunofluorescent microscopy showed that after PARP inhibition, Y-H2AX focalization was strongly increased but not significantly different among all cell lines, indicating that the amount of DNA damage conferred by inhibition of PARP is independent of BRCA1 status. By comet assays after one week of PARP inhibition, however, we found that the amount of persistent DNA damage was significantly enhanced in both BRCA1 deficient lines, whereas it was low and similar to controls in MDA-MB-231 cells. This argues for defects in DNA damage repair in both BRCA1 deficient cell lines, most likely implicating disrupted homologous recombination integrity. Since BRCA1 mutations are associated with the triple-negative breast cancer subtype, we also determined the frequency of BRCA1 promoter methylation in non-inherited triple-negative breast cancers by methylation-specific PCR. Of the analyzed samples, a high fraction showed hypermethylation in BRCA1 (n=25/68; 37%). Conclusions: In conclusion, our results show for the first time that in addition to BRCA1 or BRCA2 mutation, also BRCA1 methylated breast cancer cell lines are considerably more sensitive to PARP inhibitors than BRCA1 wildtype cells. We therefore suggest to include BRCA1 methylation as a potential biomarker of drug sensitivity against these compounds in current and future prospective clinical trials of breast and ovarian cancer. Further in vitro experiments on the effects of PARP inhibition in BRCA1 methylated breast cancer cells are underway.

Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr S4-8.

Knockdown of protein tyrosine phosphatase SHP-1 inhibits G1/S progression in prostate cancer cells through the regulation of components of the cell-cycle machinery

SHP-1, a haematopoietic cell-specific tyrosine phosphatase, is also expressed in human prostate. In this study, we report that SHP-1 depletion in PC-3 cells induced by small interfering RNAs causes G1 phase cell-cycle arrest accompanied by changes in some components of the cell-cycle machinery. SHP-1 knockdown increases p27(Kip1) (p27) protein stability, its nuclear localization and p27 gene transcription. These effects could be mediated by PI3K-AKT pathway as SHP-1 interacts with PI3K regulating its activity and p110 catalytic subunit phosphorylation. The increase in p27 protein stability could also because of reduced cyclin-dependent kinase (CDK2) activity. SHP-1 knockdown decreases the CDK6 levels, inducing retinoblastoma protein hypophosphorylation, downregulation of cyclin E and thereby a decrease in the CDK2 activity. However, the codepletion of SHP-1 and p27 does not produce re-entry into the cycle, implying that p27 is not required to maintain cell-cycle arrest induced by SHP-1 depletion. The maintenance of the PC-3 cell anti-proliferative response after p27 loss could be because of mislocalization of CDK2 induced by SHP-1 knockdown. This study shows that SHP-1 depletion promotes cell-cycle arrest by modulating the activity of cell-cycle regulators and suggests that SHP-1 may be required for the proper functioning of events governing cell-cycle progression.

Salermide, a Sirtuin inhibitor with a strong cancer-specific proapoptotic effect

Sirtuin 1 (Sirt1) and Sirtuin 2 (Sirt2) belong to the family of NAD+ (nicotinamide adenine dinucleotide-positive)-dependent class III histone deacetylases and are involved in regulating lifespan. As cancer is a disease of ageing, targeting Sirtuins is emerging as a promising antitumour strategy. Here we present Salermide (N-{3-[(2-hydroxy-naphthalen-1-ylmethylene)-amino]-phenyl}-2-phenyl-propionamide), a reverse amide with a strong in vitro inhibitory effect on Sirt1 and Sirt2. Salermide was well tolerated by mice at concentrations up to 100 muM and prompted tumour-specific cell death in a wide range of human cancer cell lines. The antitumour activity of Salermide was primarily because of a massive induction of apoptosis. This was independent of global tubulin and K16H4 acetylation, which ruled out a putative Sirt2-mediated apoptotic pathway and suggested an in vivo mechanism of action through Sirt1. Consistently with this, RNA interference-mediated knockdown of Sirt1, but not Sirt2, induced apoptosis in cancer cells. Although p53 has been reported to be a target of Sirt1, genetic p53 knockdowns showed that the Sirt1-dependent proapoptotic effect of Salermide is p53-independent. We were finally able to ascribe the apoptotic effect of Salermide to the reactivation of proapoptotic genes epigenetically repressed exclusively in cancer cells by Sirt1. Taken together, our results underline Salermide's promise as an anticancer drug and provide evidence for the molecular mechanism through which Sirt1 is involved in human tumorigenesis.

Transforming pathways unleashed by a HDAC2 mutation in human cancer

Although disruption of histone modification patterns is a common hallmark of human cancer, our knowledge of the mechanistic role of histone-modifying enzymes in its generation is very limited. We have recently identified an inactivating mutation in the histone deacetylase-2 (HDAC2) in sporadic carcinomas with microsatellite instability and in tumors arising in individuals with hereditary nonpolyposis colorectal cancer syndrome. Since HDAC2 seems to be a central player in epigenetic gene repression, we wondered whether HDAC2-truncating mutations conferred a particular expression signature on these cancer cells. Using unsupervised clustering analysis in microsatellite-unstable colorectal cancer cell lines, we have found that HDAC2 mutant cells (RKO and Co115) show a characteristically different expression microarray signature from HDAC2 wild-type cells (HCT-116, SW48, HCT-15 and LoVo). HDAC2 mutant cells exhibit upregulation of tumor-promoting genes, such as those of tyrosine kinases, mediators of cell cycle progression and angiogenic factors. The overexpression of these genes is associated with a loss of HDAC2 recruitment and a gain of histone H4 hyperacetylation in their particular 5'-end promoters, as observed by chromatin immunoprecipitation. Transfection of wild-type HDAC2 in mutant cells reverted this epigenetic pattern by repressing the transforming genes in association with HDAC2 promoter occupancy. These results suggest a role for HDAC2 mutations in human tumorigenesis through the derepression of key genes from multiple cellular transformation pathways.

The dioxin receptor is silenced by promoter hypermethylation in human acute lymphoblastic leukemia through inhibition of Sp1 binding

The transcription factor aryl hydrocarbon receptor (AhR) has relevant functions in cell proliferation. Interestingly, the AhR can either promote or inhibit proliferation depending on the cell phenotype. Although recent data reveal potential pathways for AhR signaling in cell proliferation, the mechanisms that regulate its activity in tumor cells remain unknown. Here, we have analyzed promoter hypermethylation as a potential mechanism controlling AhR expression in human tumor cells. AhR promoter CpG methylation was sporadic in a panel of 19 tumor cell lines except for the chronic myeloid leukemia (CML) K562 and the acute lymphoblastic leukemia (ALL) REH. When compared with normal lymphocytes, REH had very low constitutive AhR expression that could be attributed to promoter hypermethylation since treatment with the DNA demethylating agent 5-aza-2'-deoxycitidine (AZA) significantly increased AhR mRNA and protein. These results in leukemia-derived cell lines were further confirmed in primary ALL, where 33% of the patients (7/21) had AhR promoter hypermethylation. Chromatin immunoprecipitation (ChIP) showed that methylation impaired binding of the transcription factor Sp1 to the AhR promoter, thus providing a mechanism for AhR downregulation in REH cells. Therefore, promoter hypermethylation represents a novel epigenetic mechanism downregulating AhR activity in hematological malignancies such as ALL.

Modulation of guanosine triphosphatase activity of G proteins by arachidonic acid in rat Leydig cell membranes

Previous results from our group have indicated that arachidonic acid decrease cAMP production through a modification of heterotrimeric G proteins. In the present study, we have characterized the high affinity GTPase activity present in Leydig cell membranes and its regulation by fatty acids. The high-affinity GTPase activity, measured as [gamma32P] GTP hydrolysis rate, was both time and protein concentration dependent. Arachidonic acid elicited a dose-dependent inhibition of enzyme activity with an IC50 = 26.7+/-1.1 microM. The existence of only two double bonds in linoleic acid is reflected by a decrease in its inhibitory activity (IC50 = 34+/-2.3 microM). Saturated fatty acids showed no effect at this level. The kinetic analysis as interpreted by Lineweaver-Burk plots, indicated that 50 microM arachidonic acid had no effect on the apparent affinity for GTP, but resulted in a 40% decreases in the maximal velocity of the reaction. Arachidonic acid modulation of GTPase activity was not attenuated by blocking eicosanoid metabolism with inhibitors of 5'-lipoxygenase, cyclooxygenase, or epoxygenase P-450. The addition of arachidonic acid to pertussis toxin-treated membranes had no effect on the enzyme activity, indicating that arachidonic acid does not modify the GTPase activity present in Galphas protein. However, ADP-ribosylation with cholera toxin followed by arachidonic acid treatment led to a further 40% inhibition when compared with cholera toxin treatment alone. These results allowed us to postulate that arachidonic acid inhibits the GTPase activity of Gi protein family. To further analyze the mechanism of arachidonic acid inhibition of GTPase activity, the effect of arachidonic acid on the [35S]GTPgammaS binding was studied. No effect of this fatty acid on GTP binding was found. Combining our previous results with those found here, we can conclude that arachidonic acid maintains Gi proteins in their active state, which in turn inhibit adenylate cyclase and results in decrease cAMP levels.

Mutagenesis in the switch IV of the helical domain of the human Gsalpha reduces its GDP/GTP exchange rate

The Galpha subunits of heterotrimeric G proteins are constituted by a conserved GTPase "Ras-like" domain (RasD) and by a unique alpha-helical domain (HD). Upon GTP binding, four regions, called switch I, II, III, and IV, have been identified as undergoing structural changes. Switch I, II, and III are located in RasD and switch IV in HD. All Galpha known functions, such as GTPase activity and receptor, effector, and Gbetagamma interaction sites have been found to be localized in RasD, but little is known about the role of HD and its switch IV region. Through the construction of chimeras between human and Xenopus Gsalpha we have previously identified a HD region, encompassing helices alphaA, alphaB, and alphaC, that was responsible for the observed functional differences in their capacity to activate adenylyl cyclase (Antonelli et al. [1994]: FEBS Lett 340:249-254). Since switch IV is located within this region and contains most of the nonconservative amino acid differences between both Gsalpha proteins, in the present work we constructed two human Gsalpha mutant proteins in which we have changed four and five switch IV residues for the ones present in the Xenopus protein. Mutants M15 (hGsalphaalphaS133N, M135P, P138K, P143S) and M17 (hGsalphaalphaS133N, M135P, V137Y, P138K, P143S) were expressed in Escherichia coli, purified, and characterized by their ability to bind GTPgammaS, dissociate GDP, hydrolyze GTP, and activate adenylyl cyclase. A decreased rate of GDP release, GTPgammaS binding, and GTP hydrolysis was observed for both mutants, M17 having considerably slower kinetics than M15 for all functions tested. Reconstituted adenylyl cyclase activity with both mutants showed normal activation in the presence of AlF(4)(-), but a decreased activation with GTPgammaS, which is consistent with the lower GDP dissociating rate they displayed. These data provide new evidence on the role that HD is playing in modulating the GDP/GTP exchange of the Gsalpha subunit.

Effect of mevalonate availability on the association of G-protein alpha-subunits with the plasma membrane in GH4C1 cells

We show that the levels and activity of the alpha-subunits of Gs and Gi proteins in plasma membrane of GH4C1 cells are regulated by the availability of mevalonate (MVA), and not by changes in cholesterol cell content. Changes in the levels of MVA, induced by modulation of 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase, determine the amount of both membrane-bound G alpha-subunits, which correlated with the activity of their effector adenylyl cyclase. Lipoprotein deficient serum (LPDS) decreases cholesterol content and increases both HMG-CoA reductase activity and G alpha-subunits in the membrane. Cholesterol and 25-hydroxycholesterol (25-HC) each repress HMG-CoA reductase and diminish G alpha-subunit levels. However, while cholesterol cell content is also decreased by 25-HC, exogenous cholesterol increases it. In addition, the decrease of both G alpha-subunits is reversed by the presence of MVA. This regulation appears to be mediated by nonsterol products generated from MVA. We assume that the first is the prenylation of the gamma-subunits, since the attachment of G alpha-subunits to the membrane is dependent on this modification. However, as neither of our treatments completely abolished protein prenylation, we conclude that another MVA derivative is required in addition to prenyl residues to the presence and activity of alpha-subunits in the membrane.