5-Aza-2'-deoxycytidine-mediated reductions in G9A histone methyltransferase and histone H3 K9 di-methylation levels are linked to tumor suppressor gene reactivation.

The epigenetic silencing of tumor suppressor genes is a common event during carcinogenesis, and often involves aberrant DNA methylation and histone modification of gene regulatory regions, resulting in the formation of a transcriptionally repressive chromatin state. Two examples include the antimetastatic, tumor suppressor genes, desmocollin 3 (DSC3) and MASPIN, which are frequently silenced in this manner in human breast cancer. Treatment of the breast tumor cell lines MDA-MB-231 and UACC 1179 with 5-aza-2'-deoxycytidine (5-aza-CdR) induced transcriptional reactivation of both genes in a dose-dependent manner. Importantly, DSC3 and MASPIN reactivation was closely and consistently linked with significant decreases in promoter H3 K9 di-methylation. Moreover, 5-aza-CdR treatment also resulted in global decreases in H3 K9 di-methylation, an effect that was linked to its ability to mediate dose-dependent, post-transcriptional decreases in the key enzyme responsible for this epigenetic modification, G9A. Finally, small interfering RNA (siRNA)-mediated knockdown of G9A and DNMT1 led to increased MASPIN expression in MDA-MB-231 cells, to levels that were supra-additive, verifying the importance of these enzymes in maintaining multiple layers of epigenetic repression in breast tumor cells. These results highlight an additional, complimentary mechanism of action for 5-aza-CdR in the reactivation of epigenetically silenced genes, in a manner that is independent of its effects on DNA methylation, further supporting an important role for H3 K9 methylation in the aberrant repression of tumor suppressor genes in human cancer.

Quinol-glutathione conjugate-induced mutation spectra in the supF gene replicated in human AD293 cells and bacterial MBL50 cells.

Hydroquinone is a nephrocarcinogen in rats but generally tests negative in standard mutagenicity assays. However, 2,3,5-tris-(glutathion-S-yl)hydroquinone, a potent nephrotoxic metabolite of hydroquinone, and 2-bromo-bis-(glutathion-S-yl)hydroquinone, another cytotoxic quinol-glutathione (GSH) conjugate, cause extensive single strand breaks in DNA in a manner that is dependent on the formation of reactive oxygen species. We, therefore, investigated whether quinol-GSH conjugates have the potential to behave as genotoxicants. The shuttle vector pSP189, containing the supF gene, was treated with 2,3,5-tris-(glutathion-S-yl)hydroquinone and replicated in both human AD293 cells and Escherichia coli MBL50 cells. The mutation frequency increased 4.6- and 2.6-fold in human AD293 and bacterial MBL50 cells, respectively. Base substitutions were the major type of mutations, and they occurred predominantly at G:C sites in both cell types. A high frequency of deletions (30%), including < 10- and > 10-bp deletions, were observed in AD293-replicated plasmids. The most common types of mutations in AD293 cells were G:C to A:T transitions (33.8%) and G:C to T:A (29.4%) and G:C to C:G (19.1%) transversions. In MBL50 cells, the major mutations were G:C to T:A (33.8%) and G:C to C:G (31.3%) transversions and G:C to A:T transitions (27.5%). The mutation spectra were similar to those reported for *OH-induced mutations, suggesting that *OH generated from polyphenolic-GSH conjugates not only plays a role in cytotoxicity but also provides a basis for their mutagenicity and carcinogenicity.

Metabolism of arachidonic acid in human lung cancer cell lines.

The metabolism of arachidonic acid (AA) was studied in two pulmonary bronchioloalveolar-carcinoma cell lines (NCI-H322 and NCI-H358) and two small cell lung carcinoma cell lines (NCI-H69 and NCI-H128). Exogenous AA was metabolized only in the NCI-H322 and NCI-H358 cells. There was no detectable metabolism of AA in NCI-H69 or NCI-H128 cells, either in the presence or the absence of the calcium ionophore A23187. The major metabolite of AA isolated from both NCI-H322 and NCI-H358 cells was prostaglandin E2 (PGE2). Prostaglandin endoperoxide synthase activities, expressed as immunoreactive PGE2 (pmol/min/mg protein), were 10.3 +/- 0.28 (SD) and 4.8 +/- 0.48 in NCI-H358 and NCI-H322 cells, respectively. The rate of production of PGE2 by both NCI-H358 and NCI-H322 cells was linear up to 10 min. Production of PGE2 in both cell lines was dependent upon substrate concentration and was maximal above 17 microM AA. Moreover, PGE2 did not undergo further metabolism by either the NCI-H358 or the NCI-H322 cells. Aspirin (0.1 mM), a cyclooxygenase inhibitor, decreased PGE2 production by 77 and 60% in NCI-H358 and NCI-H322 cells, respectively. In the presence of exogenous AA the calcium ionophore, A23187 (20 microM), stimulated PGE2 production in NCI-H322 cells by almost 2-fold, although it did not affect PGE2 production in the NCI-H358 cells. In contrast, A23187 stimulated the endogenous production of PGE2 in both NCI-H322 and NCI-H358 cells by 4- and 9-fold respectively. In addition, both the NCI-H358 and NCI-H322 cell lines were susceptible to the cytotoxic effects of the anticancer agent mitoxantrone in both a time and concentration dependent manner. In contrast, the two cell lines lacking detectable prostaglandin synthesis activity, NCI-H69 and NCI-H128 were unaffected by treatment with mitoxantrone. These results illustrate that there are major differences in the abilities of human lung cancer cell lines to biosynthesize and release PGE2. It is conceivable that such differences might have exploitable diagnostic and/or therapeutic implications.

Glutathione conjugates of tert-butyl-hydroquinone, a metabolite of the urinary tract tumor promoter 3-tert-butyl-hydroxyanisole, are toxic to kidney and bladder.

3-tert-Butyl-4-hydroxyanisole and tert-butyl-hydroquinone (TBHQ) are antioxidants known to promote renal and bladder carcinogenesis in the rat, although the mechanisms of these effects are unclear. Because glutathione (GSH) conjugates of a variety of hydroquinones are nephrotoxic, and because 2-tert-butyl-5-(glutathion-S-yl)hydroquinone [5-(GSyl)TBHQ], 2-tert-butyl-6-(glutathion-S-yl)hydroquinone [6-(GSyl)TBHQ], and 2-tert-butyl-3,6-bis-(glutathion-S-yl)hydroquinone [3,6-bis-(GSyl)-TBHQ] have been identified recently as metabolites of TBHQ in the male rat, we investigated the effects of these metabolites in the male rat. At the highest dose tested (400 micromol/kg,i.v.) 5-(Gsyl)TBHQ and 6-(GSyl)TBHQ caused 2-fold increases in the urinary excretion of gamma-glutamyl transpeptidase and alkaline phosphatase, and pigments arising from the polymerization of metabolites were deposited in the kidney. 3,6-bis-(GSyl)TBHQ (200 micromol/kg) was the most potent of the GSH conjugates tested and produced significant increases in the urinary excretion of gamma-glutamyl transpeptidase, alkaline phosphatase, lactate dehydrogenase, and glucose (2-, 2-, 22-, and 11-fold increases, respectively). Alterations in the biochemical parameters correlated with the degree of single cell and tubular necrosis in the S(3)-M segment of the proximal tubule, as observed by light microscopy. In addition to nephrotoxicity, 3,6-bis-(GSyl)TBHQ increased the bladder wet weight 2-fold and caused severe hemorrhaging of the bladder. The half-wave oxidation potentials of 5-(Gsyl)TBHQ and 6-(GSyl)TBHQ were similar to that of TBHQ, whereas the half-wave oxidation potential of 3,6-bis-(Gsyl)TBHQ was approximately 100 mV higher than that of TBHQ. The TBHQ-GSH conjugates also catalyzed the formation of 8- hydroxydeoxyguanosine, indicating that GSH conjugation does not impair the redox activity of TBHQ. Because some chemicals may induce carcinogenesis by a mechanism involving cytotoxicity followed by sustained regenerative hyperplasia, our results suggest that the toxicity of GSH conjugates of TBHQ to kidney and bladder may contribute to the promoting effect of 3-tert-butyl-4-hydroxyanisole and TBHQ in these tissues.

Carrageenan-stimulated release of arachidonic acid and of lactate dehydrogenase from rat pleural cells.

Cells isolated from the rat pleural cavity consist mainly of macrophages, mast cells, eosinophils, and lymphocytes. Isolated pleural cells labeled with [14C]arachidonic acid released appreciable amounts (approximately 12%) of radiolabel upon exposure to pharmacological concentrations of carrageenan (1-100 micrograms/ml). The release of radiolabel was decreased by an inhibitor of phospholipase A2 (p-bromophenacyl bromide) but not by an inhibitor of arachidonate cyclooxygenase (indomethacin). The released products were arachidonic acid and, to a much lesser extent, prostaglandin E2 and leukotriene C4. The release of radiolabel was associated with release of cytosolic lactate dehydrogenase over the same range of carrageenan concentrations. Time-course studies indicated that release of radiolabel preceded that of lactate dehydrogenase. Since p-bromophenacyl bromide blocked stimulated release of radiolabel but did not prevent release of lactate dehydrogenase, it is unlikely that increase in arachidonate causes carrageenan-induced cell damage. Nevertheless, the question of whether the activation of phospholipase A2 in the pleural cells, most probably the macrophages, was sufficient to initiate the carrageenan-induced inflammatory response requires further study. Cytotoxicity which was apparent with as little as 5 micrograms/ml of carrageenan, may have been a significant consequence of carrageenan action.

Hepatic expression of hepatitis B virus genome in chronic hepatitis B virus infection.

The expression of hepatitis B virus (HBV) DNA in the liver was studied by nonisotopic in situ hybridization and correlated with liver histology, different phases in the natural evolution of chronic hepatitis B, and hepatic expression of HBV antigens in 251 Chinese patients with chronic HBV infection. A good correlation was found between the detection of HBV-DNA by in situ hybridization and serum HBV-DNA (P < .01). Chronic active hepatitis had the highest HBV-DNA detected in cytoplasm and nuclei, compared with livers showing minimal change, chronic persistent hepatitis, cirrhosis, and hepatocellular carcinoma. HBV-DNA in cytoplasm exceeded HBV-DNA in nucleus in all patients except in livers with hepatocellular carcinoma. Hepatic HBV-DNA correlated with disease activity (P < .02) and the correlation was highly significant with intralobular activity (P < .001). Patients in the early viral replicative phase of infection had higher levels of cytoplasmic and nuclear HBV-DNA compared with the late viral nonreplicative phase. Cytoplasmic and nuclear HBV-DNA correlated with hepatic expression of HBcAg and HBsAg (P < .05 in both cases), but not with HBeAg. These data indicate that hepatic expression of HBV-DNA follows the natural history of chronic HBV infection and is associated with active liver disease.

2,5-Bis-(glutathion-S-yl)-alpha-methyldopamine, a putative metabolite of (+/-)-3,4-methylenedioxyamphetamine, decreases brain serotonin concentrations.

3,4-(+/-)-Methylenedioxyamphetamine (MDA) and 3,4-(+/-)-methylenedioxymethamphetamine (MDMA) are serotonergic neurotoxicants. However, when injected directly into brain, MDA and MDMA are not neurotoxic, suggesting that systemic metabolism plays an important role in the development of neurotoxicity. The nature of the metabolite(s) responsible for MDA- and MDMA-mediated neurotoxicity is unclear. alpha-Methyldopamine is a major metabolite of MDA and is readily oxidized to the o-quinone, followed by conjugation with glutathione (GSH). Because the conjugation of quinones with GSH frequently results in preservation or enhancement of biological (re)activity, we have been investigating the role of quinone-thioethers in the acute and long-term neurochemical changes observed after administration of MDA. Although intracerebroventricular (i.c.v.) administration of 5-(glutathion-S-yl)-alpha-methyldopamine (4 x 720 nmol) and 5-(N-acetylcystein-S-yl)-alpha-methyldopamine (1 x 7 nmol) to Sprague-Dawley rats produced overt behavioral changes similar to those seen following administration of MDA (93 mumol/kg, s.c.) they did not produce long-term decreases in brain serotonin (5-hydroxytryptamine, 5-HT) concentrations. In contrast, 2,5-bis-(glutathion-S-yl)-alpha-methyldopamine (4 x 475 nmol) decreased 5-HT levels by 24%, 65% and 30% in the striatum, hippocampus and cortex, respectively, 7 days after injection. The relative sensitivity of the striatum, hippocampus and cortex to 2,5-bis-(glutathion-S-yl)-alpha-methyldopamine was the same as that observed for MDA; the absolute effects were greater with MDA. The effects of 2,5-bis-(glutathion-S-yl)-alpha-methyldopamine were also selective for serotonergic nerve terminal fields, in that 5-HT levels were unaffected in regions of the cell bodies. Because 2,5-bis-(glutathion-S-yl)-alpha-methyldopamine caused long-term depletion in 5-HT without adversely affecting the dopaminergic system, it also mimics the selectivity of MDA/MDMA. The data imply a possible role for quinone-thioethers in the neurobehavioral and neurotoxicological effects of MDA/MDMA.

Reactive intermediates and their toxicological significance.

Many chemicals that cause toxicity do so via metabolism to biologically reactive metabolites. However, the nature of the interaction between such reactive metabolites and various cellular components, and the mechanism(s) by which these interactions eventually lead to cell death are poorly understood. The relative importance of macromolecular alkylation (covalent binding), lipid peroxidation, alterations in thiol, calcium and energy homeostasis are discussed with reference to specific toxicants. It is concluded that the cytotoxic effects of reactive metabolites are a consequence of simultaneous and/or sequential alterations in several cellular processes. Further studies are required to determine the relationship between these alterations and cell death.

Species differences in renal gamma-glutamyl transpeptidase activity do not correlate with susceptibility to 2-bromo-(diglutathion-S-yl)-hydroquinone nephrotoxicity.

Administration of 2-bromo-(diglutathion-S-yl)hydroquinone (2-Br-[diGSyl]HQ) (10-30 mumol/kg; i.v.) to rats causes severe renal proximal tubular necrosis. gamma-Glutamyl transpeptidase (gamma-GT) catalyses the first step in the metabolism of glutathione (GSH) and its S-conjugates and the toxicity of 2-Br-(diGSyl)HQ can be emeliorated by inhibition of renal gamma-GT. Species differences in the specific activity of renal gamma-GT have been reported and we now describe the relationship between renal gamma-GT and species differences in susceptibility to 2-Br-(diGSyl)HQ nephrotoxicity. Although rats exhibited the highest specific activity of renal gamma-GT, and were the most sensitive species toward 2-Br-(diGSyl)HQ-mediated nephrotoxicity, renal gamma-GT activity did not correlate with susceptibility in the other species examined. Indeed, the guinea pig, which expressed the lowest activity of renal gamma-GT between the species (8% of the rat) was the only other rodent found to be responsive toward 2-Br-(diGSyl)HQ at the highest dose tested (200 mumol/kg; intracardiac). Thus, factors other than gamma-GT activity probably play an important role in modulating species susceptibility to 2-Br-(diGSyl)HQ nephrotoxicity. Although the reason(s) for the interspecies variation in response to 2-Br-(diGSyl)HQ are unclear at present, it seems possible that differences in both renal biochemistry, such as differences in the relative activities of cysteine conjugate N-acetyl transferase and deacetylase, and renal physiology, contribute to the observed results.

The contribution of bromobenzene to our current understanding of chemically-induced toxicities.

The metabolism and toxicity of bromobenzene has been investigated for well over one hundred years. The urinary excretion of mercapturic acids was first reported in 1879, in animals treated with bromobenzene. Bromobenzene has since proven to be a valuable tool in efforts to unravel the complexities involved in chemical- induced toxicities. For example, the importance of metabolic activation via the cytochrome(s) P-450; the role of glutathione in the detoxification of reactive metabolites; and the toxicological significance of covalent binding, enzyme inactivation and lipid peroxidation have all been illustrated in studies with bromobenzene. Thus, many of the principles involved in chemical-induced toxicity have been exemplified in studies with bromobenzene. These studies have provided substantial insight into the role of chemically reactive metabolites in the genesis of xenobiotic-mediated cytotoxicity.

Activation and detoxification of bromobenzene in extrahepatic tissues.

Bromobenzene causes hepatic and extrahepatic toxicity in rats. Toxicity is related to the presence of covalently bound material in these tissues. A major bromobenzene metabolite, p-bromophenol, has been shown to give rise to covalently bound material in liver, lung and kidney in vivo, but is not toxic. p-Bromophenol is formed from bromobenzene in liver, lung and kidney microsomes and is subsequently metabolized to 4-bromocatechol and covalently bound material. Bromobenzene-3,4-oxide generated in situ by liver microsomes, is detoxified by kidney, liver and lung cytosol. The results suggest that the kidney toxicity caused by bromobenzene is probably not mediated by either bromobenzene-3,4-oxide or the reactive metabolites of p-bromophenol. In contrast, bromobenzene-3, 4-oxide may play a role in the lung toxicity observed after bromobenzene administration. However, the covalently bound material found in extrahepatic tissues may be derived from both bromobenzene-3,4-oxide or the reactive metabolites of p-bromophenol, which may be formed directly by these tissues or transported there from the liver.

Redox stress and hepatic DNA fragmentation induced by diquat in vivo are not accompanied by increased 8-hydroxydeoxyguanosine contents.

Administration of 0.1 mmol/kg of diquat to Fischer-344 rats causes acute hepatic necrosis by mechanisms that appear to involve increased generation of reactive oxygen species, but the critical targets of the proposed oxidations have not been identified. In the present study the effects of diquat-induced redox stresses on hepatic protein thiol status were determined by derivatization of subcellular fractions with monobromobimane and separation of the fluorescent derivatives by SDS-PAGE. No differences in hepatic thiol status were seen in animals 2 or 6 h after diquat, relative to saline-treated controls, despite documentation of injury by elevated plasma transaminase activities. Hepatic DNA fragmentation was increased in diquat-treated animals (24.9±5.1 vs 6.7±0.3% (controls) at 2 h; 57.2±4.1 vs 4.6±0.3% (controls) at 6 h, P<0.001). However, 8-hydroxydeoxyguanosine (8-OHdG) contents in hepatic DNA were not increased by diquat (35.3±6.2 µmol 8-OHdG/mol deoxyguanosine (dG)) over saline-treated controls (28.3±2.6). Plasma NH3 concentrations increased in diquat-treated rats from 49 µM in controls to 170 µM 6 h after treatment with diquat. Hepatic activities of glutamine synthetase (GS) were lower in diquat-treated rats (39.7±13.0 mU/mg protein) than in controls (65.8±13.4, P<0.001), but activities of carbamyl phosphate synthetase-I (CPS-I), were not decreased significantly. The oxidation of proteins to forms reactive with 2,4-dinitrophenylhydrazine (DNPH) was investigated in subcellular fractions by Western blot analyses with a monoclonal antibody to DNP-derivatized bovine serum albumin (BSA). N-terminal sequencing of bands exhibiting reactivity with anti-DNP-BSA antibodies indicated protein carbonyl formation in malate dehydrogenase, protein disulfide isomerase, and glutathione transferase. The functional consequences of oxidation of these proteins are not known but the observation of protein carbonyl formation and no measurable loss of protein thiol content are consistent with iron chelate-mediated oxidation in the transformation critical to expression of tissue damage. The time course data are consistent with DNA fragmentation as a mechanism contributing to the development of cell injury, but the absence of increases in 8-OHdG indicates that direct oxidation of DNA may not be responsible.

Nutrient and grazing factors in relation to phytoplankton level in a eutrophic shallow lake: the effect of low macrophyte abundance.

The ability of submerged macrophytes to moderate the structure of food webs with respect to lake eutrophication management has been intensively studied in recent years. Many lake managers have adopted the option of increasing macrophyte abundance in order to restore eutrophic waters, with a view to improve water quality, increase water transparency and reduce phytoplankton biomass. These studies are mostly based upon the comparison of macrophyte-rich and macrophyte-free waters. There is less quantitative information on how different levels of macrophyte abundance, in particular very low levels, influence ecosystem structure, or vice versa, within the same ecosystem. This paper takes a multivariate statistical approach to examine ecosystem function with low macrophyte abundance (ca. 5%) in the context of nutrient-phytoplankton-zooplankton interaction in a eutrophic shallow lake. It shows that with low levels of macrophyte coverage, phytoplankton biomass was only limited by nutrient (TP and Si) availability, whilst nutrient (Si) as well as grazing (zooplankton and Daphnia) factors were both limiting phytoplankton levels in macrophyte-free waters. With respect to interactions between Daphnia and chlorophyll-a, the results suggest that low macrophyte cover does not result in a reduction of phytoplankton biomass due to the presence of Daphnia. Rather, low macrophyte cover provides a refuge that maintains constant Daphnia predation, so reducing chlorophyll-a fluctuation. These results add credence to the causal interpretation of different interactions amongst nutrients, phytoplankton and zooplankton at very low macrophyte densities.

Biological and chemical factors influencing shallow lake eutrophication: a long-term study.

The focus of eutrophication research has tended to be upon short-term and experimental studies. However, given the range of factors that can influence eutrophication dynamics, and that these matter over a range of time scales, some discrete, some continuous, eutrophication dynamics may only be fully investigated when long-term, time-series data are available. The present study aims to evaluate the interacting effects of abiotic processes and biotic dynamics in explaining variations of phytoplankton biomass in a eutrophic shallow lake, Barton Broad (Norfolk, UK) using a long-term data set. Multivariate statistical analysis shows that the inter-relationships between phytoplankton variability, nutrient and grazing factors were highly sensitive to seasonal periodicity. In spring phytoplankton biomass was related to phosphorus, nitrogen and silicon. In summer phytoplankton biomass was associated with phosphorus, nitrogen and zooplankton. In autumn phytoplankton was related to phosphorus, nitrogen, silicon and zooplankton. In winter, no significant relationship could be established between phytoplankton and environmental variables. This paper improves our understanding of the governing role of nitrogen, phosphorus, silicon and zooplankton upon phytoplankton variability, and hence, improves management methods for eutrophic lakes.

Glutathione conjugation as a mechanism for the transport of reactive metabolites.

From this and other chapters in this volume, it should be clear that GSH conjugation no longer represents a mechanism for the detoxication of xenobiotics or their metabolites. Although the majority of conjugations with GSH do facilitate the efficient excretion of xenobiotics from the body, many examples now exist where this process results in enhanced biological reactivity (Monks et al., 1990a; Monks and Lau, 1992, 1994). The number of examples in which GSH conjugation plays an important role in the generation of biologically reactive intermediates is expanding rapidly and GSH-dependent toxicity is manifested in many diverse ways. As emphasized in this chapter, GSH can act as a transport form for reactive metabolites, permitting the delivery of such metabolites to target tissues distal to the site of the initial conjugation. This type of GSH conjugate may be important in the mutagenic, carcinogenic, nephrotoxic, embryotoxic, cataractogenic, methemoglobinemic, and neurotoxic properties of a variety of redox active compounds (Monks and Lau, 1992).