Quantitative proteomic analysis of mouse liver response to the peroxisome proliferator diethylhexylphthalate (DEHP).

Peroxisome proliferators (PPs) are a diverse group of chemicals that cause hepatic proliferation, suppression of apoptosis, peroxisome proliferation and liver tumours in rodents. The biochemical response to PPs involves changes in the expression of peroxisomal beta-oxidation enzymes and fatty acid transport proteins such as acyl-CoA oxidase and liver fatty acid binding protein. The response to PPs is mediated by the peroxisome proliferator-activated receptor alpha (PPARalpha) and the livers of PPARalpha-null transgenic mice do not develop tumours in response to PPs. In order to identify the molecular pathways underlying the adverse effects of PPs in rodent liver, we carried out two-dimensional differential gel electrophoresis to provide quantitative proteomic analyses of diethylhexylphthalate (DEHP)-treated wild-type or PPARalpha-null mouse livers. Since tumourigenesis is both PP- and PPARalpha-dependent, analyses were focused on these changes. Fifty-nine proteins were identified where altered expression was both PPARalpha- and PP-dependent. In addition, six proteins regulated by the deletion of PPARalpha were identified, possibly indicating an adaptive change in response to the loss of this receptor. The proteins that we identified as being regulated by PPARalpha are known to be involved in lipid metabolism pathways, but also in amino acid and carbohydrate metabolism, mitochondrial bioenergetics and in stress responses including several genes not previously reported to be regulated by PPARalpha. These data provide novel insights into the pathways utilised by PPs and may assist in the identification of early markers rodent nongenotoxic hepatocarcinogenesis.

Validation and development of fluorescence two-dimensional differential gel electrophoresis proteomics technology.

Fluorescence two-dimensional differential gel electrophoresis (2-D DIGE*) is a new development in protein detection for two-dimensional gels. Using mouse liver homogenates (control and paracetamol (N-acetyl-p-aminophenol, APAP)-treated), we have determined the quantitative variation in the 2-D DIGE process and established statistically valid thresholds for assigning quantitative changes between samples. Thresholds were dependent on normalised spot volume, ranged from approximately 1.2 fold for large volume spots to 3.5 fold for small volume spots and were not markedly affected by the particular cyanine dye combination or by multiple operators carrying out the dye labelling reaction. To minimise the thresholds, substantial user editing was required when using ImageMaster 2D-Elite software. The difference thresholds were applied to the test system and quantitative protein differences were determined using replicate gels of pool samples and single gels from multiple individual animals (control vs treated in each gel). Throughout, the differences revealed with a particular cyanine dye combination were mirrored almost without exception when the dye combination was reversed. Both pool and individual sample analyses provided unique data to the study. The inter-animal response variability in inbred mice was approximately nine times that contributed by the 2-D DIGE process. A number of the most frequently observed protein changes resulting from APAP-treatment were identified by mass spectrometry. Several of these can be rationalised based on available data on the mechanism of APAP hepatotoxicity but others cannot, indicating that proteomics can provide further insights into the biochemical basis of APAP toxicity.

Enhanced acetaminophen hepatotoxicity in transgenic mice overexpressing BCL-2.

Mitochondria play an important role in the cell death induced by many drugs, including hepatotoxicity from overdose of the popular analgesic, acetaminophen (APAP). To investigate mitochondrial alterations associated with APAP-induced hepatotoxicity, the subcellular distribution of proapoptotic BAX was determined. Based on the antiapoptotic characteristics of BCL-2, we further hypothesized that if a BAX component was evident then BCL-2 overexpression may be hepatoprotective. Mice, either with a human bcl-2 transgene (-/+) or wild-type mice (WT; -/-), were dosed with 500 or 600 mg/kg (i.p.) APAP or a nonhepatotoxic isomer, N-acetyl-m-aminophenol (AMAP). Immunoblot analyses indicated increased mitochondrial BAX-beta content very early after APAP or AMAP treatment. This was paralleled by disappearance of BAX-alpha from the cytosol of APAP treated animals and, to a lesser extent, with AMAP treatment. Early pathological evidence of APAP-induced zone 3 necrosis was seen in bcl-2 (-/+) mice, which progressed to massive panlobular necrosis with hemorrhage by 24 h. In contrast, WT mice dosed with APAP showed a more typical, and less severe, centrilobular necrosis. AMAP-treated bcl-2 (-/+) mice displayed only early microvesicular steatosis without progression to extensive necrosis. Decreased complex III activity, evident as early as 6 h after treatment, correlated well with plasma enzyme activities at 24 h (AST r(2) = 0.89, ALT r(2) = 0.87) thereby confirming a role for mitochondria in APAP-mediated hepatotoxicity. In conclusion, these data suggest for the first time that BAX may be an early determinant of APAP-mediated hepatotoxicity and that BCL-2 overexpression unexpectedly enhances APAP hepatotoxicity.

PPARalpha-dependent alteration of GRP94 expression in mouse hepatocytes.

The adverse effects of the peroxisome proliferators (PPs), a class of rodent nongenotoxic hepatocarcinogens, include suppression of apoptosis, induction of hepatocyte proliferation, and liver enlargement which eventually leads to tumours. The response to PPs is mediated by the peroxisome proliferator activated receptor alpha (PPARalpha). We carried out proteomic analyses of PP-treated hepatocytes from wild-type and PPARalpha-null mice to identify the molecular pathways underlying the adverse effects of PPs. We have identified eighteen protein spots exhibiting differential expression in PP-treated wild-type mouse hepatocytes. Several proteins involved in lipid metabolism pathways, but also ATP synthase beta subunit, which are regulated by PPs were identified. In addition, both 2D silver-stained gels and Western blotting analysis indicated that the anti-apoptotic glucose-regulated protein 94 (GRP94) is consistently overexpressed upon stimulation with PPs, providing us with novel insights into the anti-apoptotic mechanism activated by PPs.

Proteomic analysis of differential protein expression in primary hepatocytes induced by EGF, tumour necrosis factor alpha or the peroxisome proliferator nafenopin.

Peroxisome proliferators are nongenotoxic rodent-liver carcinogens that have been shown to cause both an induction of hepatocyte proliferation and a suppression of apoptosis. Both epidermal growth factor (EGF) and the peroxisome proliferator nafenopin induce DNA replication in primary rat hepatocyte cultures, but apparently through different signalling pathways. However, both EGF and nafenopin require tumour necrosis factor alpha (TNFalpha) signalling to induce DNA replication. By examining proteins isolated from rat primary hepatocyte cultures using two-dimensional gel electrophoresis and mass spectrometry, we found that proteins showing an altered expression pattern in response to nafenopin differed from those showing altered expression in response to EGF. However, many proteins showing altered expression upon stimulation with TNFalpha were common to both the EGF and nafenopin responses. These proteome profiling experiments contribute to a better understanding of the molecular mechanisms involved in the response to peroxisome proliferators. We found 32 proteins with altered expression upon stimulation with nafenopin, including muscarinic acetylcholine receptor 3, intermediate filament vimentin and the beta subunit of the ATP synthase. These nonperoxisomal protein targets offer insights into the mechanisms of peroxisome proliferator-induced carcinogenesis in rodents and provide opportunities to identify toxicological markers to facilitate early identification of nongenotoxic carcinogens.

Protein and nonprotein cysteinyl thiol modification by N-acetyl-p-benzoquinone imine via a novel ipso adduct.

N-acetyl-p-benzoquinone imine (NAPQI), a reactive metabolite of acetaminophen (APAP), can arylate and oxidize protein and nonprotein thiols in the pathogenesis of APAP-induced hepatotoxicity. We report the first direct evidence for the formation of a labile ipso adduct between glutathione (GSH) and NAPQI using a combination of techniques including liquid chromatography/tandem mass spectrometry and liquid chromatography/NMR spectroscopy. Decomposition kinetics of the GSH-NAPQI ipso adduct and product ratios suggested that the ipso adduct was readily reversible back to NAPQI under neutral and basic conditions. The significance of the ipso adduct is that it may migrate from its site of formation to other cell compartments where it can either oxidize protein thiols or covalently modify them. Ipso adduct formation with protein thiols was demonstrated with a cysteine protease, papain, whose catalytic activity relies on the presence of an active site cysteinyl thiol. The formation and reactions of cysteinyl thiol ipso adducts of NAPQI provides significant new insights into possible reactions of quinone imines with cellular peptides and proteins.

Role of CYP1A2 in the hepatotoxicity of acetaminophen: investigations using Cyp1a2 null mice.

Acetaminophen (APAP) is known to cause centrilobular hepatic necrosis under overdose conditions. This is thought to be mediated via the P450-generated reactive intermediate N-acetyl-p-benzoquinone imine (NAPQI). Initially, NAPQI is detoxified by conjugation with glutathione (GSH), but once GSH is depleted, NAPQI reacts more extensively with hepatic proteins leading to hepatocellular damage. The P450 isoforms thought to be responsible for APAP hepatotoxicity in humans are CYP2E1, CYP1A2, and CYP3A4, and thus, we have investigated the effect of murine Cyp1a2 on APAP hepatotoxicity using Cyp1a2 knockout mice (Liang et al., Proc. Natl. Acad. Sci. USA 93, 1671-1676, 1996). Doses of 250 mg/kg were markedly hepatotoxic in these mice, and surprisingly, deaths only occurred in the knock-out and heterozygote mice over a 24-h period after dosing. Furthermore, there were no significant differences among survivors of any genotype in serum ALT concentrations, a well correlated indicator of APAP hepatotoxicity in mice. Finally, no differences were observed in the urinary metabolites excreted ove the 24-h period, including those derived from GSH conjugation of the major reactive metabolite NAPQI. Consistent with the effects on hepatotoxicity and metabolism, 2 h after hepatotoxic doses (500 mg/kg, i.p.) of APAP no significant differences were observed in total whole liver homogenate nonprotein thiol concentrations among the three genotypes even though hepatic thiols were decreased compared to control animals (> 90%). In addition, when the liver cytosol and microsome samples were examined by immunoblotting for the presence of APAP-protein adducts using a specific antiserum, there were no observable differences in either the intensity of staining or in the spectrum of adducts formed between APAP-dosed mice of any genotype. The cumulative data suggest that Cyp1a2 doses not play a significant role in APAP hepatotoxicity in these mice.