Glutathione S-transferases (GSTs) metabolize drugs and xenobiotics. Yet despite high protein sequence homology, expression of π-class GSTs, the most abundant of the enzymes, varies significantly between species. In mouse liver, hepatocytes exhibit high mGstp expression, while in human liver, hepatocytes contain little or no hGSTP1 mRNA or hGSTP1 protein. π-class GSTs are known to be critical determinants of liver responses to drugs and toxins: when treated with high doses of acetaminophen, mGstp1/2+/+ mice suffer marked liver damage, while mGstp1/2−/− mice escape liver injury.
To more faithfully model the contribution of π-class GSTs to human liver toxicology, we introduced hGSTP1, with its exons, introns, and flanking sequences, into the germline of mice carrying disrupted mGstp genes. In the resultant hGSTP1+mGstp1/2−/− strain, π-class GSTs were regulated differently than in wild-type mice. In the liver, enzyme expression was restricted to bile duct cells, Kupffer cells, macrophages, and endothelial cells, reminiscent of human liver, while in the prostate, enzyme production was limited to basal epithelial cells, reminiscent of human prostate. The human patterns of hGSTP1 transgene regulation were accompanied by human patterns of DNA methylation, with bisulfite genomic sequencing revealing establishment of an unmethylated CpG island sequence encompassing the gene promoter. Unlike wild-type or mGstp1/2−/− mice, when hGSTP1+mGstp1/2−/− mice were overdosed with acetaminophen, liver tissues showed limited centrilobular necrosis, suggesting that π-class GSTs may be critical determinants of toxin-induced hepatocyte injury even when not expressed by hepatocytes.
Citation: Vaughn MP, Biswal Shinohara D, Castagna N, Hicks JL, Netto G, De Marzo AM, et al. (2011) Humanizing π-Class Glutathione S-Transferase Regulation in a Mouse Model Alters Liver Toxicity in Response to Acetaminophen Overdose. PLoS ONE6(10): e25707. https://doi.org/10.1371/journal.pone.0025707
Editor: Anna Carla Goldberg, Albert Einstein Institute for Research and Education, Brazil
Received: July 22, 2011; Accepted: September 8, 2011; Published: October 11, 2011
Copyright: © 2011 Vaughn et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: The studies were supported by National Institutes of Health/National Cancer Institute grants CA91409, CA70196, and CA58236. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Cytosolic glutathione S-transferases (GSTs) are encoded by a superfamily of genes grouped intoα, μ, π, σ, θ, ζ, and ω classes by primary amino acid sequence , . The enzymes, which can exist as homo- or heterodimers of subunit polypeptides, catalyze reactions involving the conjugation of reduced glutathione (GSH) to electrophilic substrates. By targeting electrophiles generated by cytochrome P450s for conjugation with GSH, GSTs contribute to coordinated drug and xenobiotic metabolism, promoting the ultimate elimination of potential toxins by the ATP-dependent glutathione S-conjugate export pump . The π-class GST, a homodimeric enzyme which was first described as a placental isoform, has since been found in many different tissues and is now known to be the most abundant of the GSTs . The amino acid sequence of π-class GST subunit polypeptides is remarkably conserved across species, perhaps because the enzymes have been implicated in a broad array of vital cell and tissue functions, including xenobiotic metabolism, cell signaling, mutagen/carcinogen defense, and antineoplastic drug resistance .
Human π-class GSTs are encoded by a single gene, hGSTP1, while mice have two such genes, mGstp1 and mGstp2. Each of the genes has 7 exons that encode enzyme subunit polypeptides of 210 amino acids . The degree of protein homology is striking: of the 210 amino acids, 170 (81%) are completely conserved, and accounting for conserved and semi-conserved substitutions, the amino acid sequence across species has a 94% homology. hGSTP1, mGstp1, and mGstp2 also share some common transcriptional promoter elements, including an AP1 (TGA[C/G]TCA) transcription factor binding site and GC boxes (GGGCGG) permitting SP1 transcription factor binding just upstream of a TATA box. hGSTP1 and mGstp2 have two SP1 binding sites; mGstp1 has only one. For hGSTP1, the AP1 site and at least one of the SP1 sites must be intact for hGSTP1 transcription to occur , .
Despite the extensive coding sequence homology, the transcription promoter regions and introns display little similarity across species, aside from the AP1 and SP1 sites. hGSTP1 contains several unique promoter elements that could contribute to transcriptional regulation, including a pentad B sequence repeated 18 to 21 times and an overlapping NF-κB and C/EBP site (TTAAGGGAATTTCC). Also noteworthy is the abundance of CpG dinucleotides (n = 39) found between an [ATAAA]n repeat region and the hGSTP1 transcription start site. These CpGs, which are unmethylated in normal cells, are targeted for de novo methylation in many human cancers, leading to somatic epigenetic silencing of transcription , .
As a consequence of the likely differences in transcriptional regulation, π-class GST expression patterns vary widely across species, and these expression differences appear to have functional consequences. As an example, shortly after the discovery of rGST-P, the rat π-class GST, levels of the enzyme were found to be stereotypically elevated in preneoplastic hepatic foci, and in neoplastic lesions, induced by the chemical carcinogens diethylnitrosamine (DEN) and 2-acetylaminofluorene (AAF) . Because rGST-P is not expressed in rat hepatocytes under normal conditions, the appearance of rGST-P-expressing cells has emerged as a reliable biomarker of hepatocarcinogenesis in the rat , . However, this rat model has been difficult to extend to mice, which exhibit constitutive production of mGstp1/2 in normal hepatocytes, or to humans, which like rats fail to express π-class GSTs in hepatocytes, but do not induce π-class GST expression in preneoplastic or neoplastic lesions , , , , , , . Of interest, many strains of mice appear much less susceptible than rats to liver carcinogenesis induced by chemical carcinogens , .
In an attempt to recapitulate human patterns of π-class GST expression in mice, we introduced the full-length hGSTP1 gene, along with its cis-transcriptional regulatory sequences, into the germline of mice carrying disrupted mGstp genes. The resultant “humanized” strain of hGSTP1+mGstp1/2−/− mice showed marked differences from wild-type mGstp1/2+/+ mice in the expression of π-class GSTs in liver tissues, with an absence of the enzymes in hepatocytes that was reminiscent of the lack of enzymes in human hepatocytes, but maintenance of expression in bile duct cells, Kupffer cells, macrophages, and endothelial cells. These expression differences were responsible for functional differences in liver toxicology, with hGSTP1+mGstp1/2−/− mice exhibiting far less liver injury than mGstp1/2+/+ mice upon administration of high doses of acetaminophen . Nonetheless, hGSTP1+mGstp1/2−/− mice did manifest limited centrilobular necrosis not seen in mGstp1/2−/− mice, suggesting that π-class GSTs remained critical determinants of liver damage even when not present in hepatocytes.
π-class GST expression in hGSTP1+ mGstp1/2−/− mice
To create humanized hGSTP1+ mGstp1/2−/− mice, a linearized DNA fragment containing full length hGSTP1 gene (−1138 to +3600) was microinjected into C57BL/6 mouse oocytes, yielding some 16 offspring. Before attempting to target this gene to the germline of mice, this construct or a 5′ deletion construct with truncated 5′ regulatory regions lacking the [ATAAA]n pentad repeat (Fig. 1A), was transfected into Hep3B human liver cancer cells, which are normally devoid of π-class GSTs as a result of epigenetic silencing of hGSTP1 attributable to somatic CpG island hypermethylation . hGSTP1 transfection promoted expression of π-class GST subunit polypeptides and an increase in GST activity (Fig. 1B–C). Three hGSTP1+ founder mice were then crossed to C57BL/6 mGstp1/2−/− mice; after two generations of breeding the resultant hGSTP1+mGstp1/2−/− mouse strain was maintained by breeding with mGstp1/2−/− mice and selecting progeny carrying hGSTP1+ alleles. hGSTP1+mGstp1/2−/− mice were not grossly different in appearance or behavior from mGstp1/2+/+ or mGstp1/2−/− mice.
When transfected into Hep3B cells, which do not express hGSTP1 mRNA or hGSTP1 polypeptides, (A) both full length hGSTP1 (−1138 to +3600) and hGSTP1 with [ATAAA]n ([AT]n) repeat sequences deleted (−78 to +3600) directed expression of (B) GSTP1 polypeptides, as detected by immunoblot analysis, and (C) active GSTπ, as detected by CDNB assay using the indicated volumes of HEP3B cell lysates, with appearance of the glutathione conjugate of CDNB monitored as change in OD340 nm. (A–C) Yellow box, position of [AT]n repeat; Green box, position of CpG island; positions are shown with respect to the hGSTP1 transcriptional start site; blue boxes, position of exons; rhGSTP1, recombinant human GSTP1 polypeptides in vitro. Transfection controls directed expression of GFP polypeptides.
Patterns of π-class GST expression differ substantially between humans and mice
To ascertain whether human π-class GST expression patterns were recapitulated in mice by transfer of hGSTP1, with all of its known cis-regulatory sequences, various tissues from mice and from humans were subjected to immunohistochemical staining analyses of enzyme content using anti-π-class GST antibodies (Fig. 2A). For liver tissues, in mGstp1/2+/+mice, π-class GSTs were present in the nucleus and cytoplasm of hepatocytes, but not prominently in Kupffer cells, bile duct cells, and endothelial cells, while in hGSTP1+ mGstp1/2−/− mice, enzyme expression was restricted to non-hepatocytes, including endothelial cells, Kupffer cells, and bile duct cells. In this way, the pattern of π-class GST expression in human liver, where the hepatocytes are devoid of the enzymes but show expression in bile duct, endothelial, and Kupffer cells, was better recapitulated in hGSTP1+mGstp1/2−/− mice than in mGstp1/2+/+mice. Human prostate tissues produce π-class GSTs predominantly in basal epithelial cells, and much less so in luminal epithelial cells. As in the liver, this human pattern of expression was better recapitulated in hGSTP1+mGstp1/2−/− mice than in mGstp1/2+/+mice. Furthermore, the antibodies used could detect both mouse and human π-class GST with great specificity, as no immunoreactive peptides were evident in tissues from mGstp1/2−/− mice.
(A) Immunohistochemistry using anti-GSTπ antibodies for liver and prostate tissues from a wild type mouse, mGstp1/2−/− mouse, hGSTP1+mGstp1/2−/− mouse, and human. Left panels show liver staining; arrows, bile ducts; V, central vein. Right panels show prostate staining, arrows, representative luminal cells; arrowheads, representative basal cells; Lu, lumen of prostate gland acini. All images are 200× original magnification. (B) Quantitative RT-PCR for mGstp1 and hGSTP1, normalized to mGapdh or hGAPDH, using RNA from various tissues.
Next, mRNAs from various tissues from mice and from humans were subjected to quantitative RT-PCR for mGstp1, hGSTP1, mGapdh, and hGAPDH (Fig. 2B). In tissues from mature male C57BL/6 wild type mice, mGstp1 mRNA was present at high levels in the liver, relative to hGSTP1 mRNA, and expressed at lower levels in the kidney and in each of the lobes of the prostate. In contrast, in human tissues, hGSTP1 mRNA, relative to hGAPDH, was barely detectable in the liver, but expressed at low levels in the kidney and prostate. For hGSTP1+mGstp1/2−/− mice, hGSTP1 mRNA expression patterns, relative to mGSTP1 mRNA, resembled that of hGSTP1 mRNA, relative to hGAPDH, in human tissues: hGSTP1 mRNA was nearly absent from the liver while present in the kidney and in each of the prostate lobes. Of note, anatomically distinct mouse prostate lobes may be analogous to non-anatomically distinct human prostate zones, so each of the mouse lobes were considered independently, while the human prostate was assessed as a single entity.
GSTP1 CpG island DNA methylation patterns in hGSTP1+ mice
In cancers arising in human liver and human prostate, absence of π-class GST expression is a consequence of somatic CpG methylation changes at a CpG island sequence encompassing the hGSTP1 transcriptional regulatory region , . Remarkably, though the gene microinjected into mouse oocytes was without any 5-meC bases, when maintained in the germline, the hGSTP1 alleles present in hGSTP1+ mGstp1/2−/− mice established patterns of CpG dinucleotide methylation in adult cells that was similar to those seen in hGSTP1 in normal adult human cells (Fig. 3) . This suggested that cis-elements in hGSTP1 were not only sufficient to establish human patterns of hGSTP1 expression in hGSTP1+ mGstp1/2−/− mice, but also sufficient to establish human patterns of CpG dinucleotide/CpG island methylation. Finally, the lack of expression of π-class GSTs in hepatocytes and prostate luminal cells in hGSTP1+ mGstp1/2−/− mice could not be explained by hypermethylation of hGSTP1 CpG island sequences.
Bisulfite genomic sequencing for 5-meC in four amplicons (labeled BiSeq 1–4)from the hGSTP1 CpG island using DNA from various hGSTP1+mGstp1/2−/− mouse tissues (hGSTP1+ liver, prostate, kidney), along with DNA from human white blood cells (normal hGSTP1 5-meC pattern) and Hep3B cells (hypermethylated at hGSTP1 CpG island). Each row represents a summary of multiple independently cloned alleles from each sample as indicated. Circles indicate the relative position of CpG dinucleotides within the bisulfite sequencing amplicon. The color of each circle is scaled to the fraction of alleles methylated for the indicated CpG according to the heatmap shown at the bottom.
Toxicity of acetaminophen (APAP)
Data obtained from in vitro studies using purified enzymes suggested that π-class GSTs might be major contributors to the liver disposition of acetaminophen (N-acetyl-p-aminophenol; APAP) by catalyzing reactions between its reactive metabolite N-acetyl-p-benzoquinoneimine (NAPQI) and glutathione . However, when mGstp1/2−/− mice were challenged with an overdose of acetaminophen (APAP), the mice exhibited a marked decreased, not increased, hepatotoxicity when compared to mGstp1/2+/+ mice . To ascertain whether in hGSTP1+mGstp1/2−/− mice, which like mGstp1/2−/− mice, fail to produce π-class GSTs in hepatocytes, were sensitive or resistant to liver damage caused by APAP, we treated mGstp1/2+/+ mice, hGSTP1+mGstp1/2−/− mice, and mGstp1/2−/− mice with 300 mg/kg APAP and assessed the mice for liver damage using serum assays for alanine aminotransferase (ALT) and histopathology examination of liver tissues (Fig. 4, 5). Male mGstp1/2+/+ mice that had been fasted for 24 hours and then given APAP by intraperitoneal injection exhibited marked increases in serum ALT levels relative to saline-treated controls (p<0.04) and also showed massive centrilobular and bridging necrosis in the liver with around 40–50% hepatocyte death. In contrast, mGstp1/2−/− mice showed very little evidence of liver damage in response to APAP, and no significant differences in serum ALT levels. The humanized hGSTP1+mGstp1/2−/− mice treated with APAP showed low levels of liver injury, with small serum ALT elevations, more limited centrilobular necrosis without bridging, and mild hepatocellular necrosis (less than 10%). ALT levels were significantly greater than those observed in the APAP-treated mGstp1/2−/− mice (p = 0.03), but lower than wildtype mice. As previously reported, significant gender differences were observed in response to the acetaminophen treatment in mGstp1/2+/+ mice. Female wildtype mice did not show significant elevations in ALT following acetaminophen treatment relative to saline-treated control. This gender bias in APAP mediated ALT elevations was also present in the acetaminophen-treated hGSTP1+mGstp1/2−/− mice, which also did not show significant elevations in ALT relative to saline-treated controls. In fact, regardless of genotype, female mice did not exhibit elevations in ALT following acetaminophen treatment.
Serum ALT levels 24 hours after intraperitoneal injection of 300 mg/kg APAP or saline are shown. Compared to mGstp1/2−/− mice, mGstp1/2+/+ (p<0.04) and hGSTP1+mGstp1/2−/− (p<0.03) mice exhibited significantly higher ALT elevations in response to APAP overdose. Data was analyzed by Wilcoxon Rank Sum test.
Arrowheads denote central veins; arrows highlight liver injury/necrosis. Left panels are 200× original magnification and Right panels are 400× original magnification.
In a second experiment, mice were administered 300 mg/kg acetaminophen or vehicle by oral gavage in order to more faithfully model human exposure to the drug. Serum markers of parenchymal liver damage (as measured by ALT and AST) as well as biliary duct damage (as measured by alkaline phosphatase and total bilirubin) were assessed at various timepoints (Fig. 6A). In wildtype mice levels of ALT and AST increased relative to baseline measurements 4 hours after drug administration and peaked at 24 hours. In contrast, mGstp1/2−/− mice were impervious to acetaminophen administration, and serum markers remained normal. Interestingly, hGSTP1+mGstp1/2−/− mice exhibited an intermediate level of sensitivity to acetaminophen relative to the wildtype and mGstp1/2−/− mice – elevations of AST and ALT at 24 hours in hGSTP1+mGstp1/2−/− mice were significantly greater than those in mGstp1/2−/− mice but showed a trend to be lower than those seen in wildtype mice (Figure 6A–B). Acetaminophen treatment did not result in increases in serum levels of total bilirubin or alkaline phosphate outside of the normal range, suggesting that acetaminophen-induced damage was primarily occurring in the hepatocytes.
Baseline serum measurements were taken prior to administration of the drug. Mice were fasted for 12 or 24 hours and then administered vehicle or 300 mg/kg acetaminophen via gavage. (A) Serum AST, ALT, ALP and total bilirubin measurements taken in serum collected at 4, 24, and 48 hours. (B) Serum measurements of AST and ALT at 24 hours after acetaminophen administration. N = 5 mice per group * = p<0.05 by Students t-test.
The generation of a “humanized” hGSTP1-transgenic mouse strain has created a new mouse model for the study of human metabolism, toxicology, and carcinogenesis. The different patterns of π-class GST expression in mGstp1/2+/+ mice and hGSTP1+mGstp1/2−/− mice, with accompanying differences in response to drugs like acetaminophen, may provide new insights not just into the biochemistry of drug and toxin disposition, but rather the cell and tissue specific responses to stress and injury inflicted by such agents. In this way, hGSTP1+mGstp1/2−/− mice can join an increasing family of “humanized” transgenic mice created to better understand the xenobiotic metabolism .
To make such “humanized” transgenic mice, three general approaches have been pursued. One technique involves the construction of transgenes containing cDNA encoding human metabolizing enzymes ligated to transcription regulatory sequences to drive cell- or organ-specific enzyme expression. A second strategy features recombination (or “knock-in”) of human coding sequences into an orthologous mouse gene. For mice generated in this way, human enzymes can be produced in a mouse pattern of expression. To create mice in which human enzymes are present in a human pattern of expression, introduction of the entire human gene into the germline of mice, as was done for hGSTP1, may be required. This last approach relies upon the recognition of human transcriptional regulatory sequences by murine transcription factors, which can direct chromatin assembly and lineage-specific gene expression. For hGSTP1+mGstp1/2−/− mice, the observed patterns of hGSTP1 expression were most consistent with significant conservation in transcriptional trans-regulation between human and mouse. Finally, for any of the “humanized” transgenic mice producing human enzymes, there is a concern that simultaneous expression of both human and mouse enzymes might interfere generally with metabolic processes. The availability of gene knockout strains, such as mGstp1/2−/− mice, for breeding crosses with mice carrying human transgenes can reduce this worry.
One observation made during the expansion of the hGSTP1+mGstp1/2−/− mouse colony was a general increase in litter size of the hGSTP1+mGstp1/2−/− mice as compared to mGstp1/2−/− mice. The mechanistic basis for this difference has not been established. However, π-class GSTs were originally assigned the “P” (or “π”) family designation because the enzymes were first found in the placenta . Perhaps, π-class GSTs provide some sort of fetal development or survival advantage, whether mouse or human. This advantage may also be manifest in a greater than ∼60% transmission of the hGSTP1 transgene on the mGstp1/2−/− background (the expected frequency is 50%). Aside from the slight increase in litter size and gene transmission, there were no other obvious gross phenotypes in the hGSTP1+mGstp1/2−/− mouse strain.
When subjected to APAP overdoses, hGSTP1+mGstp1/2−/− mice responded differently than mGstp1/2+/+ mice. There were also notable differences between the serum ALT levels of male and female mice after administration of acetaminophen. This phenomenon is in agreement with previous reports ,  and is likely due to gender-dependent differences in expression of π-class GSTs; female mice have significantly lower expression of GST-π and are resistant to damage . Despite intensive study, the mechanism(s) by which APAP, and APAP metabolism, cause liver injury has not been fully elucidated. Although APAP can be converted by cytochrome P450 enzymes to NAPQI, a reactive intermediate capable of inflicting significant damage, NAPQI can be detoxified by conjugation to GSH , . The notion that π-class GSTs might afford protection against NAPQI generation from an overdose of APAP was undermined by the finding of a decrease, rather than increase, in liver injury accompanying APAP poisoning in mGstp1/2−/− versus mGstp1/+/+ mice, a phenomenon confirmed by the current study . Remarkably, between mGstp1/2−/− and mGstp1/+/+ mice treated with APAP there were no differences in GSH depletion, though GSH levels recovered more quickly in mGstp1/2−/− mice. In addition, the difference in toxicity could not be attributed to differential expression of enzymes, such as CYP2E1 and other Phase I enzymes, responsible for the metabolism to acetaminophen to NAPQI , . This π-class GST effect on APAP hepatotoxicity was reflected in the pattern and amount of damage generated by APAP overdose in hGSTP1+mGstp1/2−/− mice, which express little, if any π-class GSTs in hepatocytes, but exhibit abundant expression in bile duct cells, Kupffer cells, macrophages, and endothelial cells. Taken together, the data strongly suggest that π-class GSTs act to promote liver injury both by intrinsically augmenting hepatocyte death when expressed in hepatocytes, and also by indirectly triggering hepatocyte death when present in other liver cell types. One possibility for the intrinsic mechanism may be the influence of π-class GSTs on intracellular signaling pathways, such as the c-Jun N-terminal kinase (Jnk) pathway typically activated by cellular stress and proinflammatory cytokines . In wild-type mice, APAP has been found to trigger prolonged activation of the Jnk pathway that contributes to hepatocyte death . mGstp1/2−/− mice have been shown to have significantly higher basal levels of Jnk signaling compared wild type mice, but show only modest, if any, increase from this basal signaling after exposure to acetaminophen . π-class GSTs, encoded by genes responsive to AP-1 transcription factors (and thus Jnk signaling), therefore interfere with constitutive Jnk activation, providing an autoregulatory loop that may be perturbed differently in the hepatocytes from mGstp1/2+/+, mGstp1/2−/−, and hGSTP1+mGstp1/2−/− mice subjected to the oxidative stress of APAP overdose, and this may partly underlie the differential levels of acetaminophen toxicity seen in these mice. The mechanism by which π-class GSTs present in bile duct cells, Kupffer cells, macrophages, and endothelial cells can trigger limited centrilobular necrosis in hepatocytes devoid of the enzyme is less clear. Nonetheless, centrilobular necrosis constitutes a common mode of human liver injury in response to liver toxin exposure. Perhaps, hGSTP1+mGstp1/2−/− mice may better model this type of liver response to toxins than wild-type mouse strains which contain high levels of π-class GSTs in hepatocytes.
Finally, the availability of hGSTP1+mGstp1/2−/− mice may permit new studies of carcinogenesis that better recapitulate human cancer development. By comparing responses of mGstp1/2+/+ and Gstp1/2−/− mice to chemical carcinogens, π-class GSTs have been implicated in protection against tumor development in several different organ sites, including the skin, the colon, and the lungs , , . For human cancers, the contributions of π-class GSTs appear to be more complex. During the pathogenesis of some human cancers, such as cancers of the colon, lung, kidney, and stomach, hGSTP1 has been reported to be expressed at high levels and contribute to anti-neoplastic drug resistance , . In other human cancers, such as cancers of the prostate, liver, and breast, hGSTP1 is conspicuously absent , , . Absence of hGSTP1 expression appears mostly attributable to somatic epigenetic hGSTP1 silencing attributable to de novo changes in DNA methylation at a CpG island encompassing the gene promoter. Epigenetic silencing of π-class GST expression has not been reported for either mGstp1/2 or for rGST-P. Since hGSTP1 contains a much higher density of CpG dinucleotides than either of the rodent orthologs, perhaps hGSTP1 may be more vulnerable than mGstp1/2 or for rGST-P to epigenetic silencing during carcinogenesis. This hypothesis can be tested using hGSTP1+mGstp1/2−/− mice, which have established both a more human pattern of hGSTP1 expression in the liver and prostate and a human pattern of 5-meCpG distribution at the hGSTP1 CpG island.
Materials and Methods
All experimental protocols were approved and performed in accordance with the standards established by the U.S. Animal Welfare Acts, as set forth in the National Institutes of Health guidelines and in the Policy and Procedures Manual of the Johns Hopkins University Animal Care and Use Committee (Approval Numbers MO10M447, MO08M163, MO08M465, MO08M487).
Generation of hGSTP1 transgenic mice
The full length hGSTP1 gene (−1138 to +3600) was amplified from a bacterial artificial chromosome (BAC) containing sequence from chromosome 11q (accession number AP001184; from BACPAC Resource Center (BPRC) at Children's Hospital Oakland Research Institute , Oakland, CA) using PfuUltra polymerase (Stratagene). A plasmid containing full length hGSTP1, along with a plasmid that had the 5′ [ATAAA]n repeat sequences deleted (−78 to +3600), were transfected into Hep3B cells, known to be devoid of hGSTP1 mRNA and hGSTP1 polypeptides as a result of epigenetic hGSTP1 silencing, to ensure that the gene was capable of directing expression of catalytically-active GSTπ . Transfection was accomplished using Lipofectamine Plus (Invitrogen) in serum free minimal essential medium (MEM). The full length hGSTP1-containing construct was then provided to Xenogen (Hopkinton, MA) for pronuclear microinjection and development of the founder mice on a C57BL/6 strain background. To identify hGSTP1+ progeny, genotyping specifically for hGSTP1, even in the presence of wild type mouse DNA, was accomplished by PCR using the forward primer 5′-AGGCGTGCAGATCACCTAAG-3′ and the reverse primer 5′- GCCACATCTGGCTGATTTTT-3′ with cycling conditions of 95°C for 3 minutes, 35 cycles of 95°C for 30 seconds, 59°C for 30 seconds, and 72°C for 45 seconds, yielding a 100 base pair product.
Breeding to create hGSTP1+Gstp1/2−/− mice
All mice were housed in a pathogen-free environment, allowed free access to food and water, and were maintained on a 12 hour light/dark cycle. mGstp1/2−/− that had been extensively backcrossed to C57BL/6 mice were available for breeding crosses with hGSTP1+ mice. Genotyping for mGstp1/2+/+ and mGstp1/2−/− was performed using the same PCR reaction conditions as for hGSTP1, but with the following primers: mGstp1/2+/+ forward 5′-GGCCACCCAACTACTGTGAT -3′, mGstp1/2+/+ reverse 5′-AGAAGGCCAGGTCCTAAAGC -3′, mGstp1/2−/− forward 5′-CTGTAGCGGCTGATGTTGAA -3′, and mGstp1/2−/−reverse 5′-ATGGCGATTACCGTTGATGT -3′. To assess patterns of π-class GST expression and DNA methylation, mice were euthanized in a CO2 chamber according to Johns Hopkins Animal Care and Use Committee (ACUC) guidelines. After sacrifice, tissues were harvested and immediately placed in formalin or snap frozen in liquid nitrogen for further processing.
Assessment of π-class GST expression by quantitative RT-PCR, immunoblot analysis, immunohistochemistry, and enzyme activity assay
RNA isolated from various human and mouse tissues was assayed for the expression of specific mRNAs using a QuantiTect Probe RT-PCR Kit (Qiagen, Valencia, CA) and primer sets for hGSTP1, hGAPDH, mGstp1, mGstp2, and mGAPDH from Applied Biosystems (Foster City, CA). π-class GST subunit polypeptide expression was detected in protein lysates from cells and tissues by immunoblot analysis with anti-GSTπ antibodies as previously described ; GST activity in the lysates was monitored using 1-chloro-2,4-dinitrobenzene (CDNB) as a substrate, with recombinant human GSTπ (Calbiochem) as a positive control . Human and mouse tissues were evaluated for patterns of π-class GST expression via immunohistochemical staining of formalin fixed, paraffin embedded, tissue sections. Briefly, the sections were deparaffinized by rinsing in xylene and then in a gradient of ethanols (100%→70%) until clear; antigen retrieval was accomplished by citrate steaming for 20 minutes. After incubating for 5 minutes in hydrogen peroxide and rinsed, the slides were probed with anti-GSTπ antibody (Stressgen, Ann Arbor, MI) at a 1∶1500 dilution, overnight at 4°C. Antibody binding was revealed using a horseradish peroxidase-conjugated secondary antibody (Dako, Denmark) applied to tissue sections for 30 minutes. For color generation and counterstaining, slides were first incubated in diaminobenzidine (Sigma), and then in hematoxylin (Dako).
Bisulfite genomic sequencing for assessment of hGSTP1 CpG island methylation
Sequence mapping of 5-meC bases at the hGSTP1 CpG island was undertaken using bisulfite conversion methods and PCR primers for four amplicons as described by Millar et al. .
Toxicology analyses. Mice were administered acetaminophen (N-acetyl-p-aminophenol; APAP; Sigma) via intraperitoneal (IP) injection or via oral gavage. APAP was used at a concentration of 300 mg/kg in phosphate buffered saline (PBS) and injections were undertaken using mice that were fasted for 12 hours prior to injection. Acetaminophen dosage was selected based on previous reports of hepatotoxicity in mice , . To assess the injurious effects of APAP on liver histology and function, serum was collected from mice via submandibular venipuncture or cardiac puncture (at terminal timepoints). Serum samples were taken at 4, 24, and 48 hours after gavage. Mice were sacrificed via cardiac puncture/exsanguination following administration of 14 mg Avertin (20 mg/ml, Sigma) or via CO2 asphyxiation. The blood was collected in BD Microtainer SST tubes (BD, Franklin Lakes, NJ), allowed to clot for at least 1 hour at room temperature, and then centrifuged at 16,000× g for 2 minutes at 4°C to allow separation of serum. Serum specimens were assayed for alanine aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase (ALP) and total bilirubin levels using VetAce (Alfa Wassermann, West Caldwell, NJ). For histopathology, dissected liver tissues were fixed in buffered formalin, embedded in paraffin, and then processed for tissue section staining with hematoxylin and eosin.
Conceived and designed the experiments: MV DBS SY WN. Performed the experiments: MV DBS NC JH GN ADM JS ZR BK-A. Analyzed the data: MV DBS GN ADM SY WN. Contributed reagents/materials/analysis tools: CH CW. Wrote the paper: MV DBS GN ADM CH CW SY WN.
- 1. Mannervik B, Awasthi YC, Board PG, Hayes JD, Di Ilio C, et al. (1992) Nomenclature for human glutathione transferases. Biochem J 282(Pt 1): 305–306.B. MannervikYC AwasthiPG BoardJD HayesC. Di Ilio1992Nomenclature for human glutathione transferases.Biochem J282Pt 1305306
- 2. Mannervik B, Board PG, Hayes JD, Listowsky I, Pearson WR (2005) Nomenclature for mammalian soluble glutathione transferases. Methods Enzymol 401: 1–8.B. MannervikPG BoardJD HayesI. ListowskyWR Pearson2005Nomenclature for mammalian soluble glutathione transferases.Methods Enzymol40118
- 3. Ishikawa T (1992) The ATP-dependent glutathione S-conjugate export pump. Trends Biochem Sci 17: 463–468.T. Ishikawa1992The ATP-dependent glutathione S-conjugate export pump.Trends Biochem Sci17463468
- 4. Sato K, Kitahara A, Satoh K, Ishikawa T, Tatematsu M, et al. (1984) The placental form of glutathione S-transferase as a new marker protein for preneoplasia in rat chemical hepatocarcinogenesis. Gann 75: 199–202.K. SatoA. KitaharaK. SatohT. IshikawaM. Tatematsu1984The placental form of glutathione S-transferase as a new marker protein for preneoplasia in rat chemical hepatocarcinogenesis.Gann75199202
- 5. Eaton DL, Bammler TK (1999) Concise review of the glutathione S-transferases and their significance to toxicology. Toxicol Sci 49: 156–164.DL EatonTK Bammler1999Concise review of the glutathione S-transferases and their significance to toxicology.Toxicol Sci49156164
- 6. Moffat GJ, McLaren AW, Wolf CR (1994) Involvement of Jun and Fos proteins in regulating transcriptional activation of the human pi class glutathione S-transferase gene in multidrug-resistant MCF7 breast cancer cells. J Biol Chem 269: 16397–16402.GJ MoffatAW McLarenCR Wolf1994Involvement of Jun and Fos proteins in regulating transcriptional activation of the human pi class glutathione S-transferase gene in multidrug-resistant MCF7 breast cancer cells.J Biol Chem2691639716402
- 7. Moffat GJ, McLaren AW, Wolf CR (1996) Sp1-mediated transcriptional activation of the human Pi class glutathione S-transferase promoter. J Biol Chem 271: 1054–1060.GJ MoffatAW McLarenCR Wolf1996Sp1-mediated transcriptional activation of the human Pi class glutathione S-transferase promoter.J Biol Chem27110541060
- 8. Lee WH, Morton RA, Epstein JI, Brooks JD, Campbell PA, et al. (1994) Cytidine methylation of regulatory sequences near the pi-class glutathione S-transferase gene accompanies human prostatic carcinogenesis. Proc Natl Acad Sci U S A 91: 11733–11737.WH LeeRA MortonJI EpsteinJD BrooksPA Campbell1994Cytidine methylation of regulatory sequences near the pi-class glutathione S-transferase gene accompanies human prostatic carcinogenesis.Proc Natl Acad Sci U S A911173311737
- 9. Nelson WG, De Marzo AM, Yegnasubramanian S (2009) Epigenetic alterations in human prostate cancers. Endocrinology 150: 3991–4002.WG NelsonAM De MarzoS. Yegnasubramanian2009Epigenetic alterations in human prostate cancers.Endocrinology15039914002
- 10. Kitahara A, Satoh K, Nishimura K, Ishikawa T, Ruike K, et al. (1984) Changes in molecular forms of rat hepatic glutathione S-transferase during chemical hepatocarcinogenesis. Cancer Res 44: 2698–2703.A. KitaharaK. SatohK. NishimuraT. IshikawaK. Ruike1984Changes in molecular forms of rat hepatic glutathione S-transferase during chemical hepatocarcinogenesis.Cancer Res4426982703
- 11. Satoh K, Kitahara A, Soma Y, Inaba Y, Hatayama I, et al. (1985) Purification, induction, and distribution of placental glutathione transferase: a new marker enzyme for preneoplastic cells in the rat chemical hepatocarcinogenesis. Proc Natl Acad Sci U S A 82: 3964–3968.K. SatohA. KitaharaY. SomaY. InabaI. Hatayama1985Purification, induction, and distribution of placental glutathione transferase: a new marker enzyme for preneoplastic cells in the rat chemical hepatocarcinogenesis.Proc Natl Acad Sci U S A8239643968
- 12. Dragan YP, Campbell HA, Baker K, Vaughan J, Mass M, et al. (1994) Focal and non-focal hepatic expression of placental glutathione S-transferase in carcinogen-treated rats. Carcinogenesis 15: 2587–2591.YP DraganHA CampbellK. BakerJ. VaughanM. Mass1994Focal and non-focal hepatic expression of placental glutathione S-transferase in carcinogen-treated rats.Carcinogenesis1525872591
- 13. Terrier P, Townsend AJ, Coindre JM, Triche TJ, Cowan KH (1990) An immunohistochemical study of pi class glutathione S-transferase expression in normal human tissue. Am J Pathol 137: 845–853.P. TerrierAJ TownsendJM CoindreTJ TricheKH Cowan1990An immunohistochemical study of pi class glutathione S-transferase expression in normal human tissue.Am J Pathol137845853
- 14. Howie AF, Forrester LM, Glancey MJ, Schlager JJ, Powis G, et al. (1990) Glutathione S-transferase and glutathione peroxidase expression in normal and tumour human tissues. Carcinogenesis 11: 451–458.AF HowieLM ForresterMJ GlanceyJJ SchlagerG. Powis1990Glutathione S-transferase and glutathione peroxidase expression in normal and tumour human tissues.Carcinogenesis11451458
- 15. Bakker J, Lin X, Nelson WG (2002) Methyl-CpG binding domain protein 2 represses transcription from hypermethylated pi-class glutathione S-transferase gene promoters in hepatocellular carcinoma cells. J Biol Chem 277: 22573–22580.J. BakkerX. LinWG Nelson2002Methyl-CpG binding domain protein 2 represses transcription from hypermethylated pi-class glutathione S-transferase gene promoters in hepatocellular carcinoma cells.J Biol Chem2772257322580
- 16. Tchou JC, Lin X, Freije D, Isaacs WB, Brooks JD, et al. (2000) GSTP1 CpG island DNA hypermethylation in hepatocellular carcinomas. Int J Oncol 16: 663–676.JC TchouX. LinD. FreijeWB IsaacsJD Brooks2000GSTP1 CpG island DNA hypermethylation in hepatocellular carcinomas.Int J Oncol16663676
- 17. Zhang YJ, Chen Y, Ahsan H, Lunn RM, Chen SY, et al. (2005) Silencing of glutathione S-transferase P1 by promoter hypermethylation and its relationship to environmental chemical carcinogens in hepatocellular carcinoma. Cancer Lett 221: 135–143.YJ ZhangY. ChenH. AhsanRM LunnSY Chen2005Silencing of glutathione S-transferase P1 by promoter hypermethylation and its relationship to environmental chemical carcinogens in hepatocellular carcinoma.Cancer Lett221135143
- 18. Zhong S, Tang MW, Yeo W, Liu C, Lo YM, et al. (2002) Silencing of GSTP1 gene by CpG island DNA hypermethylation in HBV-associated hepatocellular carcinomas. Clin Cancer Res 8: 1087–1092.S. ZhongMW TangW. YeoC. LiuYM Lo2002Silencing of GSTP1 gene by CpG island DNA hypermethylation in HBV-associated hepatocellular carcinomas.Clin Cancer Res810871092
- 19. Morel F, Fardel O, Meyer DJ, Langouet S, Gilmore KS, et al. (1993) Preferential increase of glutathione S-transferase class alpha transcripts in cultured human hepatocytes by phenobarbital, 3-methylcholanthrene, and dithiolethiones. Cancer Res 53: 231–234.F. MorelO. FardelDJ MeyerS. LangouetKS Gilmore1993Preferential increase of glutathione S-transferase class alpha transcripts in cultured human hepatocytes by phenobarbital, 3-methylcholanthrene, and dithiolethiones.Cancer Res53231234
- 20. McGlynn KA, Hunter K, LeVoyer T, Roush J, Wise P, et al. (2003) Susceptibility to aflatoxin B1-related primary hepatocellular carcinoma in mice and humans. Cancer Res 63: 4594–4601.KA McGlynnK. HunterT. LeVoyerJ. RoushP. Wise2003Susceptibility to aflatoxin B1-related primary hepatocellular carcinoma in mice and humans.Cancer Res6345944601
- 21. Henderson CJ, Wolf CR, Kitteringham N, Powell H, Otto D, et al. (2000) Increased resistance to acetaminophen hepatotoxicity in mice lacking glutathione S-transferase Pi. Proc Natl Acad Sci U S A 97: 12741–12745.CJ HendersonCR WolfN. KitteringhamH. PowellD. Otto2000Increased resistance to acetaminophen hepatotoxicity in mice lacking glutathione S-transferase Pi.Proc Natl Acad Sci U S A971274112745
- 22. Millar DS, Paul CL, Molloy PL, Clark SJ (2000) A distinct sequence (ATAAA)n separates methylated and unmethylated domains at the 5′-end of the GSTP1 CpG island. J Biol Chem 275: 24893–24899.DS MillarCL PaulPL MolloySJ Clark2000A distinct sequence (ATAAA)n separates methylated and unmethylated domains at the 5′-end of the GSTP1 CpG island.J Biol Chem2752489324899
- 23. Coles B, Wilson I, Wardman P, Hinson JA, Nelson SD, et al. (1988) The spontaneous and enzymatic reaction of N-acetyl-p-benzoquinonimine with glutathione: a stopped-flow kinetic study. Arch Biochem Biophys 264: 253–260.B. ColesI. WilsonP. WardmanJA HinsonSD Nelson1988The spontaneous and enzymatic reaction of N-acetyl-p-benzoquinonimine with glutathione: a stopped-flow kinetic study.Arch Biochem Biophys264253260
- 24. Cheung C, Gonzalez FJ (2008) Humanized mouse lines and their application for prediction of human drug metabolism and toxicological risk assessment. J Pharmacol Exp Ther 327: 288–299.C. CheungFJ Gonzalez2008Humanized mouse lines and their application for prediction of human drug metabolism and toxicological risk assessment.J Pharmacol Exp Ther327288299
- 25. Dai G, He L, Chou N, Wan YJ (2006) Acetaminophen metabolism does not contribute to gender difference in its hepatotoxicity in mouse. Toxicol Sci 92: 33–41.G. DaiL. HeN. ChouYJ Wan2006Acetaminophen metabolism does not contribute to gender difference in its hepatotoxicity in mouse.Toxicol Sci923341
- 26. Chan K, Han XD, Kan YW (2001) An important function of Nrf2 in combating oxidative stress: detoxification of acetaminophen. Proc Natl Acad Sci U S A 98: 4611–4616.K. ChanXD HanYW Kan2001An important function of Nrf2 in combating oxidative stress: detoxification of acetaminophen.Proc Natl Acad Sci U S A9846114616
- 27. Bessems JG, Vermeulen NP (2001) Paracetamol (acetaminophen)-induced toxicity: molecular and biochemical mechanisms, analogues and protective approaches. Crit Rev Toxicol 31: 55–138.JG BessemsNP Vermeulen2001Paracetamol (acetaminophen)-induced toxicity: molecular and biochemical mechanisms, analogues and protective approaches.Crit Rev Toxicol3155138
- 28. Jaeschke H, Bajt ML (2006) Intracellular signaling mechanisms of acetaminophen-induced liver cell death. Toxicol Sci 89: 31–41.H. JaeschkeML Bajt2006Intracellular signaling mechanisms of acetaminophen-induced liver cell death.Toxicol Sci893141
- 29. Elsby R, Kitteringham NR, Goldring CE, Lovatt CA, Chamberlain M, et al. (2003) Increased constitutive c-Jun N-terminal kinase signaling in mice lacking glutathione S-transferase Pi. J Biol Chem 278: 22243–22249.R. ElsbyNR KitteringhamCE GoldringCA LovattM. Chamberlain2003Increased constitutive c-Jun N-terminal kinase signaling in mice lacking glutathione S-transferase Pi.J Biol Chem2782224322249
- 30. Adler V, Yin Z, Fuchs SY, Benezra M, Rosario L, et al. (1999) Regulation of JNK signaling by GSTp. Embo J 18: 1321–1334.V. AdlerZ. YinSY FuchsM. BenezraL. Rosario1999Regulation of JNK signaling by GSTp.Embo J1813211334
- 31. Gunawan BK, Liu ZX, Han D, Hanawa N, Gaarde WA, et al. (2006) c-Jun N-terminal kinase plays a major role in murine acetaminophen hepatotoxicity. Gastroenterology 131: 165–178.BK GunawanZX LiuD. HanN. HanawaWA Gaarde2006c-Jun N-terminal kinase plays a major role in murine acetaminophen hepatotoxicity.Gastroenterology131165178
- 32. Henderson CJ, Smith AG, Ure J, Brown K, Bacon EJ, et al. (1998) Increased skin tumorigenesis in mice lacking pi class glutathione S-transferases. Proc Natl Acad Sci U S A 95: 5275–5280.CJ HendersonAG SmithJ. UreK. BrownEJ Bacon1998Increased skin tumorigenesis in mice lacking pi class glutathione S-transferases.Proc Natl Acad Sci U S A9552755280
- 33. Ritchie KJ, Henderson CJ, Wang XJ, Vassieva O, Carrie D, et al. (2007) Glutathione transferase pi plays a critical role in the development of lung carcinogenesis following exposure to tobacco-related carcinogens and urethane. Cancer Res 67: 9248–9257.KJ RitchieCJ HendersonXJ WangO. VassievaD. Carrie2007Glutathione transferase pi plays a critical role in the development of lung carcinogenesis following exposure to tobacco-related carcinogens and urethane.Cancer Res6792489257
- 34. Ritchie KJ, Walsh S, Sansom OJ, Henderson CJ, Wolf CR (2009) Markedly enhanced colon tumorigenesis in ApcMin mice lacking glutathione S-transferase Pi. Proc Natl Acad Sci U S A. KJ RitchieS. WalshOJ SansomCJ HendersonCR Wolf2009Markedly enhanced colon tumorigenesis in ApcMin mice lacking glutathione S-transferase Pi.Proc Natl Acad Sci U S A
- 35. Volm M, Mattern J, Samsel B (1991) Overexpression of P-glycoprotein and glutathione S-transferase-pi in resistant non-small cell lung carcinomas of smokers. Br J Cancer 64: 700–704.M. VolmJ. MatternB. Samsel1991Overexpression of P-glycoprotein and glutathione S-transferase-pi in resistant non-small cell lung carcinomas of smokers.Br J Cancer64700704
- 36. Esteller M, Corn PG, Urena JM, Gabrielson E, Baylin SB, et al. (1998) Inactivation of glutathione S-transferase P1 gene by promoter hypermethylation in human neoplasia. Cancer Res 58: 4515–4518.M. EstellerPG CornJM UrenaE. GabrielsonSB Baylin1998Inactivation of glutathione S-transferase P1 gene by promoter hypermethylation in human neoplasia.Cancer Res5845154518
- 37. Habig WH, Pabst MJ, Jakoby WB (1974) Glutathione S-transferases. The first enzymatic step in mercapturic acid formation. J Biol Chem 249: 7130–7139.WH HabigMJ PabstWB Jakoby1974Glutathione S-transferases. The first enzymatic step in mercapturic acid formation.J Biol Chem24971307139
- 38. Henderson NC, Pollock KJ, Frew J, Mackinnon AC, Flavell RA, et al. (2007) Critical role of c-jun (NH2) terminal kinase in paracetamol- induced acute liver failure. Gut 56: 982–990.NC HendersonKJ PollockJ. FrewAC MackinnonRA Flavell2007Critical role of c-jun (NH2) terminal kinase in paracetamol- induced acute liver failure.Gut56982990