![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Division of Clinical Pharmacology, Department of Molecular Pharmacology and Experimental Therapeutics (B.M., O.E.S., L.L.P., I.M., R.M.W.), Department of Biochemistry and Molecular Biology (B.W.E., E.D.W.), and Department of Health Sciences Research (D.J.S.), Mayo Clinic College of Medicine, Rochester, Minnesota
(Received February 2, 2006; accepted April 19, 2006)
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
; Hayes et al., 2005; Yang et al., 2006). Glutathione conjugation plays an important role in the detoxification of many drugs, carcinogens, and endogenous compounds (Mulder and Ouwerkerk-Mahadevan, 1997; Pool-Zobel et al., 2005; Yang et al., 2006). Genetic variation in GSTs has been reported to represent a risk factor for a variety of cancers (Landi, 2000; Strange et al., 2000; Strange et al., 2001). The cytosolic GSTs are made up of alpha (A), mu (M), omega (O), pi (P), sigma (S), theta (T), and zeta (Z) families (Strange et al., 2001). GSTO1 and GSTO2 are the most recently discovered cytosolic GSTs. Members of this family do not display activity with "classical" GST substrates such as 1-chloro-2,4-dinitrobenzene, dichloromethane, or ethacrynic acid, but they do exhibit several enzymatic characteristics of glutaredoxins (Board et al., 2000; Whitbread et al., 2003). Although most GSTs have a serine or tyrosine residue at the active site, the GSTOs have cysteine, as well as a 19-amino acid extension at the N terminus that is not present in other cytosolic GSTs (Board et al., 2000). The genes encoding GSTO1 and GSTO2 map to chromosome 10 and are separated by approximately 1.5 kb (Whitbread et al., 2003).
Both GSTO1 and GSTO2 catalyze the reduction of monomethyl arsenate from an oxidation state of +V to +III (Zakharyan et al., 2001; Schmuck et al., 2005). The use of enzyme purified from rabbit liver played an important role in the identification of GSTO1 as an enzyme that can catalyze monomethyl arsenate reduction (Maiorino and Aposhian, 1985; Zakharyan et al., 1995; Zakharyan and Aposhian, 1999). This reduction reaction is the rate-limiting step in arsenic biotransformation in humans (Zakharyan et al., 2001). Although arsenic also undergoes methylation in humans, that conjugation reaction requires prior reduction. The methylation of arsenic is catalyzed by a recently identified arsenic methyltransferase (AS3MT). We recently also resequenced and performed functional genomic studies with the human AS3MT gene (Wood et al., 2006).
Arsenic contamination of drinking water is a public health problem worldwide (Tchounwou et al., 1999; Ratnaike, 2003; Tchounwou et al., 2003; Minamoto et al., 2005). Chronic arsenic exposure can cause skin, bladder, kidney, liver, and lung cancer (Chen et al., 1992; Hopenhayn-Rich et al., 1998; Paul et al., 2004), and levels of arsenic metabolites vary greatly in the urine of individuals drinking the same contaminated water (Vahter, 2000; Marnell et al., 2003). Genetic variation in GSTO1 and/or GSTO2—enzymes that catalyze the rate-limiting step in arsenic biotransformation in humans—might explain a portion of that variation. Arsenic is also used as a therapeutic agent, and arsenic trioxide is used to treat a variety of neoplasms (Maeda et al., 2001, 2004; Ivanov and Hei, 2005). Patients treated with this drug can display striking arsenic trioxide-induced toxicity (Soignet et al., 1998; Mathews et al., 2006), another situation in which knowledge of GSTO pharmacogenomics might prove helpful.
There have been previous pharmacogenetic studies of the GSTOs (Marnell et al., 2003; Whitbread et al., 2003; Yu et al., 2003), but there remains need for a systematic, comprehensive study of genetic variation in GSTO1 and GSTO2, as well as studies of the functional implications of that variation. The experiments described subsequently represent an attempt to perform comprehensive studies of genetic variation in GSTO1 and GSTO2, as well as functional genomic characterization of polymorphisms that alter the amino acid sequence of proteins encoded by variant GSTO alleles. We observed large ethnic variation in the presence of common genetic polymorphisms in GSTO1 and GSTO2, and several of those polymorphisms had functional implications. Finally, our studies of four different ethnic groups used DNA samples that were deposited in the Coriell Cell Repository by the National Institutes of Health (NIH)—samples that are available to the biomedical investigative community for use in future experiments.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
GSTO1 and GSTO2 Resequencing. Twenty-two separate PCR amplifications were performed with each DNA sample, eight for GSTO1, and 14 for GSTO2. Exons, splice junctions, and a portion of the 5'-flanking region (5'-FR) were amplified for both genes. Primers were designed to hybridize within introns, within 5'-FRs, or within 3'-untranslated regions (3'-UTRs). This approach ensured that a GSTO-processed pseudogene, which maps to chromosome 3 (Whitbread et al., 2003), would not be amplified. The 5'-end of each primer used to amplify the GSTO1 and GSTO2 coding regions included an M13 tag to make it possible to use dye primer-sequencing chemistry (Chadwick et al., 1996). Resequencing of the GSTO2 5'-FR and the intergenic area used dye terminator chemistry. The sequences of the primers used to perform the resequencing experiments are listed in Supplementary Table 1.
|
PCR amplifications were performed with AmpliTaq Gold DNA polymerase (PerkinElmer Life and Analytical Sciences, Boston, MA). The 50-µl reaction mixtures consisted of 1 U of DNA polymerase, 5 µl of a 10-fold diluted DNA sample (160–190 ng of DNA), 10 pmol of each primer, 0.05 mM dNTP (Roche Diagnostics, Indianapolis, IN), and 5 µl of 10x PCR buffer that contained 15 mM MgCl2 (PerkinElmer). Amplifications were performed with a PerkinElmer model 9700 thermal cycler. PCR cycling parameters involved a 12-min hot start at 94°C, 30 cycles at 94°C for 30 s, 30 s at the annealing temperature, and a final 10-min extension at 72°C. Amplicons were sequenced on both strands in the Mayo Molecular Biology Core Facility with an ABI 377 DNA sequencer using BigDye (PerkinElmer) dye primer-sequencing chemistry or, for the GSTO2 5'-FR and the intergenic area, dye terminator-sequencing chemistry. The sequences of samples in which a SNP was observed only once or those with ambiguous chromatograms were verified by performing independent amplifications, followed by resequencing. Chromatograms were analyzed using the PolyPhred 3.0 (Nickerson et al., 1997) and Consed 8.0 (Gordon et al., 1998) programs from the University of Washington (Seattle, WA) as well as Mutation Surveyor (Soft Genetics, LLC, State College, PA) (Bardelli et al., 2003). The GenBank accession number for the GSTO1 and GSTO2 reference sequences were NT_030059
[GenBank]
.12. Reference sequences for GSTO1 and GSTO2 cDNAs were NM_004832
[GenBank]
.1 and NM_183239
[GenBank]
.1, respectively.
GSTO1 and GSTO2 Transient Expression. WT GSTO1 and GSTO2 cDNA open reading frame sequences were amplified from a pooled human liver cDNA library (Clontech, Mountain View, CA). Expression constructs for the nonsynonymous cSNPs observed during the resequencing studies were generated by site-directed mutagenesis using "circular PCR". Sequences of the site-directed mutagenesis primers are also listed in Supplementary Table 1. The site-directed mutagenesis amplification reactions contained 1 U of Pfu Turbo DNA polymerase (Stratagene, La Jolla, CA) in a final volume of 25 µl, 10 pmol primers, 0.24 mM dNTPs, and 2.5 µl of 10x PCR buffer. The sequences of expression constructs were confirmed by sequencing in both directions after cloning into the eukaryotic expression vector pCR3.1 (Invitrogen, Carlsbad, CA). WT and variant GSTO1 and GSTO2 expression constructs were then transfected into COS-1 cells, together with pSV-ß-galactosidase DNA (Promega) to make it possible to use ß-galactosidase activity to correct for transfection efficiency. Transfections were performed with Lipofectamine 2000 (Invitrogen), and the cells were cultured for 48 h in Dulbecco's modified Eagle's medium (DMEM) (Cambrex Bio Science Walkersville, Inc., Walkersville, MD) with 10% fetal bovine serum (Clontech). They were then homogenized with a Polytron homogenizer (Brinkmann Instruments, Westbury, NY); the homogenates were centrifuged at 100,000g for 1 h; and supernatant preparations were stored at –80°C.
GSTO1 and GSTO2 Western Blot Analysis. Rabbit polyclonal antibodies were generated to GSTO1 and GSTO2 (Cocalico Biologicals, Reamstown, PA) using polypeptides that corresponded to amino acids 10 to 30 for each of the proteins, with an additional cysteine residue at the amino terminus. These synthetic polypeptides differed at two amino acids for the two GSTO isoforms and were conjugated to keyhole limpet hemocyanin before the immunization of rabbits. For quantitative Western blot analysis, cytosol preparations of COS-1 cells transfected with the expression constructs were loaded on a 12% polyacrylamide gel on the basis of ß-galactosidase activity to correct for possible variation in transfection efficiency. After SDS-PAGE, the proteins were transferred to a nitrocellulose membrane that was probed with rabbit antiserum to the appropriate GSTO isoform, diluted 1:2000 with "blocking buffer". Goat anti-rabbit horseradish peroxidase, diluted 1:20,000, was used as the secondary antibody. Bound antibody was detected with the ECL Western blotting system (Amersham Pharmacia Biotech, Piscataway, NJ). The level of immunoreactive protein was quantitated using IP Gel Lab (Biosystemetica, Plymouth, UK).
GSTO1 and GSTO2 in Vitro Translation and Degradation. Selected GSTO1 and GSTO2 expression constructs were also transcribed and translated in vitro in the presence of [35S]methionine (1000 Ci/mmol, 2.5 mCi total activity) (Amersham Pharmacia Biotech) using the TNT-coupled rabbit reticulocyte lysate (RRL) system (Promega, Madison, WI). The reaction mixture used to generate radioactively labeled protein was incubated at 30°C for 90 min, and 5-µl aliquots were used to perform SDS-PAGE with 10% gels. Protein degradation studies were performed as described previously (Wang et al., 2003). Specifically, 50 µl of an adenosine 5'-triphosphate (ATP) generating system, 50 µl of "untreated" RRL, and 10 µl of [35S]methionine radioactively labeled GSTO allozyme were mixed. This mixture was incubated at 37°C, and 10-µl aliquots were removed at various time intervals, followed by SDS-PAGE. After electrophoresis, the gels were dried and exposed to X-ray film, and the bands of radioactively labeled protein were quantified.
Data Analysis. Values for 
, 
, and Tajima's D were calculated as described by Tajima (1989), followed by correction for length. D' values as a measure of linkage disequilibrium were calculated as described by Hartl and Clark (2000) and Hendrick (2000) and were displayed graphically. Haplotype analysis was performed as described by Schaid et al. (2002) using the E-M algorithm. Group mean values were compared using Student's t test. Degradation data were plotted using GraphPad Prism (GraphPad Software, San Diego, CA).
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
AG|GT sequence, resulting in the creation of a new splice donor site and the deletion of codon 155, as reported previously (Board et al., 2000). Twenty-four polymorphisms were present in AA, 12 in CA, 15 in HCA, and 18 in MA subjects. Seven of these polymorphisms were specific to AA, four to CA, and three to MA. Twenty-one of the polymorphisms in AA subjects had frequencies 
1%, and 10 had frequencies 10%. Comparable data for samples from CA subjects were eight polymorphisms with frequencies 1% and four 10%, whereas for HCA subjects, seven had frequencies of 1% and four had frequencies 10%, and in MA subjects, eight polymorphisms had frequencies 1%, and three 10%. Four nonsynonymous cSNPs were observed in GSTO1. Polymorphisms that resulted in Cys32Tyr and Ala236Val alterations in the encoded amino acid sequence were observed only in CA and MA subjects, respectively, whereas SNPs that resulted in Ala140Asp and Glu208Lys changes in amino acid sequence were present in all four populations.
|
We also calculated nucleotide diversity, a measure of genetic variation, adjusted for the number of alleles studied, for both genes and for the intergenic region. Two standard measures of nucleotide diversity are , average heterozygosity per site, and , a population mutation measure that is theoretically equal to the neutral mutation parameter (Fullerton et al., 2000). Values for Tajima's D, a test for the neutral mutation hypothesis (Fullerton et al., 2000), were also calculated (Table 2). Under conditions of neutrality, Tajima's D should equal 0. AA subjects showed greater nucleotide diversity for GSTO1, based on and values, than did the other groups (Table 2). For GSTO1, the Tajima's D value for HCA subjects approached statistical significance. For GSTO2, none of the values for Tajima's D were statistically significant. For the intergenic region, Tajima's D value for AA subjects was statistically significant, raising the possibility that this area may be undergoing selection.
|
GSTO1 and GSTO2 Linkage Disequilibrium and Haplotype Analysis. Haplotypes are proving of increasing value for use in epidemiologic association studies (Drysdale et al., 2000; Brodde and Leineweber, 2005). Therefore, we performed both pairwise studies of linkage disequilibrium and haplotype analysis with our gene resequencing data. For the linkage disequilibrium analysis, pairwise calculations of D' values were performed for all polymorphisms. D' values can range from 0 to 1.0, with a value of 1.0 indicating that two polymorphisms are maximally associated, whereas 0 indicates they are randomly associated (Hartl and Clark, 2000; Hendrick, 2000). Graphical representations of statistically significant pairwise population-specific D' values for all GSTO1 and GSTO2 polymorphisms are shown in Fig. 2. Because GSTO1 and GSTO2 are separated by only approximately 1.5 kb, it is not surprising that we observed linkage disequilibrium for polymorphisms located in the two genes. Patterns for possible haplotype blocks were not well defined in the CA and HCA DNA, but they seemed to be similar for the AA and MA samples (Fig. 2).
|
Haplotype is the combination of alleles on a single chromosome or, in a more restricted definition, all polymorphisms present on a single allele (Altshuler et al., 2005). We also analyzed haplotypes for GSTO1 and GSTO2. It is possible to determine a haplotype unequivocally if not more than one polymorphism per allele is heterozygous, but haplotypes can also be inferred computationally (Schaid et al., 2002). GSTO1 and GSTO2 haplotypes, both unequivocal and inferred, with frequencies 1% in at least one ethnic group, are listed in Tables 3 and 4. Haplotype designations were made on the basis of the amino acid sequence of the encoded allozyme. For example, the "WT", the most common amino acid sequence in AA subjects, was designated *1. Letter designations were then added based on descending frequencies of haplotype encoding that amino acid sequence, beginning with the AA population. Therefore, *1A was more frequent than *1B, and *1B was more frequent than *1C, etc. For GSTO1, *2, *3, *4, and *5 designations were assigned to the Tyr32, Asp140, Lys208, and Val236 variants, respectively. The *4 haplotype with the Lys208 alteration in encoded amino acid sequence also included the deletion of codon 155 as a result of the 3-bp deletion at the exon 4-intron 4 splice donor site since these two variants always occurred together. The *6 haplotype included two changes in encoded amino acid sequence plus the loss of codon 155. For GSTO2, *2, *3, and *4 designations were assigned to encoded allozymes that included Ile41, Tyr130, and Asp142 changes in amino acids, respectively. The GSTO2*5 designation was used for an Asp142, Ile158 "double-variant" that existed on the basis of an inferred haplotype.
|
|
Western Blot Analysis of GSTO1 and GSTO2. As an initial step in the study of the possible functional implications of nonsynonymous cSNPs in GSTO1 and GSTO2, expression constructs were created for all variant allozymes, including the GSTO1 codon 155 deletion, and those constructs were used to transfect COS-1 cells. For GSTO1, the codon 155 deletion was linked to a polymorphism that resulted in a Glu208Lys alteration in encoded amino acid; so a construct with both the codon 155 deletion and the Lys208 variant amino acid was also created. A mammalian cell line was transfected to ensure the presence of mammalian post-translational modification and the mammalian machinery for protein degradation. The GSTO expression constructs were cotransfected with ß-galactosidase to make it possible to correct for variations in transfection efficiency. The antibodies for both GSTO1 and GSTO2 were directed against peptides that did not include any of the genetically variant amino acids. When GSTO1 was expressed in COS-1 cells, the Tyr32 variant allozyme displayed a striking decrease in the quantity of expressed immunoreactive protein and the Val236 allozyme showed a significant increase, whereas values for the other constructs, including those with the codon 155 deletion, were similar to those for the WT protein (Fig. 3 and Table 5). It should be pointed out that the Cys32Tyr polymorphism resulted in the loss of the critical GSTO1 active site cysteine (Board et al., 2000). When GSTO2 constructs were transiently expressed in COS-1 cells, levels of Tyr130, Ile158, and the double variant Asp142/Ile158 allozyme proteins were strikingly reduced, whereas the Ile41 and Asp142 allozymes were expressed at approximately 80% of the level of the WT allozyme (Fig. 4 and Table 5). The GSTO2 antibody cross-reacted with a protein in the COS-1 cells with a molecular mass slightly higher than that for recombinant GSTO2 (Fig. 4). We attempted to complement these expression studies with parallel assays of level of enzyme activity, but the assays currently available for GSTO1 and GSTO2 activities—assays that have been used to study purified, bacterially expressed protein (Schmuck et al., 2005)—lacked adequate sensitivity for application to these cultured mammalian cell cytosol preparations.
|
|
|
GSTO1 and GSTO2 Degradation Studies. Studies of a series of genetic polymorphisms and mutations have shown that the change of only one or two amino acids can alter the quantity of encoded protein during expression in mammalian cells (Thomae et al., 2002; Adjei et al., 2003; Thomae et al., 2003; Wang et al., 2003; Ji et al., 2005; Salavaggione et al., 2005; Martin et al., 2006). A common mechanism responsible for this effect is accelerated protein degradation (Wang et al., 2003; Weinshilboum and Wang, 2004). In an attempt to study mechanisms responsible for the striking decrease in expression of the GSTO1 Tyr32 and the GSTO2 Tyr130 and Ile158 variant allozymes, protein degradation studies were performed using the RRL. The RRL is commonly used to perform this type of study (Wang et al., 2003). Specifically, radioactive WT and variant allozymes were synthesized using "treated" RRL (Figs. 5A and 6A), and the recombinant allozymes were then incubated for various periods of time with "untreated" RRL and an ATP generating system to follow the time course of protein degradation. The Tyr32 GSTO1 variant allozyme was degraded much more rapidly than was the WT allozyme (Fig. 5, B and C). For example, only 20% of Tyr32 protein remained after 24 h, whereas 60% of the WT allozyme remained (Fig. 5C). For GSTO2, the Tyr130, Ile158, and Asp142/Ile158 allozymes were also degraded much more rapidly than was the WT, with Ile158 and Asp142/Ile158 being degraded even more rapidly than was Tyr130 (Fig. 6, B and C). After 24 h, 26% of Tyr130, 6.6% of Ile158, 5.6% of Asp142/Ile158, and 73% of the WT protein remained (Fig. 6C). Therefore, accelerated degradation seemed to be responsible, at least in part, for differences in levels of the GSTO1 and GSTO2 variant allozymes after expression in COS-1 cells.
|
|
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
). A large number of haplotypes were also observed for both genes (Tables 3 and 4). For the initial functional genomic studies, expression constructs for GSTO1 and GSTO2 variant allozymes were used to transfect COS-1 cells, and quantitative Western blot analyses were performed. Variant allozymes for each gene showed striking decreases in levels of expressed protein (Figs. 3 and 4) because of, at least in part, accelerated degradation (Figs. 5 and 6). Studies of genetic polymorphisms and mutations in other phase II drug-metabolizing enzymes have provided similar observations, most often as a result of proteasome-mediated degradation (Thomae et al., 2003; Wang et al., 2003; Ji et al., 2005) or, on the basis of recent data, as a result of intracellular protein aggregation and aggresome formation (Wang et al., 2005). Obviously, the implications of these observations for in vivo function must be pursued in the course of future studies. Because the rate-limiting step in arsenic biotransformation in humans is reduction catalyzed, at least in part, by GSTO1 and GSTO2 (Zakharyan et al., 2001; Schmuck et al., 2005), these pharmacogenomic results, when combined with similar data for the human arsenic methyltransferase gene AS3MT (Wood et al., 2006) may help us to understand differences among individuals in the metabolism and effect of arsenic in humans.
In summary, we have identified a large number of novel GSTO1 and GSTO2 polymorphisms, as well as a series of haplotypes for both genes. Functional studies of common nonsynonymous cSNPs in these genes revealed striking differences in levels of protein expression and the mechanism responsible—at least in part—was rapid degradation of the variant allozyme. Striking variations among ethnic groups in allele and haplotype frequencies were also observed, as they have been for many other human genes (Adjei et al., 2003; Wang et al., 2003; Ji et al., 2005; Salavaggione et al., 2005; Martin et al., 2006; Wood et al., 2006). These results represent an important step toward future pharmacogenomic studies of these new members of the GST family of proteins. It will now be possible to use these results for GSTO1 and GSTO2, together with similar data for the human AS3MT gene (Wood et al., 2006) to test the hypothesis that genetic variation in all or some of these genes might represent risk factors for arsenic-dependent carcinogenesis and/or variation in response to therapy with arsenic trioxide.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
The gene resequencing data described in this article have been deposited in the NIH-funded database PharmGKB with submission numbers PS205018, PS205019, PS205020, PS205021, PS205022, PS205026, PS205027, and PS205028 for GSTO1 and PS204659, PS205029, PS205030, PS205031, PS205032, and PS205033 for GSTO2.
Article, publication date, and citation information can be found at http://dmd.aspetjournals.org.
ABBREVIATIONS: GST, glutathione S-glutathione; GSTO, glutathione S-transferase omega; AS3MT, arsenic methyltransferase; NIH, National Institutes of Health; AA, African-American; CA, Caucasian-American; HCA, Han Chinese American; MA, Mexican-American; PCR, polymerase chain reaction; FR, flanking region; UTR, untranslated region; SNP, single nucleotide polymorphism; PAGE, polyacrylamide gel electrophoresis; RRL, rabbit reticulocyte lysate; bp, base pair(s); WT, wild type.
The online version of this article (available at http://dmd.aspetjournals.org) contains supplemental material. 
Address correspondence to: Dr. Richard M. Weinshilboum, Division of Clinical Pharmacology, Department of Molecular Pharmacology and Experimental Therapeutics, Mayo Clinic College of Medicine, 200 First Street SW, Rochester, MN 55905. E-mail: weinshilboum.richard{at}mayo.edu
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  L. Wang, S. Kumar, B. L. Fridley, K. R. Kalari, I. Moon, L. L. Pelleymounter, M. A.T. Hildebrandt, A. Batzler, B. W. Eckloff, E. D. Wieben, et al. Proteasome {beta} Subunit Pharmacogenomics: Gene Resequencing and Functional Genomics Clin. Cancer Res., June 1, 2008; 14(11): 3503 - 3513. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. A. Leil, C. Endo, A. A. Adjei, G. K. Dy, O. E. Salavaggione, J. R. Reid, M. M. Ames, and A. A. Adjei Identification and Characterization of Genetic Variation in the Folylpolyglutamate Synthase Gene Cancer Res., September 15, 2007; 67(18): 8772 - 8782. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
