Search

A nomenclature system for the cytosolic sulfotransferase (SULT) superfamily has been developed. The nomenclature guidelines were applied to 65 SULT cDNAs and 18 SULT genes that were characterized from eukaryotic organisms. SULT cDNA and gene sequences were identified by querying the GenBank databases and from published reports of their identification and characterization. These sequences were evaluated and named on the basis of encoded amino acid sequence identity and, in a few cases, a necessity to maintain historical naming convention. Family members share at least 45% amino acid sequence identity whereas subfamily members are at least 60% identical. cDNAs which encode amino acid sequences of at least 97% identity to each other were assigned identical isoform names. We also attempted to categorize orthologous enzymes between various species, where these have been identified, and the nomenclature includes a species descriptor. We present recommendations for the naming of allelic variants of SULT genes and their derived allozymes arising from single nucleotide polymorphisms and other genetic variation. The superfamily currently comprises 47 mammalian SULT isoforms, one insect isoform and eight plant enzymes, and collectively these sequences represent nine separate SULT families and 14 subfamilies. It is hoped that this nomenclature system will be widely adopted and that, as novel SULTs are identified and characterized, investigators will name their discoveries according to these guidelines.

Thiopurine S-methyltransferase (TPMT) plays an important role in the metabolism of thiopurine drugs. In humans, a common genetic polymorphism for TPMT is a major factor responsible for individual variation in the toxicity and therapeutic efficacy of these drugs. Dogs (Canis familiaris) are also treated with thiopurine drugs and, similar to humans, they display large individual variations in thiopurine toxicity and efficacy. We set out to determine whether dogs might also display genetically determined variation in TPMT activity. As a first step, we observed that canine red blood cell (RBC) TPMT activity in samples from 145 dogs varied over a nine-fold range. That variation was not associated with either the age or sex of the animal. Subsequently, we cloned the canine TPMT cDNA and gene. The canine cDNA encoded a protein that was 81.2% identical to the enzyme encoded by the most common TPMT allele in humans. A genotype-phenotype correlation analysis was performed by resequencing the canine gene using DNA samples from 39 animals selected for high, low or intermediate levels of RBC TPMT activity. We observed nine polymorphisms in these 39 DNA samples, including three insertion/deletion events and six single nucleotide polymorphisms (SNPs), one of which was a nonsynonymous cSNP (Arg97Gln). However, when the variant allozyme at codon 97 was expressed in COS-1 cells, it did not display significant differences in either basal levels of TPMT activity or in substrate kinetics compared with the wild-type allozyme. Six of the nine canine TPMT polymorphisms were associated with 67% of the variation in level of RBC TPMT activity in these 39 blood samples. When those six SNPs were assayed using DNA from all 145 animals studied, 40% of the phenotypic variance in the entire population sample could be explained by these polymorphisms. Therefore, inheritance is a major factor involved in the regulation of variation in RBC TPMT in the dog, just as it is in humans. These observations represent a step towards the application of pharmacogenetic and pharmacogenomic principles to companion animal drug therapy., (Copyright 2002 Lippincott Williams & Wilkins)

3'-Phosphoadenosine 5'-phosphosulfate (PAPS) is the sulfate donor cosubstrate for all sulfotransferase (SULT) enzymes. SULTs catalyze the sulfate conjugation of many endogenous and exogenous compounds, including drugs and other xenobiotics. In humans, PAPS is synthesized from adenosine 5'-triphosphate (ATP) and inorganic sulfate (SO2-4) by two isoforms, PAPSS1 and PAPSS2. Rare mutations that inactivate PAPSS2 are associated with human spondyloepimetaphyseal dysplasia and murine brachymorphism. To determine whether more common genetic polymorphisms that do not completely inactivate the enzyme might be one factor responsible for individual differences in sulfate conjugation, we previously cloned the human PAPSS2 gene. In the present studies, we 'resequenced' all twelve PAPSS2 exons and splice junctions, as well as approximately 500 bp of the 5'-flanking region, using 90 Polymorphism Discovery Resource (PDR) DNA samples from the Coriell Cell Repository. Twenty-two single nucleotide polymorphisms (SNPs) were observed, including four nonsynonymous coding region SNPs (cSNPs) that altered the following amino acids: Glu10Lys, Met281Leu,Val291Met and Arg432Lys. We also observed four insertions/deletions, including one sample that was homozygous for an 81-bp deletion in the 5'-flanking region 286 bp upstream from the site of transcription initiation. Transient expression studies showed that two of the nonsynonymous cSNPS, those that resulted in Glu10Lys and Val291Met alterations in encoded amino acids, showed significant decreases in levels of PAPSS activity. In the case of Glu10Lys, decreased activity was paralleled by a decrease in immunoreactive protein, while the Val291Met allozyme displayed a significant decrease in affinity for both ATP and Na2SO4 when compared to 'wild-type' enzyme, but without a significant alteration in level of immunoreactive protein. It will now be possible to test the hypothesis that these common, functionally significant PAPSS2 genetic polymorphisms might contribute to variations in sulfate conjugation in vivo.

Sulfotransferase (SULT) enzymes catalyze an important phase II reaction in the biotransformation of many drugs and other xenobiotics. We previously cloned the human SULT1C1 cDNA and gene as steps toward pharmacogenetic studies. We have now 'resequenced' the exons, portions of introns flanking exons and approximately 315 bp of the 5' flanking region of SULT1C1 in 89 DNA samples from Caucasian subjects to identify common genetic polymorphisms. Nineteen separate polymorphisms were observed, including four nonsynonymous coding region single nucleotide polymorphisms (cSNPs) and five insertions/deletions. These data were also used to determine and/or infer common SULT1C1 haplotypes. Three of the four nonsynonymous cSNPs had allele frequencies greater than 1%, including one with a frequency of 6.7%. Expression constructs were created for all of the nonsynonymous cSNPs observed, and those constructs were used to transfect COS-1 cells. Three of the four SULT1C1 variant allozymes had significantly reduced enzyme activity when compared with the wild-type enzyme. Among the variant allozymes, apparent Km values for 3'-phosphoadenosine 5'-phosphosulfate (PAPS), the sulfate donor for the reaction, varied 7-fold, and quantitative Western blot analysis showed variable levels of immunoreactive protein when compared to the wild-type enzyme. Therefore, mechanisms responsible for decreased activity involved both alterations in levels of enzyme protein and alterations in substrate kinetics. In summary, application of a 'genotype to phenotype' strategy has resulted in the identification of a series of functionally significant common genetic polymorphisms for SULT1C1. It will now be possible to evaluate the possible contribution of these polymorphisms to variation in the sulfate conjugation of drugs, other xenobiotics and/or disease pathophysiology.

Sulfotransferase (SULT) enzymes catalyze the sulfate conjugation of drugs, other xenobiotics, neurotransmitters and hormones. The genes for SULT1A1 and SULT1A2 contain common genetic polymorphisms that are associated with individual variations in levels of enzyme activity as well as variations in biochemical and physical properties. We set out to compare the frequencies of common SULT1A1 and SULT1A2 alleles in Caucasian, Chinese and African-American subjects. Allele frequencies for SULT1A1*1, *2 and *3 in 242 Caucasian subjects were 0.656, 0.332 and 0.012, respectively. Frequencies of those same alleles were significantly different in 290 Chinese subjects: 0.914, 0.080 and 0.006, respectively, as were frequencies in 70 African-American subjects: 0.477, 0.294 and 0.229, respectively. Ethnic variation in allele frequencies was also observed for SULT1A2, with frequencies in Caucasian subjects for SULT1A2*1, *2 and *3 of 0.507, 0.389 and 0.104; frequencies in Chinese of 0.924 and 0.076 with no *3 alleles observed; and, finally, in African-Americans frequencies of 0.637, 0.249 and 0.114, respectively. We also found that SULT1A1*2 and SULT1A2*2, the most common variant alleles for these two genes, were in positive linkage disequilibrium in all three populations studied, with D' values of 0.776 in Caucasian (P < 0.001), 0.915 in Chinese (P < 0.001) and 0.864 in African-American subjects (P < 0.001). These observations represent a step towards determining the possible functional implications for individual variations in sulfate conjugation of common genetic polymorphisms for SULT1A1 and SULT1A2.

Histamine is involved in the pathophysiology of asthma, and histamine N-methyltransferase (HNMT) plays the dominant role in histamine metabolism in human bronchial epithelium. Levels of HNMT activity in human tissues are controlled, in part, by inheritance. A common C314T polymorphism within the HNMT gene results in a Thr105Ile change in encoded amino acid, and the T314 allele is associated with decreased levels of both HNMT enzymatic activity and immunoreactive protein. Therefore, presence of the T314 allele would be expected to result in reduced histamine metabolism and increased bronchoconstriction. We characterized this common, functionally significant polymorphism in DNA samples from 237 randomly selected Caucasian control subjects and 192 samples from Caucasian asthmatic patients. Allele frequencies for the T314 HNMT allele were 0.08 in the control samples and 0.14 in samples from Caucasian asthmatic patients (odds ratio = 1.9, P < 0.01), indicating a significant increase in the frequency of subjects with low HNMT activity among asthmatics. The association between a common, functionally significant genetic polymorphism for HNMT and asthma suggests that individual variation in histamine metabolism might contribute to the pathophysiology and/or response to therapy of this disease.

Nicotinamide N-methyltransferase (NNMT) catalyses the N-methylation of nicotinamide and structurally related pyridines. NNMT enzymatic activity in human liver varies over a five-fold range with a bimodal frequency distribution - raising the possibility of regulation by a genetic polymorphism. We set out to characterize molecular genetic mechanisms that might be involved in the regulation of individual variation in human liver NNMT activity. After Northern blot analysis confirmed that NNMT is highly expressed in the liver, eight human hepatic biopsy samples, four each with 'low' or 'high' levels of activity, were used to perform quantitative Western blot analysis. There was a highly significant correlation (r(s) = 0.96, P < 0.0001) between NNMT activity and immunoreactive protein in these samples. We next determined that a potent promoter was located within the initial 700 bp of the 5'-flanking region of the human NNMT gene. That gene consists of 3 exons, with an initial 1240 bp intron and a second intron that is approximately 14 kb in length. We subsequently isolated DNA from 27 human liver biopsy samples with low, intermediate or high levels of NNMT activity. The three exons, all 1240 bp of intron 1 and approximately 700 bp of the 5'-flanking region of the NNMT gene were amplified from each of these samples with the polymerase chain reaction, followed by DNA sequencing to identify genetic polymorphisms that might correlate with 'NNMT phenotype'. No single nucleotide polymorphisms (SNPs) or insertion/deletion events were detected within either the exons or 5'-flanking regions of NNMT for these 27 samples. Although there were eight SNPs within intron 1, none were systematically related to level of NNMT activity. These results indicate that the exons and 5'-flanking region of the NNMT gene display little or no sequence variation. Therefore, polymorphisms within these areas of the gene are unlikely to be related to wide individual variations in the level of this enzyme activity in the human liver.

Sulfate conjugation catalysed by phenol sulfotransferase (PST) is an important pathway in the metabolism of many drugs. Two isoforms of PST have been characterized biochemically in human tissues-a thermostable (TS), or phenol-metabolizing (P) and a thermolabile (TL), or monoamine-metabolizing (M) form. Pharmacogenetic studies of TS and TL PST activities in the human blood platelet showed that the activities of these two isoforms were regulated by separate genetic polymorphisms. Subsequently, a series of TS PST cDNAs were cloned, and, based on sequence homology, those cDNAs could be classified as members of two separate subgroups, designated here as 'TS PST1' and 'TS PST2'-indicating the existence of three rather than two PST isoforms; TS PST1, TS PST2 and TL PST. The genes encoding TS PST2, STP2, and TL PST, STM, have been cloned, structurally characterized and mapped to chromosome 16-the same chromosome on which the TS PST1 gene, STP1, is localized. As a step toward molecular pharmacogenetic studies of sulfate conjugation in humans, we set out to clone and structurally characterize STP1, the remaining uncharacterized human PST gene. We found that STP1 spanned approximately 4.4 kb and contained 9 exons. The first two exons, IA and IB, were identified by performing 5'-rapid amplification of cDNA ends (RACE) with human liver cDNA as template. Exons IA and IB were noncoding and represented two different cDNA 5'-untranslated region sequences. No canonical TATA box sequences were present within the 5'-flanking regions of the gene, i.e. regions flanking exons IA and IB. Finally, use of the long polymerase chain reaction made it possible to determine that STP1 is located approximately 45 kb 5'-upstream from STP2 on the short arm of human chromosome 16. Cloning and structural characterization of STP1, when combined with knowledge of the structures of STP2 and STM, will make it possible to study the molecular basis for the genetic regulation of PST activity in human tissue.

Catechol-O-methyltransferase (COMT) inactivates catecholamines and catechol drugs such as L-DOPA. A common genetic polymorphism in humans is associated with a three-to-four-fold variation in COMT enzyme activity and is also associated with individual variation in COMT thermal instability. We now show that this is due to G-->A transition at codon 158 of the COMT gene that results in a valine to methionine substitution. The two alleles can be identified with a PCR-based restriction fragment length polymorphism analysis using the restriction enzyme Nla III. The identification of a gentic marker associated with significant alterations in enzyme activity will facilitate the analysis of a possible role for the COMT gene in neuropsychiatric conditions in which abnormalities in catecholamine neurotransmission are believed to occur, including mood disorders, schizophrenia, obsessive compulsive disorder, alcohol and substance abuse, and attention deficit hyperactivity disorder. In addition, this polymorphism may have pharmacogenetic significance in that it will help make it possible to identify patients who display altered metabolism of catechol drugs.

Nicotinamide N-methyltransferase (NNMT) catalyses the N-methylation of nicotinamide and other pyridines. Human liver NNMT activity shows large individual variations and a bimodal frequency distribution, raising the possibility that this activity, like those of many other methyltransferase enzymes, might be regulated by a genetic polymorphism. In an attempt to develop an experimental animal model for pharmacogenetic studies of NNMT, we determined optimal conditions for the measurement of hepatic NNMT activity in C57BL/6J mice. Mouse liver NNMT was a cytoplasmic enzyme with a pH optimum of 7.4 and apparent Km values for nicotinamide and S-adenosyl-L-methionine, cosubstrates for the reaction, of 370 and 6.5 microM, respectively. These properties were very similar to those of human liver NNMT, as was the relative sensitivity of the mouse liver enzyme to a series of methyltransferase inhibitors. Hepatic NNMT activity was then measured in tissue from male mice of 10 inbred strains. Average levels of NNMT activity in these strains varied by up to 14-fold and ranged from 1.13 +/- 0.18 U per mg protein (mean +/- SEM, n = 6) for C3H/HeJ mice to 16.0 +/- 1.16 U per mg protein in C57BR/cdJ animals. Average hepatic NNMT activities in female mice of six strains in which both sexes were studied varied from five-fold higher than those in males for "low activity' strains, to not significantly different for "high activity' strains. A series of properties of NNMT was then compared in hepatic cytosol from male mice of three different strains - one with "low' (C3H/HeJ), one with "intermediate' (DBA/2J), and one with "high' (C57BL/6J) hepatic NNMT activity. There were no striking differences among these three strains in hepatic NNMT pH optimum, substrate kinetics, IC50 values for inhibitors, thermal stability or behavior during ion exchange chromatography. The existence of large strain and gender-dependent variation in hepatic NNMT activity will make it possible to use inbred mice for studies of the role of inheritance and gender in the regulation of NNMT activity in this species, as well as for studies of the potential pharmacological and toxicological consequences of variation in this important drug-metabolizing enzyme activity.

The level of catechol O-methyltransferase (COMT) activity and COMT thermal stability in human tissue are controlled by a common genetic polymorphism. We studied individual hepatic biopsy samples shown previously to have phenotypically high, low or intermediate COMT activities and thermal stabilities to test the hypothesis that the molecular mass (M(r)) and/or isoelectric point (pI) of the enzyme might differ in tissue from subjects with different presumed genotypes for the COMT genetic polymorphism. COMT was partially purified from each hepatic tissue sample by sequential ion exchange and gel filtration chromatography, and photoaffinity labelling was performed with [3H-methyl]-S-adenosyl-L-methionine ([3H-methyl]-Ado-Met), the methyl donor for the COMT enzymatic reaction. Two-dimensional sodium dodecylsulfate polyacrylamide gel electrophoresis (2-D SDS-PAGE) analysis of individual samples consistently showed the presence of three [3H-methyl]-Ado-Met photoaffinity labelled proteins with pI values of 5.4, 5.5 and 5.7, all three of which had M(r) values of approximately 27.1 kDa. The same pattern was observed in all samples irrespective of COMT phenotype. Western blot analysis of 2-D SDS-PAGE gels performed with rabbit polyclonal antibodies to partially purified human kidney COMT showed a pattern similar to that found during photoaffinity labelling. Once again, the same pattern was found in all samples irrespective of COMT phenotype. Therefore, neither photoaffinity labelling nor Western blot analysis revealed differences in either M(r) or pI of cytoplasmic COMT in hepatic tissue from subjects selected on the basis of different phenotypic expression of the COMT genetic polymorphism.

Thiopurine methyltransferase (TPMT) catalyses the S-methylation of thiopurine drugs such as 6-mercaptopurine (6-MP). TPMT activity in the human red blood cell (RBC) is controlled by a common genetic polymorphism. Gene frequencies for this polymorphism are such that approximately one in 300 subjects is homozygous for the allele for low activity and lacks RBC TPMT activity, 11% of subjects are heterozygous and have intermediate levels of enzyme activity and 89% are homozygous for the allele for high activity. Our experiments were performed to determine whether the properties of TPMT in an important human drug metabolizing organ, the liver, were similar to those of RBC TPMT and to test the hypothesis that the genetic polymorphism which controls TPMT activity in the human RBC might also regulate the level of this enzyme activity in hepatic tissue. Human liver TPMT is a cytoplasmic enzyme and the Km values for 6-MP and S-adenosyl-L-methionine, cosubstrates for the reaction, were 580 microM and 2.7 microM, respectively. These properties, as well as the sensitivity of human liver TPMT to a panel of methyltransferase inhibitors, were similar to those of RBC TPMT. The enzyme activity was then measured in 119 surgical biopsy samples of hepatic tissue. Average hepatic TPMT activity was 13.6% higher in samples from male than in those from female patients. Frequency distribution histograms demonstrated the presence of a subgroup with intermediate enzyme activity that included 8.4% of samples. In addition, when TPMT activity was measured in both RBCs and hepatic tissue for 35 patients, those with inherited intermediate levels of RBC TPMT activity also had intermediate hepatic enzyme activity. Finally, ion exchange chromatography demonstrated the presence of two isozymes of TPMT in human hepatic tissue, but the isozymes did not appear to explain the molecular mechanism responsible for the genetic polymorphism. These results were compatible with the conclusion that the genetic polymorphism which controls TPMT activity in the RBC also controls levels of this important enzyme activity in a major human drug metabolizing organ, the liver.