![[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}
|
|
|
|
|
||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
ABSORPTION, DISTRIBUTION, METABOLISM, AND EXCRETION
Departments of Molecular Pharmacology and Experimental Therapeutics (O.E.S., C.Y., B.A.T., R.M.W.) and Health Sciences Research (V.S.P.), Mayo Medical School-Mayo Clinic-Mayo Foundation, Rochester, Minnesota; and Department of Medical Sciences (L.B.K., L.A.T.), School of Veterinary Medicine, University of Wisconsin, Madison, Wisconsin
Received August 23, 2003; accepted October 24, 2003.
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
; Lennard, 1992; White et al., 2000). In humans, these drugs are used as cytotoxic agents to treat neoplasia and as immune suppressants (Paterson and Tidd, 1975; Lennard, 1992), with a similar spectrum of therapeutic applications in companion animals (Beale, 1988). However, thiopurine drugs have a relatively narrow therapeutic index and are capable of causing life-threatening toxicity, most often myelosuppression (Paterson et al., 1975; Lennard, 1992). Thiopurines also provide an excellent example of the successful use of pharmacogenetics to individualize human drug therapy (Weinshilboum et al., 1999; Weinshilboum, 2001). That is true because a genetic polymorphism for thiopurine S-methyl-transferase (TPMT) (Weinshilboum and Sladek, 1980), an enzyme that catalyzes the S-methylation of these drugs (Remy, 1963; Woodson and Weinshilboum, 1983), plays an important role in individual variation in their toxicity and therapeutic efficacy (Weinshilboum et al., 1999; Weinshilboum, 2001).
In the present study, we set out to test the hypothesis that the most popular companion animal, the domestic cat, might also display inherited variation in TPMT activity. As a first step, TPMT activity was measured in the cat red blood cell (RBC), an easily accessible cell that has been widely used in human TPMT pharmacogenetic studies (Weinshilboum and Sladek, 1980; Weinshilboum et al., 1999; Yan et al., 2000). There were large individual variations in level of feline RBC TPMT activity, but the average level of activity was much lower than that in humans or dogs, the two other species in which large population studies have been conducted (Weinshilboum and Sladek, 1980; Yan et al., 2000; Salavaggione et al., 2002). The cat TPMT cDNA and gene were then cloned and characterized and that information was used to perform a genotype-phenotype correlation analysis by resequencing the feline TPMT gene using DNA from animals selected for extreme RBC TPMT phenotype. The GenBank accession number for the cat TPMT cDNA is AY324659
[GenBank]
, and those for the cat TPMT gene are AY324660
[GenBank]
to AY324667
[GenBank]
. A total of 31 SNPs were identified, and 12 "informative" polymorphisms were then assayed in all animals for which DNA was available. Those 12 SNPs were associated with approximately 30% of TPMT phenotypic variance. Finally, functional genomic studies were conducted with all of the feline TPMT allozymes identified, i.e., all variant cat TPMT amino acid sequences. Five of the eight variant allozymes displayed reduced levels of both enzyme activity and immunoreactive protein compared with the "wild-type" sequence. These results enhance our understanding of TPMT comparative genomics. They also represent a step toward the application of pharmacogenetic principles to companion animal drug therapy.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
). Hematocrits of the washed RBCs were determined before cell lysis to make it possible to express enzyme activity per milliliter of packed RBCs. Buffy coats were also prepared for use in DNA isolation.
TPMT Enzyme Assay and Substrate Kinetic Studies. TPMT activity was measured with the radiochemical method of Weinshilboum et al. (1978) as modified by Szumlanski et al. (1992) to omit the Chelex-100 step. Other modifications included an extension of the incubation time for the RBC lysates to 2 h and a decrease of the reaction pH to 6.5 from the 7.5 value used to assay human RBC TPMT activity (Weinshilboum et al., 1978). One unit of TPMT activity represented the formation of 1 nmol of 6-methylmercaptopurine per hour of incubation. This same enzyme assay was also used to perform substrate kinetic studies with a pooled cat RBC lysate and, with an incubation time of 30 min, with recombinant cat TPMT allozymes. For the substrate kinetic experiments, eight 6-MP concentrations that varied from 60 µM to 7.5 mM and eight AdoMet concentrations that ranged from 0.19 to 24.2 µM were studied.
Cat TPMT cDNA and Gene Cloning. The feline TPMT cDNA was cloned by use of the PCR and a cross-species sequence homology-based strategy. Because TPMT is highly expressed in the kidney, the template was cDNA synthesized using mRNA isolated from a cat CCL-94 renal cell line (American Type Culture Collection, Manassas, VA). Sequences of the primers used to perform this amplification, as well as the sequences of all primers used in subsequent experiments, are listed in Table 1. The numbering scheme used for primers and the locations of polymorphic nucleotides is described in the legend for Table 1. Specifically, 5'- and 3'-rapid amplification of cDNA ends (RACE) (Frohman et al., 1988) were used to obtain partial cat TPMT cDNA sequences, and the complete cDNA open reading frame (ORF) was then amplified using primers that hybridized to the cDNA 5'- and 3'-untranslated regions (UTRs). Knowledge of the cat cDNA sequence was then used to clone and characterize the feline TPMT gene.
|
The cat TPMT gene structure was partially determined by use of the PCR and exon-specific primers with cat genomic DNA as template. Those experiments were performed using DNA isolated from buffy coat preparations obtained from 10 cats with the QIAamp blood kit (QIAGEN, Chatsworth, CA). Buffy coat DNA was also used in subsequent experiments in which TPMT genotypes were determined for individual cats. However, amplifications performed with genomic DNA were only partially successful in characterizing the gene structure. Therefore, the cat TPMT cDNA was used to probe a cat genomic DNA BAC library created by Dr. P. J. de Jong at the Oakland Children's Hospital. Three positive clones were identified (BACs 523M14, 457C19, and 398H2) and, on the basis of PCR amplification, all three seemed to include the full TPMT gene. BAC 523M14 was then selected to characterize the TPMT gene structure by direct sequencing using TPMT cDNA ORF-specific sequencing primers. Use of this approach made it possible to characterize exon-intron splice junctions and to obtain intron sequence that flanked each exon. That information, in turn, was used to design intron-based primers that were used both to resequence the cat gene and to verify partial intron sequences determined initially by exon-to-exon amplifications. The cat TPMT gene structure was also compared with those of the human, dog, rat, and mouse genes. Specifically, we made comparisons with the dog gene structure described by Salavaggione et al. (2002) and the human TPMT gene structure described by Szumlanski et al. (1996). The mouse and rat TPMT gene structures were obtained by performing database searches with mouse and rat TPMT cDNA sequences having GenBank accession numbers AF046887
[GenBank]
and AF120100
[GenBank]
, respectively.
Cat TPMT Polymorphism Detection. DNA isolated from blood obtained from 24 cats selected for either high or low levels of RBC TPMT activity was used to resequence the cat TPMT gene to perform genotype-phenotype correlation analysis. DNA sequencing was performed in the Mayo Molecular Biology Core Facility with ABI 377 DNA sequencers using BigDye (Applied Biosystems, Foster City, CA) dye primer sequencing chemistry. Dye primer sequencing was used to enhance our ability to detect heterozygosity (Chadwick et al., 1996). All resequencing primers contained 18 additional M13 nucleotides at their 5' ends to make it possible to use dye-primer sequencing chemistry (Table 1). After DNA sequence analysis had been completed for these initial 24 cats, restriction fragment length polymorphism (RFLP) assays were designed for SNPs found to be associated with variations in level of RBC TPMT activity, with the exception of SNPs located in exons 2, 5, and 6. Those polymorphisms were assayed by the use of dye primer DNA sequencing. Specifically, the RFLP assay for the nucleotide (–30) polymorphism in intron 3 involved digestion of the variant sequence with ApoI (New England Biolabs, Beverly, MA). NlaIII (New England Biolabs) digested the variant sequence for the exon 4 nucleotide 337 polymorphism. The wild-type sequence for the intron 8 SNP at nucleotide 75 was digested by DraI (New England Biolabs), and AccI (New England Biolabs) digested the variant sequence for the nucleotide 698 polymorphism in exon 9. Primers for the exon 8 and 9 amplifications were the same as those used during the resequencing experiments. All RFLP assays included controls for both genotypes being analyzed. After restriction digestion, samples were analyzed by electrophoresis with a 3% agarose gel (Invitrogen, Carlsbad, CA).
Site-Directed Mutagenesis and Transient Expression. Sitedirected mutagenesis was then performed using a "circular PCR" approach to create expression constructs for eight variant allozymes (five observed and three inferred) that resulted from the five nonsynonymous cSNPs at cDNA nucleotides 20, 23, 43, 337, and 698 (see Table 1 for primer sequences). These ORF sequences were then cloned into the eukaryotic expression vector p91023(b) (Kaufman, 1985), and the insert was sequenced in both directions to ensure that only the desired sequence change had been introduced during amplification. These expression constructs were then used to transfect COS-1 cells as described previously (Honchel et al., 1993). The cells were also cotransfected with pSV-
-galactosidase DNA (Promega, Madison, WI) to make it possible to correct for transfection efficiency.
TPMT Antibodies and Western Blot Analysis. A peptide corresponding to cat TPMT amino acids 40–59, with an additional cysteine residue at the amino terminus, was synthesized for use in the generation of antibodies. This polypeptide sequence had been compared with those in GenBank, the EMBL database, and the SWISS-PROT protein sequence database to avoid use of a peptide with high homology to non-TPMT proteins. The synthetic peptide was conjugated to keyhole limpet hemocyanin and was used to generate rabbit polyclonal antibodies (Cocalico Biologicals, Inc., Reamstown, PA). Those antibodies were used to perform quantitative Western blot analyses with recombinant cat TPMT allozymes expressed in COS-1 cells. In the course of those experiments, the quantity of COS-1 cytosol loaded on a 12% acrylamide gel was adjusted on the basis of -galactosidase activity to correct for transfection efficiency. Bound antibody was detected by chemiluminescence using the ECL Western blotting system (Amersham Biosciences Inc., Piscataway, NJ). The AMBIS Radioanalytic Imaging System, Quant Probe version 4.31 (Ambis, Inc., San Diego, CA), was used to quantitate immunoreactive protein, and the data were expressed as a percentage of the intensity of the wild type cat TPMT band on the same gel.
Data Analysis. DNA sequences obtained during the gene resequencing studies were analyzed with the PolyPhred 3.0 (Nickerson et al., 1997) and Consed 8.0 (Gordon et al., 1998) programs. The University of Wisconsin GCG software package, version 10, was also used to analyze nucleotide sequence data. Apparent Km values were calculated with the method of Wilkinson (1961) using a computer program written by Cleland (1963). Points that deviated from linearity on double inverse plots as a result of substrate inhibition were not included in those calculations. Specifically, four concentrations of 6-MP, 0.24 to 1.9 mM, and four concentrations of AdoMet from 1.5 to 12.1 µM were used to calculate the final apparent Km values. Analysis of variance (ANOVA) was used to compare TPMT activity levels among female, male, spayed, and neutered animals. Correlation analyses were performed by calculating Spearman's rank correlation coefficients. TPMT activity levels in healthy animals and cats with minor or chronic health complaints were compared by the use of Student's t test. The association between polymorphisms and RBC TPMT activity level was determined by the use of ANOVA, and R2 values from the ANOVA models were used to summarize the proportion of variability in TPMT activity that could be explained by the polymorphisms. Linkage disequilibrium was explored by computing D' values for all possible pairwise combinations of SNPs. D' is a measure of linkage disequilibrium that is independent of polymorphism frequency (Hartl and Clark, 1997; Hedrick, 2000). Haplotype analyses were performed by the use of a program based on the E-M algorithm (Excoffier and Slatkin., 1995; Long et al., 1995; Schaid et al., 2002).
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
; Yan et al., 2000). RBC TPMT activity was then measured in blood from cats that presented to a veterinary medical teaching hospital. After observing large individual variations in level of feline RBC TPMT activity, the cat TPMT cDNA and gene were cloned and genotype-phenotype correlation analysis was performed by resequencing exons and splice junctions of the cat TPMT gene using DNA from animals with high or low levels of RBC enzyme activity. Thirty-one TPMT polymorphisms were present in these selected samples, and 12 of those SNPs were associated with approximately 30% of the total variance in level of enzyme activity when assayed in all 89 animals for which DNA was available. Finally, functional genomic studies were performed for all cat TPMT variant allozymes observed during the gene resequencing studies. Those experiments demonstrated that five of the eight variant allozymes showed decreases in levels of both enzyme activity and quantity of immunoreactive protein.
TPMT Assay and Activity. Assay conditions for cat RBC TPMT activity were optimized for enzyme quantity, time of incubation, pH, and concentrations of the two cosubstrates for TPMT: 6-MP and AdoMet. The final assay used 100 µl of RBC lysate and a 2-h incubation. This incubation time was necessary because of the relatively low level RBC TPMT activity in the cat compared with that present in human RBCs. Both the quantity of RBC lysate and incubation time were within the linear range (data not shown). The pH optimum for TPMT in cat RBC lysates was 6.5, different from the optimal value of 7.5 for human RBC TPMT (Weinshilboum et al., 1978). Apparent Km values of the enzyme for the two cosubstrates for the reaction, 6-MP and AdoMet, determined with a pooled RBC lysate as an enzyme source, were 800 and 2.4 µM, respectively, values very similar to those that will be described subsequently for recombinant cat TPMT. These figures can be compared with reported apparent Km values of 320 and 1.7 µM for human RBC lysate TPMT (Weinshilboum et al., 1978) and 710 and 19.1 µM, respectively, for RBC lysate TPMT in the dog (Salavaggione et al., 2002).
These optimal assay conditions were then used to measure RBC TPMT activity in blood samples from 104 cats seen at the University of Wisconsin-Madison Veterinary Medical Teaching Hospital (Fig. 1). Cat RBC TPMT activity varied from 1.36 to 13.0 units/ml packed RBCs, with a mean ± S.D. of 4.47 ± 1.82 units/ml RBCs, for a 9.8-fold variation within ± 2 S.D. of the mean. This mean value was only about 20% of those that have been reported for human or dog RBC TPMT (Weinshilboum and Sladek, 1980; Yan et al., 2000; Salavaggione et al., 2002). Cat RBC TPMT activity did not differ significantly among male and female, spayed or neutered cats (Table 2). Furthermore, level of RBC TPMT activity was not significantly correlated with age in the 75 animals for which age was known (rs = –0.072; p = 0.54) (Fig. 2). Clinical information was available for 85 of these animals. Eighteen suffered from disease, and the remaining 67 cats were considered "healthy". There was not a significant difference between these two groups in average level of RBC TPMT activity (data not shown). Information on "breed" was available for 99 of the animals studied, but 96 of the 99 were identified only as "mixed breed domestic short hair" cats. The presence of large individual variations in level of cat RBC TPMT activity that were not related to age or sex raised the possibility that genetic factors might contribute to this variation. Therefore, the next series of experiments was designed to determine whether inheritance, involving the cat TPMT gene, might be one factor responsible for the variation, as it is in both humans and dogs (Weinshilboum and Sladek, 1980; Weinshilboum, 2001; Salavaggione et al., 2002). To test that hypothesis, we first had to clone and characterize the cat TPMT cDNA and gene.
|
|
|
Cat TPMT cDNA and Gene Cloning. The cat TPMT cDNA was cloned by using a cross-species sequence homology-based strategy performed with cDNA synthesized from mRNA isolated from a cat renal cortical cell line. The cat TPMT cDNA ORF was 735 bp in length and encoded a protein 82.4% identical to the amino acid sequence encoded by the most common TPMT allele in humans (Honchel et al., 1993; Szumlanski et al., 1996). Seventy-six bp of 5'-UTR and 30 bp of cat TPMT cDNA 3'-UTR sequence were obtained by performing 5'- and 3'-RACE. The cDNA was then used to clone and characterize the cat TPMT gene by using both a PCR-based strategy with cat genomic DNA as template and by directly sequencing a cat genomic DNA BAC clone isolated by probing a cat genomic DNA BAC library with the cDNA. A comparison of the domestic cat TPMT gene structure with those for humans, dogs, mice, and rats is shown in Fig. 3. The lengths of internal exons that encoded protein and the locations of splice junctions within the ORF were identical in these five species (Fig. 3). Because humans have a processed pseudogene for TPMT (Lee et al., 1995), we tested the hypothesis that the cat might also have a processed pseudogene by performing PCR amplifications using three forward primers that hybridized with cat exons 3 and 4 paired with three reverse primers for cat exons 8 and 9. No amplicons were produced during those amplifications, thus failing to provide evidence for a TPMT processed pseudogenes in the cat. Obviously, we cannot eliminate the possible existence of feline-processed pseudogenes lacking adequate sequence homology to hybridize with these six primers.
|
Cat TPMT Genotype-Phenotype Correlation Analysis. Knowledge of the cat TPMT gene structure and sequence, including knowledge of intron sequences that flanked each exon, was then used to perform a genotypephenotype correlation analysis. The first step in that process involved resequencing the cat gene using DNA samples from animals selected to have either high or low levels of RBC TPMT activity. Specifically, 24 DNA samples, 12 from animals with low activity (2.2 ± 0.44 units/ml RBCs, mean ± S.D.) and 12 from animals with high activity (7.8 ± 2.1), were selected for inclusion in this phase of the analysis. Each exon encoding protein, plus approximately 100 bp of the flanking introns, was amplified using these DNA samples as template, and amplicons were sequenced using dye primer chemistry. Thirty-one SNPs were observed in these 24 samples, including five nonsynonymous cSNPs (Table 3). No insertions or deletions were observed. Twenty-three of the SNPs had allele frequencies equal to or greater than 10% in these 24 DNA samples, including four of the five nonsynonymous cSNPs. All SNPs with a frequency adequate for analysis (>10%) were in Hardy-Weinberg equilibrium. We also determined "nucleotide diversity", a quantitative measure of genetic variation, adjusted for the number of alleles studied. Two standard measures of nucleotide diversity are 
, average heterozygosity per site, and 
, a population mutation measure theoretically equal to the neutral mutation parameter (Fullerton et al., 2000). In the 24 cat DNA samples that had been completely resequenced, was 0.27 ± 0.14 x 10–4, whereas was 6.99 ± 2.27 x 10–4. These values can be compared with and values of 0.30 ± 0.18 x 10–4 and 1.83 ± 0.75 x 10–4, respectively, for 39 dog DNA samples that were resequenced for the TPMT gene as well as 0.15 ± 0.09 x 10–4 and 2.77 ± 0.91 x 10–4 for 90 human DNA samples in which TPMT was resequenced (www.PharmGKB). It can be seen that, on the basis of values, cats seemed to have a higher degree of nucleotide diversity than either humans or dogs, with a p value for a species-dependent difference of <0.004 in both cases.
|
We next analyzed the possible association of cat TPMT polymorphisms with level of RBC TPMT activity. That analysis demonstrated that 12 of the 31 SNPs were associated with 56% of phenotypic variance in these 24 samples (Table 3). A previous analysis of human RBC TPMT population data showed that the common polymorphism for the trait of level of RBC TPMT in humans accounted for approximately 60–66% of the total variance (Vuchetich et al., 1995). A combination of RFLP analysis and dye primer DNA sequencing was then used to assay these 12 informative SNPs in DNA samples from all 89 of the 104 animals for which DNA was available (Table 3). Data from these 89 cats were then used to calculate the proportion of variance in level of RBC TPMT activity that was associated with these 12 polymorphisms in the larger population sample. A total of 30% of the variance could be explained (Table 3). Because several of the SNPs were tightly linked, as described in subsequent paragraphs, values for the proportion of variance associated with individual polymorphisms listed in Table 3 are not additive, e.g., the information listed for the intron 8 (75) SNP and the exon 9 (698) SNP is "redundant". Unfortunately, the relatively high nucleotide diversity present in the cat TPMT gene and, therefore, the relatively large number of polymorphisms resulted in such a large number of haplotypes, as described subsequently, that it proved difficult to associate individual haplotypes with level of activity in a statistically meaningful way.
We next computed pairwise linkage disequilibrium for the 12 informative SNPs in these 89 DNA samples by calculating D' values (Table 4). D' values can range from (+1) when two polymorphisms are maximally linked to (–1) when they never occur together (Hartl and Clark, 1997; Hedrick, 2000). This analysis showed, for example, that the nonsynonymous cSNPs at exon 9 ORF nucleotide 698 was in positive linkage disequilibrium with the intron 8 (75) SNP, with a D' value of 1.0, as was mentioned previously (Table 4). Because we had observed 31 polymorphisms, there were theoretically up to (230) possible haplotypes for the cat TPMT gene, too many to analyze with even the relatively sophisticated programs and hardware available to us. The same problem occurred when only the 12 SNPs that were assayed in all animals were included in the analysis, because of their relatively high frequencies, and also because of linkage disequilibrium. When only the nonsynonymous cSNPs were considered, there were six unambiguous and three inferred allozymes, for a total of eight variant amino acid sequences (Table 5). That information was of practical importance for the recombinant allozyme expression studies that will be described subsequently. We were able to determine complete haplotypes based on polymorphism data for all 12 of the SNPs assayed for the two most common allozymes. There were two unambiguous and four inferred haplotypes for each of these allozymes with, not surprisingly, the unambiguous haplotypes being those with the highest frequencies (Table 6). The final series of experiments involved functional genomic studies of the variant allozymes encoded by the five nonsynonymous cSNPs.
|
|
|
Cat TPMT Allozyme Expression in COS-1 Cells. Expression constructs were created for the nine cat TPMT allozymes listed in Table 5 (six observed and three inferred amino acid sequences), and those constructs were used to transfect COS-1 cells. Mammalian cells were used to perform these experiments to ensure appropriate post-translational modification as well as the presence of mammalian systems for protein degradation. Alteration in only a single amino acid as a result of a genetic polymorphism has been found to have significant functional implications for many enzymes, including human TPMT, most often as a result of a decrease in the quantity of protein (Szumlanski et al., 1996; Preuss et al., 1998; Thomae et al., 2002, 2003; Adjei et al., 2003; Shield et al., 2004). After expression in COS-1 cells, five of the recombinant cat allozymes had decreased levels of TPMT activity compared with the wild-type allozyme (Fig. 4A; Table 7). The values shown in Fig. 4A and Table 7 are averages of 12 independent transfections for each allozyme and all values have been corrected for transfection efficiency.
|
|
One possible explanation for these decreases in level of activity would involve a change in substrate kinetics as a result of the alteration in amino acid sequence. Therefore, apparent Km values for the two cosubstrates for the reaction, 6-MP and AdoMet, were determined for all nine allozymes. Some differences in apparent Km values were observed (Table 7). For example, although most of the recombinant allozymes had apparent Km values for 6-MP of approximately 1 mM, three had higher values of 2 to 3 mM. As a result, for those constructs, the basal level of enzyme activity in the transfected COS-1 cell preparation was reassayed using 15 mM 6-MP, twice the concentration used to perform the "standard" assay, to ensure that a saturating concentration of the methyl acceptor substrate was present. Those assays showed that use of the higher 6-MP concentration failed to result in a significant increase in basal level of activity. Therefore, alterations in substrate kinetics were unable to explain the striking differences that we had observed in levels of allozyme activity (Fig. 4A). A common mechanism by which inherited variation in amino acid sequence can result in a lower level of activity is a decrease in the quantity of the encoded protein (Szumlanski et al., 1996; Preuss et al., 1998; Adjei et al., 2003; Shield et al., 2004). Therefore, we next performed quantitative Western blot analysis to measure levels of TPMT immunoreactive protein for these recombinant cat allozymes.
Quantitative Western blot analysis was performed for each of the nine cat TPMT allozymes. Average values for levels of TPMT protein based on five independent transfections for each construct are shown in Fig. 4B and are listed in Table 7. A representative Western blot used to obtain these data is shown in Fig. 4C. Variations in levels of immunoreactive protein for the nine allozymes were significantly correlated with variations in levels of enzyme activity (rs = 0.987, p < 0.001; Fig. 5). These observations were compatible with a growing body of evidence that indicates that a change in only one or two encoded amino acids as a result of common genetic polymorphisms can result in significant alterations in levels of immunoreactive protein, most often a significant decrease (Szumlanski et al., 1996; Preuss et al., 1998; Thomae et al., 2002, 2003; Xu et al., 2002; Adjei et al., 2003; Shield et al., 2004). Because frequencies of alleles encoding the five allozymes with decreased levels of activity and immunoreactive protein were relatively low (Table 3), no samples were either homozygous or compound heterozygotes for these alleles. However, 27 of the 89 samples genotyped were heterozygous for one of the five allozymes that displayed reductions in activity after the transfection of COS-1 cells, but the average level of activity in those 27 samples, 4.28 ± 0.32 units/ml packed RBC (mean ± S.E.M.), was not statistically lower than that in the remaining 62 samples, 4.78 ± 0.24 (p = 0.23).
|
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
; Weinshilboum, 2001). The thiopurine drugs that are metabolized by TPMT are also used to treat companion animals, including cats, and companion animals also display large individual variations in thiopurine efficacy and toxicity (Houston and Taylor, 1991; Rinkhardt and Kruth, 1996). Cats in particular have been reported to be especially sensitive to myelosuppression after azathioprine therapy (Beale et al., 1992), and some authors have even recommended that these drugs not be used to treat cats (White et al., 2000). If principles of individualized therapy based, in part, on pharmacogenetics are of value when applied to humans, it is reasonable to ask whether similar principles might also apply to companion animals. We set out both to test that possibility for the cat and at the same time, expanded our understanding of TPMT comparative genomics.
Average levels of RBC TPMT activity in the cat were significantly lower than those in the two other species that have been studied in a similar manner: humans and dogs (Weinshilboum and Sladek, 1980; Yan et al., 2000; Salavaggione et al., 2002). Two previous studies that included a much smaller number of cats also reported that levels of RBC TPMT activity in this species were only approximately one-fifth of those observed in humans or dogs (Foster et al., 2000; White et al., 2000). If RBC TPMT activity reflects the level of this enzyme activity in other cat tissues, these relatively low levels of activity may help to explain the sensitivity of this species to thiopurine therapy (Beale et al., 1992). We also observed that cats, like humans, displayed large individual variations in level of RBC TPMT activity (Fig. 1). To make it possible to perform genotype-phenotype correlation analysis to study the possible contribution of inheritance to this variation, we cloned the cat TPMT cDNA and gene.
Knowledge of the sequence and structure of the cat TPMT gene made it possible to resequence all coding exons and splice junctions of the cat TPMT gene using DNA from 24 animals selected on the basis of either low or high levels of RBC TPMT activity. Thirty-one SNPs were observed in these 24 DNA samples, including five nonsynonymous cSNPs that resulted in Leu7Ser, Ile8Thr, Asp15Asn, Met113Leu, and Asp233Val alterations in encoded amino acids (Table 3). Twelve of those 31 SNPs were associated, collectively, with 56% of the total variation in level of RBC TPMT activity in these 24 animals (Table 3). When those 12 SNPs were assayed in DNA from all 89 animals for which DNA was available, 30% of the variance in level of RBC TPMT in this population sample was associated with these 12 polymorphisms, polymorphisms present only within exons and splice junctions (Table 3). Obviously, the present observations do not eliminate the possibility of additional functionally significant genetic variation located elsewhere in the gene (e.g., within introns). To place these observations in context, the well defined and functionally well characterized ORF-based polymorphisms for the human TPMT gene are associated with approximately 60 to 66% of the total variation in level of human RBC TPMT activity (Vuchetich et al., 1995). We were also able to identify six unequivocal, as well as three inferred cat TPMT allozymes (Table 5). Expression of all nine of those allozymes showed that five displayed decreases in levels of TPMT enzyme activity measured under optimal conditions, with parallel decreases in levels of immunoreactive protein, and three of the allozymes displayed significant increases in both activity and protein levels (Figs. 4 and 5). Common variant allozymes for TPMT and other genetically polymorphic enzymes in humans have often been associated with decreased levels of enzyme protein (Szumlanski et al., 1996; Preuss et al., 1998; Thomae et al., 2002, 2003; Adjei et al., 2003; Shield et al., 2004), and our observations in the cat serve to emphasize just how common that mechanism is. The decrease in level of protein for the most common TPMT variant allele in humans results from rapid degradation through a ubiquitin-proteasome-mediated process, with the involvement of molecular chaperones such as hsp90 (Tai et al., 1999; Wang et al., 2003). Whether similar mechanisms might apply in other mammalian species remains to be determined. Finally, the nature of cellular mechanisms that result in elevated levels of protein, as seen in three of our samples, remains to be explored.
In summary, we set out to determine whether inheritance might influence RBC TPMT activity level in F. domesticus, the domestic cat, as it does in humans and dogs (Weinshilboum and Sladek, 1980; Yan et al., 2000; Salavaggione et al., 2002). There was a wide range in level of feline RBC TPMT activity (Fig. 1), and the cat TPMT gene seemed to be more polymorphic than those of the other two species that have been studied in depth: humans and dogs. A significant portion of the variation in level of feline RBC TPMT activity was associated with these polymorphisms, although we cannot eliminate associations between the SNPs that we studied and other functionally significant polymorphisms located elsewhere in the cat genome. It is possible that thiopurine drug therapy in the cat, like that in humans, might eventually benefit from the application of pharmacogenetic information to help "individualize" treatment with this class of drugs. As a result, the present studies have not only broadened our understanding of the comparative genomics of an important drug-metabolizing enzyme, but they may also represent a step toward the application of pharmacogenetic principles to an important companion animal species.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
ABBREVIATIONS: 6-MP, 6-mercaptopurine; TPMT, thiopurine S-methyltransferase; RBC, red blood cell; SNP, single nucleotide polymorphism; PCR, polymerase chain reaction; RACE, rapid amplification of cDNA ends; ORF, open reading frame; UTR, untranslated region; RFLP, restriction fragment length polymorphism; ANOVA, analysis of variance; bp, base pair(s).
1 These authors contributed equally to this work. 
Address correspondence to: Dr. Richard Weinshilboum, Department of Molecular Pharmacology and Experimental Therapeutics, Mayo Medical School-Mayo Clinic-Mayo Foundation, 200 First St. SW, Rochester, MN 55905. E-mail: weinshilboum.richard{at}mayo.edu
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Adjei AA, Thomae BA, Prondzinski JL, Eckloff BW, Wieben ED, and Weinshilboum RM (2003) Human estrogen sulfotransferase (SULT1E1) pharmacogenomics: gene resequencing and functional genomics. Br J Pharmacol 139: 1373–1382.[CrossRef][Medline]
Beale KM (1988) Azathioprine for treatment of immune-mediated diseases of dogs and cats. J Am Vet Med Assoc 192: 1316–1318.[Medline]
Beale KM, Altman D, Clemmons RR, and Bolon B (1992) Systemic toxicosis associated with azathioprine administration in domestic cats. Am J Vet Res 53: 1236–1240.[Medline]
Chadwick RB, Conrad MP, McGinnis MD, Johnston-Dow L, Spurgeon SL, and Kronick MN (1996) Heterozygote and mutation detection by direct automated fluorescent DNA sequencing using a mutant Taq DNA polymerase. Biotechniques 20: 676–683.[Medline]
Cleland WW (1963) Computer programmes for processing enzyme kinetic data. Nature (Lond) 198: 463–465.[CrossRef][Medline]
Excoffier L and Slatkin M (1995) Maximum-likelihood estimation of molecular haplotype frequencies in a diploid population. Mol Biol Evol 12: 921–927.[Abstract]
Foster AP, Shaw SE, Duley JA, Shobowale-Bakre EM, and Harbour DA (2000) Demonstration of thiopurine S-methyltransferase activity in the erythrocytes of cats. J Vet Intern Med 14: 552–554.[Medline]
Frohman MA, Dush MK, and Martin GR (1988) Rapid production of full-length cDNAs from rare transcripts: amplification using single gene-specific oligonucleotide primers. Proc Natl Acad Sci USA 85: 8998–9002.
Fullerton SM, Clark AG, Weiss KM, Nickerson DA, Taylor SL, Stengard JH, Salomaa V, Vartiainen E, Perola M, Boerwinkle E, et al. (2000) Apolipoprotein E variation at the sequence haplotype level: implications for the origin and maintenance of a major human polymorphism. Am J Hum Genet 67: 881–900.[CrossRef][Medline]
Gordon D, Abajian C, and Green P (1998) Consed: a graphical tool for sequence finishing. Genome Res 8: 195–202.
Hartl DL and Clark AG (1997) Principles of Population Genetics, 3rd ed., pp 95–107, Sinauer Associates, Inc., Sunderland, MA.
Hedrick PW (2000) Genetics of Populations, 3rd ed., pp 396–405, Jones and Bartlett Publ., Sudbury, MA.
Honchel R, Aksoy I, Szumlanski C, Wood TC, Otterness DM, Wieben ED, and Weinshilboum RM (1993) Human thiopurine methyltransferase: molecular cloning and expression of T84 colon carcinoma cell cDNA. Mol Pharmacol 43: 878–887.[Abstract]
Houston DM and Taylor JA (1991) Acute pancreatitis and bone marrow suppression in a dog given azathioprine. Can Vet J 32: 496–497.[Medline]
Kaufman RJ (1985) Identification of the components necessary for adenovirus translational control and their utilization in cDNA expression vectors. Proc Natl Acad Sci USA 82: 689–693.
Lee D, Szumlanski C, Houtman J, Honchel R, Rojas K, Overhauser J, Wieben ED, and Weinshilboum RM (1995) Thiopurine methyltransferase pharmacogenetics: cloning of human liver cDNA and presence of a processed pseudogene on human chromosome 18q21.1. Drug Met Dispos 23: 398–405.[Abstract]
Lennard L (1992) The clinical pharmacology of 6-mercaptopurine. Eur J Clin Pharmacol 43: 329–339.[CrossRef][Medline]
Long JC, Williams RC, and Urbanek M (1995) An E-M algorithm and testing strategy for multiple-locus haplotypes. Am J Hum Genet 56: 799–810.[Medline]
Nickerson DA, Tobe VO, and Taylor SL (1997) PolyPhred: automating the detection and genotyping of single nucleotide substitutions using fluorescence-based resequencing. Nucleic Acids Res 25: 2745–2751.
Paterson ARP and Tidd DM (1975) 6-Thiopurines, in Antineoplastic and Immunosuppressive Agents II (Sartorelli AC and Johns DG eds) pp 384–403, Springer, New York.
Preuss CV, Wood TC, Szumlanski CL, Raftogianis RB, Otterness DM, Girard B, Scott MC, and Weinshilboum RM (1998) Human histamine N-methyltransferase pharmacogenetics: common genetic polymorphisms that alter activity. Mol Pharmacol 53: 708–717.
Remy CN (1963) Metabolism of thiopyrimidines and thiopurines: S-methylation with S-adenosylmethionine transmethylase and catabolism in mammalian tissue. J Biol Chem 238: 1078–1084.
Rinkhardt NE and Kruth SA (1996) Azathioprine-induced bone marrow toxicity in four dogs. Can Vet J 37: 612–613.[Medline]
Salavaggione OE, Kidd L, Prondzinski JL, Szumlanski CL, Pankratz VS, Wang L, Trepanier L, and Weinshilboum RM (2002) Canine red blood cell thiopurine S-methyltransferase: companion animal pharmacogenetics. Pharmacogenetics 12: 713–724.[Medline]
Schaid DJ, Rowland CM, Tines DE, Jacobson RM, and Poland GA (2002) Score tests for association between traits and haplotypes when linkage phase is ambiguous. Am J Hum Genet 70: 425–434.[CrossRef][Medline]
Shield AJ, Thomae BA, Eckloff BW, Wieben ED, and Weinshilboum RM (2004) Human catechol O-methyltransferase genetic variation: gene resequencing and functional characterization of variant allozymes. Mol Psychiatry, in press.
Szumlanski CL, Honchel R, Scott MC, and Weinshilboum RM (1992) (1992) Human liver thiopurine methyltransferase pharmacogenetics: biochemical properties, liver-erythrocyte correlation and presence of isozymes. Pharmacogenetics 2: 148–159.[Medline]
Szumlanski C, Otterness D, Her C, Lee D, Brandriff B, Kelsell D, Spurr N, Lennard L, Wieben E, and Weinshilboum R (1996) Thiopurine methyltransferase pharmacogenetics: human gene cloning and characterization of a common polymorphism. DNA Cell Biol 15: 17–30.[Medline]
Tai HL, Fessing MY, Bonten EJ, Yanishevsky Y, d'Azzo A, Krynetski EY, and Evans WE (1999) Enhanced proteasomal degradation of mutant human thiopurine S-methyltransferase (TPMT) in mammalian cells: mechanism for TPMT protein deficiency inherited by TPMT*2, TPMT*3A, TPMT*3B or TPMT*3C. Pharmacogenetics 9: 641–650.[Medline]
Thomae BA, Eckloff BW, Freimuth RR, Wieben ED, and Weinshilboum RM (2002) Human sulfotransferase SULT2A1 pharmacogenetics: genotype-to-phenotype studies. Pharmacogenomics J 2: 48–56.[CrossRef][Medline]
Thomae BA, Rifki OF, Theobald MA, Eckloff BW, Wieben ED, and Weinshilboum RM (2003) Human catecholamine sulfotransferase (SULT1A3) pharmacogenetics: common functional genetic polymorphism. J Neurochem 87: 809–819.[CrossRef][Medline]
Vuchetich JP, Weinshilboum RM, and Price RA (1995) Segregation analysis of human red blood cell (RBC) thiopurine methyltransferase (TPMT) activity. Genet Epidemiol 12: 1–11.[CrossRef][Medline]
Wang L, Sullivan W, Toft D, and Weinshilboum R (2003) Thiopurine S-methyltransferase pharmacogenetics: chaperone protein association and allozyme degradation. Pharmacogenetics 13: 555–564.[CrossRef][Medline]
Weinshilboum R (2001) Thiopurine pharmacogenetics: clinical and molecular studies of thiopurine methyltransferase. Drug Met Dispos 29: 601–605.
Weinshilboum RM, Otterness DM, and Szumlanski CL (1999) Methylation pharmacogenetics: catechol O-methyltransferase, thiopurine methyltransferase and histamine N-methyltransferase. Annu Rev Pharmacol Toxicol 39: 19–52.[CrossRef][Medline]
Weinshilboum RM, Raymond FA, and Pazmiño PA (1978) Human erythrocyte thiopurine methyltransferase: radiochemical microassay and biochemical properties. Clin Chim Acta 85: 323–333.[CrossRef][Medline]
Weinshilboum RM and Sladek SL (1980) Mercaptopurine pharmacogenetics: monogenic inheritance of erythrocyte thiopurine methyltransferase activity. Am J Hum Genet 32: 651–662.[Medline]
White SD, Rosychuk RAW, Outerbridge CA, Fieseler KV, Spier S, Ihrke PJ, and Chapman PL (2000) Thiopurine methyltransferase in red blood cells of dogs, cats and horses. Vet Intern Med 14: 499–502.
Wilkinson GN (1961) Statistical estimations in enzyme kinetics. Biochem J 80: 324–332.[Medline]
Woodson LC and Weinshilboum RM (1983) Human kidney thiopurine methyltransferase: purification and biochemical properties. Biochem Pharmacol 32: 819–826.[CrossRef][Medline]
Xu Z-H, Freimuth RR, Eckloff B, Wieben E, and Weinshilboum RM (2002) Human 3'-phosphoadenosine 5'-phosphosulfate synthetase 2 (PAPSS2) pharmacogenetics: gene resequencing, genetic polymorphisms and functional characterization of variant allozymes. Pharmacogenetics 12: 11–21.[CrossRef][Medline]
Yan L, Zhang S, Eiff B, Szumlanski CL, Powers M, O'Brien JF, and Weinshilboum RM (2000) Thiopurine methyltransferase polymorphic tandem repeat: genotype-phenotype correlation analysis. Clin Pharmacol Ther 68: 210–219.[CrossRef][Medline]
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
|
 |
 |
 |
0.7 or 
(-0.7) with p values < 0.001 have been included in the table.
