Altered mercaptopurine metabolism, toxic effects, and dosage requirement in a thiopurine methyltransferase-deficient child with acute lymphocytic leukemia

Thiopurine methyltransferase deficiency, inherited as an autosomal codominant trait, is associated with aberrant mercaptopurine metabolism leading to excessive cellular accumulation of 6-thioguanine nucleotides, the active metabolites of mercaptopurine. We describe a case of severe thiopurine methyltransferase deficiency (activity less than 1 U/8 x 10(8) erythrocytes) in a young girl with acute lymphocytic leukemia. The level of 6-thioguanine nucleotide in the patient's erythrocytes was seven times the population median value, and she had intolerable hematologic toxic effects during postremission therapy with a standard dosage of mercaptopurine (75 mg/m2 per day). Subsequent therapy with 6% of this dosage (10 mg/m2 three times weekly) yielded erythrocytic 6-thioguanine nucleotide concentrations consistently above the population median but not associated with prohibitively toxic effects. This case demonstrates that thiopurine methyltransferase deficiency does not absolutely contraindicate mercaptopurine therapy, and it also provides insight into the mechanism of excessive toxic effects of mercaptopurine sometimes observed in children with acute lymphocytic leukemia.

Analysis of N-methyl-N-nitrosourea-induced mutations in a shuttle vector plasmid propagated in mouse O6-methylguanine-DNA methyltransferase-deficient cells in comparison with proficient cells

A shuttle vector plasmid, pYZ289, was constructed from the pZ189 plasmid and polyoma virus DNA. The plasmid contains a supF gene as a marker of mutation and can replicate in both Escherichia coli and mouse cells. The pYZ289 plasmids treated with N-methyl-N-nitrosourea were passed through mouse cells originating from skin tumors, which are either proficient (HL18) or deficient (HL8) in O6-methylguanine-DNA methyltransferase activity, and mutations in the supF gene were analyzed. In the repair-deficient HL8 cells, N-methyl-N-nitrosourea-treated pYZ289 showed lower plasmid survival and higher mutation frequency than in the repair-proficient HL18 cells. DNA sequence analysis in the mutated supF gene revealed that most mutations occurred in G:C base pairs (86% for HL8, 76% for HL18), and the frequency of G:C----A:T transition was higher in HL8 cells (69%) than in HL18 cells (31%). G:C----T:A transversions occurred more frequently in HL18 cells (31%) than in HL8 cells (12%). Mutations occurred frequently at the base pair positions of 123 and 159 of supF gene in HL18 cells and at 169 in HL8 cells. Analysis of the bases neighboring the mutations appeared to be related to the mutability of the base pairs with the sequence of 5'-purine-G-G-3' being the most frequently mutated. These results show that the new pYZ289 plasmid is useful for the analysis of mutations and that a deficiency in O6-methylguanine-DNA methyltransferase enhances the N-methyl-N-nitrosourea-induced mutation with significant specificity.

Pharmacogenetics of human erythrocyte thiopurine methyltransferase activity in a French population

1. A genetic polymorphism in human erythrocyte thiopurine methyltransferase activity (RBC TPMT) resulting in a trimodal phenotypic distribution has been demonstrated both in a North American population and in British children. 2. We studied whether such a polymorphism may be also present in a white French population by testing RBC TPMT activity in 303 randomly selected blood donors. 3. We found a large inter-individual variation in RBC TPMT activity which ranged from 2 to 40 nmol ml-1 packed RBC h-1, with a mean value of 15.4 +/- 7.0 nmol ml-1 packed RBC h-1. The enzyme activity was not significantly influenced by the sex and age of the subjects. 4. In our population sample, we found no subject with undetectable enzyme activity. However, the probit plot of the log RBC TPMT activity showed a highly significant change in slope at a TPMT activity of 7.5 nmol ml-1 packed RBC h-1. Thirty four subjects (11% of our population) had TPMT activities below 7.5 nmol ml-1 packed RBC h-1. 5. These data are consistent with the view that the genetic polymorphism of TPMT activity described in populations from North America and the United Kingdom is also present in a French population, with about 89% of subjects exhibiting a high activity and 11% an intermediate activity.

Functional methionine synthase deficiency (cblE and cblG): clinical and biochemical heterogeneity

Functional methionine synthase deficiency is generally characterized by homocystinuria and hypomethioninemia in the absence of methylmalonic aciduria. Patients are divided into two classes, cblE and cblG, on the basis of complementation analysis. Presentation has usually been in the first 2 years of life, but one patient came to medical attention at age 21 years with symptoms initially diagnosed as multiple sclerosis. Common findings among 11 patients (4 with cblE and 7 with cblG) have included megaloblastic anemia (all patients) and various neurological deficits including developmental retardation (10 patients), cerebral atrophy (8 patients), hypotonia (7 patients), EEG abnormalities (6 patients), and nystagmus (5 patients). Hypertonia, seizures, blindness, and ataxia were less frequent. All patients have responded to therapy with cobalamin with resolution of anemia and biochemical abnormalities; neurological deficits resolved more slowly and in some cases incompletely. Hydroxycobalamin has been more effective than cyanocobalamin. Fibroblasts from patients with cblE (5 patients) and cblG (6 patients) all showed decreased intracellular levels of methylcobalamin (MeCbl) and decreased incorporation of label from 5-methyltetrahydrofolate into macromolecules, suggesting decreased activity of the MeCbl-dependent enzyme methionine synthase. Methionine synthase specific activity in extracts of all cblE fibroblasts was normal or near-normal under standard reducing conditions; synthase specific activity in extracts of 5 cblG patients was low but was high in a 6th patient measured in another laboratory. Thus, there is heterogeneity among patients with functional methionine synthase deficiency both in clinical presentation and in the results of biochemical studies of cultured cells.

O6-methyltransferase-deficient and -proficient CHO cells differ in their responses to ethyl- and methyl-nitrosourea-induced DNA alkylation

The mutagenic and cytotoxic effects of N-ethyl-N-nitrosourea (ENU) and N-methyl-N-nitrosourea (MNU) were compared in two isogenic Chinese hamster ovary (CHO) cell lines differing for the expression of the repair function for O6-methylguanine (O6-meGua), the O6-methyl-DNA-methyltransferase (MT). Survival and ouabain resistance (ouar) mutation frequency were similar in the two cell lines after treatment with ENU while both effects were strongly reduced in the MT-proficient (MT+) CHO cells after exposure to MNU. The slow repair kinetics of O6-ethylguanine (O6-etGua) when compared to O6-meGua, i.e. 25% versus 88% removal at 20 h after treatment, could still account for the similar mutational curves reported in the two cell lines after ENU treatment. The number of ENU-induced sister chromatid exchanges (SCE) was slightly reduced in the MT+ as compared to MT-deficient CHO cells suggesting a role for O6-etGua in SCE formation. Comparison of survival after exposure to ENU and MNU showed that, at similar levels of O6-alkylguanine on DNA, the ethyl- is more tolerated than the methyl-adduct. These data focus the attention on the importance of DNA damage processing in the cytotoxic response to alkylating agents.

Megaloblastic anemia and immune abnormalities in a patient with methionine synthase deficiency

We report a case of methionine synthase deficiency associated with cellular immune deficiency discovered in a 14-year-old boy. Principal findings were: developmental delay, recurrent upper and lower respiratory tract infections, megaloblastic anemia, discovered at 3 months of age, unresponsive to cyanocobalamin and poorly responsive to folinic acid. Biochemical studies showed: an abnormal deoxyuridine suppression test despite normal serum folate, cobalamin and transcobalamin levels; a normal intracellular uptake of these two coenzymes; and an absolute requirement of methionine for fibroblast growth, suggestive of defective methionine synthesis. An absence of methionine synthase activity in the patient's bone marrow and a profound depression of this activity in lymphocytes and liver were found. Hypergammaglobulinemia with variable lymphopenia, depressed lymphocyte transformation after lectin or recall-antigen stimulation, defective delayed-type hypersensitivity and decreased natural killer activity were noted as well. The patient died at the age of 14.

Vitamin B12-responsive neonatal megaloblastic anemia and homocystinuria with associated reduced methionine synthase activity

We present findings on an infant with neonatal megaloblastic anemia, homocystinuria, and neurologic dysfunction that included developmental delay and tonic seizures. There was no methylmalonic aciduria. Cyanocobalamin therapy was accompanied by complete hematologic and neurologic recovery, diminished homocystine excretion, and subsequently normal neurologic development. Cultured fibroblasts and lymphoblasts showed a reduced methionine synthase activity and a growth requirement for methionine. Cobalamin incorporation by the patient's lymphoblasts was normal, but the proportion of cellular methylcobalamin in the patient's lymphoblasts and fibroblasts were markedly reduced and that of adenosylcobalamin normal. The reduced methionine synthase activity was independent of assay reducing (thiol) conditions, but normal levels of activity accompanied culture of the patient's lymphoblasts in medium with markedly increased cobalamin concentration. The characteristics of the reduced methionine synthase of our patient differ significantly from that of the previously described infant with cobalamin E disease and suggest that genetic heterogeneity may characterize this mutation.

Comparison of folic acid coenzyme distribution patterns in patients with methylenetetrahydrofolate reductase and methionine synthetase deficiencies

Folic acid coenzyme distribution patterns were examined in the liver and kidney of two patients with homocystinuria due to different inborn errors of metabolism affecting the remethylation of homocysteine to methionine. One patient, with severe mental retardation (and death at 3 1/2 yr), had greatly reduced levels of methylenetetrahydrofolic acid (THF) reductase in fibroblasts as well as in liver and kidney. Chromatographic separation of folate coenzymes in liver showed an abnormal pattern with THF as the main component and almost no methyl-THF but total folate was normal. The other patient, who was dystrophic, microcephalic, and had megaloblastic anemia died at age 4 months. He had reduced levels of methionine synthetase in liver and kidney due to a defect of intracellular cobalamin metabolism. Chromatographic analysis of his tissues showed methyl-THF to be the principal folate form and a markedly reduced total folate. These results support the "methyl-THF trap" hypothesis and offer information with respect to the possible therapy of these two disorders.

Incorporation of 5-substituted pyrimidine nucleoside analogues into DNA of a thymidylate synthetase-deficient murine FM3A carcinoma cell line

Various 5-substituted 2'-deoxyuridine (dUrd), 2'-deoxycytidine (dCyd), 1-beta-D-arabinofuranosyluracil (araU) and 1-beta-D-arabinofuranosylcytosine (araC) analogues have been investigated for their stimulatory effect on the growth of a thymidylate (dTMP) synthetase-deficient murine mammary carcinoma cell line (FM3A/TS-) that is auxotrophic for thymidine (dThd). Such stimulatory effect may be considered as indicative for the incorporation of the nucleoside analogue into host cell DNA. Based on this premise, several dUrd analogues were found to be incorporated into FM3A/TS- cell DNA (in decreasing order of incorporation): 5-bromo-dUrd greater than 5-chloro-dUrd greater than 5-(3-hydroxy-1-propynyl)-dUrd greater greater than 5-(1-pentynyl)-dUrd approximately 5-(1-propynyl)-dUrd approximately 5-iodo-dUrd greater than 5-(5-carboxy-1-hexenyl)-dUrd greater than 5-(3,3-dimethyl-1-butynyl)-dUrd greater than 5-ethyl-dUrd greater than 5-(5-chloro-1-pentynyl)-dUrd greater than 5-ethynyl-dUrd approximately 5 vinyl-dUrd greater than 5-phenylethynyl-dUrd greater than 5-(5-cyano-1-pentenyl)-dUrd greater than 5-(1-propenyl)-dUrd greater than 5-(1-hexynyl)-dUrd greater than 5-(5-hexyn-1-enyl)-dUrd. Among the 5-substituted dCyd analogues, 5-methyl-dCyd, 5-chloro-dCyd, 5-bromo-dCyd and 5-iodo-dCyd were also found to stimulate cell growth, and are therefore assumed to be incorporated into FM3A/TS- cell DNA. Since the stimulatory effects of these compounds on FM3A/TS- cell proliferation were suppressed in the presence of a Cyd deaminase inhibitor (tetrahydrouridine) or dCMP deaminase inhibitor (2'-deoxytetrahydrouridine), it is surmised that the dCyd analogues are first deaminated to the corresponding dUrd analogues before they are incorporated into DNA. None of the 5-substituted araU or araC analogues tested were able to sustain the growth of FM3A/TS- cells. It is postulated, therefore, that these araU or araC analogues are not incorporated to any appreciable extent into the DNA of FM3A/TS- cells, or, if they are incorporated, prevent cell growth. Thus, the dTMP synthetase-deficient FM3A/TS- cell line represents a unique system to distinguish those pyrimidine nucleoside analogues that are able to sustain cell growth and, therefore, assumed to be incorporated into the host cell DNA from those pyrimidine nucleoside analogues that are not.

Mutants of Chinese hamster cells deficient in thymidylate synthetase

Stable mutants of Chinese hamster V79 cells deficient in thymidylate synthetase (TS; E.C. 2.1.1.45) have been selected from cultures grown in medium supplemented with folinic acid, aminopterin, and thymidine (FAT). After chemical mutagenesis, the frequency of colonies resistant to the "FAT" medium increased more than 100-fold over the spontaneous frequency. The optimal expression time of the mutant phenotype was 5-7 days after mutagen treatment. The recovery of FAT-resistant colonies in the selective medium was not affected by the presence of wild-type cells at a density below 9,000 cells per cm2. All 21 mutants tested exhibited thymidine auxotrophy; neither folinic acid nor deoxyuridine could support mutant cell growth. There was no detectable TS activity in all 11 mutants so far examined and only about 50% of wild-type activity in three prototrophic revertants, as measured by whole-cell and cell-free enzyme assays. The apparent Michaelis-Menten constant (Km) for deoxyuridine-5'-monophosphate and inhibition constant (Ki) for 5-fluoro-deoxyuridine-5'-monophosphate, measured by whole-cell enzyme assay, appear to be similar for the wild-type and revertant cell lines. Using 5-fluoro-[6-3H]-2'-deoxyuridine 5'-monophosphate as active site titrant, the relative amounts of TS in crude cell extract from the parental, revertant, and mutant cells were shown to exist in a 1:0.5:0 ratio. Furthermore, the enzymes from two revertants were more heat labile than that of V79 cells. These properties, taken together, suggest that the FAT-resistant, thymidine auxotrophic phenotype may be the result of a structural gene mutation at the TS locus. The availability of such a mutant facilitates studies on thymidylate stress in relation to DNA metabolism, cell growth, and mutagenesis.

Chromosome breakage induced by thymidylate stress in thymidylate synthase-negative mutants of mouse FM3A cells

In thymidylate synthase-negative mutants of mouse FM3A cells, thymidine starvation rapidly decreased mitotic activity and resulted in cell death (thymineless death). When the thymidine starvation was reversed by an addition of thymidine, mitotic activity was recovered, but the majority of mitotic cells exhibited extensive chromosome aberrations, including chromatid breaks, chromatid exchanges, and pulverizations. Autoradiographic examination revealed that chromosome instability was induced only in cells arrested in the S phase during thymidine starvation. Furthermore, the most sensitive sites to the chromosome-damaging effect appeared to be sites which had replicated just prior to thymidine starvation. During thymidine starvation, cells at other stages in the cell cycle were accumulated at the G1-S boundary, and they were insensitive to the chromosome-damaging effect. Thymidine starvation was also found to be recombinagenic. Complete removal from the medium of a thymidine analogue, 5-bromo-2'-deoxyuridine, resulted in a dramatic increase in the frequency of sister chromatid exchanges. These results support the view that thymidine starvation in mammalian cells results in thymineless death via induction of DNA double-strand breaks, leading to chromosome fragmentation as well as rearrangements in the cells synthesizing DNA.

Accumulation of DNA strand breaks during thymineless death in thymidylate synthase-negative mutants of mouse FM3A cells

Thymidylate synthase-negative mutants of cultured mouse cells were immediately committed to cell death upon thymidine deprivation, especially when the cells were synchronized in the S phase. Thymidylate deprivation induced single strand breaks in chromosome-size DNA strands, as measured by alkaline sucrose gradient sedimentation, giving rise to two peaks, one with large and the other with small fragments, the latter about the size of T4 DNA. An increase in the small DNA fragments paralleled that of thymineless death. Thymidine deprivation also produced double strand DNA fragments as determined by a method of neutral filter elution, and their extent paralleled that of cell death. Double-stranded DNA eluted through the filter sedimented as a single peak both in a neutral and in an alkaline sucrose gradient that coincided with that of the above small DNA fragments. Therefore, the strand breaks seemed to occur in some defined portions of the genome and in a specific manner compared to breaks induced by x-rays, which occurred rather randomly. Cycloheximide blocked both thymineless death and the production of the small DNA fragments. The strand breaks induced by thymidine starvation were not repaired but instead advanced on subsequent incubation of the cells in growth medium containing thymidine.

Rat basophilic leukemia cell lines defective in phospholipid methyltransferase enzymes, Ca2+ influx, and histamine release: reconstitution by hybridization

Variants of the rat basophilic leukemia (RBL) cell line were isolated and screened for phospholipid methyltransferase I and II activities, enzymes that convert phosphatidylethanolamine to phosphatidylcholine. Two variants were found that had decreased phospholipid methyltransferase enzyme levels and were unable to cause an influx of Ca2+ or release histamine in an IgE-mediated reaction. However, these cells were able to release histamine through an ionophore-induced reaction, indicating that the releasing mechanism distal to the Ca2+ channel was intact. One cell line, 1C1.B1, had low specific activity for phospholipid methyltransferase I. A second variant, 2H3.B6, had reduced phospholipid methyltransferase II activity. Although both variants were unable to incorporate label from [methyl-3H]methionine or [3H]serine into phosphatidylcholine, they were able to incorporate [methyl-3H]choline and myo-[2-3H(N)]inositol into phospholipids. Fusion of the two cell lines and isolation on selective media resulted in the growth of eight independent hybrids. All eight had an increased number of chromosomes and normal phospholipid methyltransferase activities. Stimulation of the hybrids with IgE resulted in CA2+ influx and histamine release. These results indicate that phospholipid methylation precedes and is necessary for Ca2+ influx, and they further support the hypothesis that methylation is a necessary early step in the IgE-mediated histamine release reaction in RBL cells.

Selection of mammalian thymidine auxotrophic cell mutants defective in thymidylate synthase by their reduced sensitivity to methotrexate

Thymidine auxotrophic mutants were selectively isolated from mutagenized mouse FM3A cells by resistance to methotrexate in the presence of thymidine and 5-methyl-tetrahydrofolate with a frequency of 10(-5)-10(-6). In most of the thymidine auxotrophs the activity of thymidylate synthase was very low or undetectable, but dihydrofolate reductase activity was normal. Upon starvation of thymidine, the mutant cells immediately stopped growing and started to lyse within one day. In the presence of thymidine, the mutant cells grew quite normally. This phenotype behaved recessively in cell-cell hybrids, and the segregation profile of its marker indicated that the lesions in the mutants are not linked to the X chromosome. Prototrophic revertants could be isolated from these mutants, and they showed almost the normal level of thymidylate synthase activity. The selection method described here should be useful for isolating large numbers of thymidylate synthase-negative mutants from various mammalian cell lines.

Mercaptopurine pharmacogenetics: monogenic inheritance of erythrocyte thiopurine methyltransferase activity

Thiopurine methyltransferase (TPMT) catalyzes thiopurine S-methylation, an important metabolic pathway for drugs such as 6-mercaptopurine. Erythrocyte (RBC) TPMT activity was measured in blood samples from 298 randomly selected subjects. Of the subjects, 88.6% were included in a subgroup with high enzyme activity (13.50 +/- 1.86 U, mean +/- SD), 11.1% were included in a subgroup with intermediate activity (7.20 +/- 1.08 U), and 0.3% had undetectable activity. This distribution conforms to Hardy-Weinberg predictions for the autosomal codominant inheritance of a pair of alleles for low and high TPMT activity, TPMT(L) and TPMT(H), with gene frequencies of .059 and .941, respectively. If RBC TPMT activity is inherited in an autosomal codominant fashion, then subjects homozygous for TPMT(H) would have high enzyme activity, subjects heterozygous for the two alleles would have intermediate activity, and subjects homozygous for TPMT(L) would have undetectable activity. The segregation of RBC TPMT activity among 215 first-degree relatives in 50 randomly selected families and among 35 members of two kindreds and one family selected because they included probands with undetectable RBC enzyme activity were also compatible with the autosomal codominant inheritance of RBC TPMT. For example, in eight matings between subjects with intermediate activity (presumed genotype TPMT(L)TPMT(H)) and subjects with high activity (presumed genotype TPMT(H)TPMT(H)), 47% (8/17) of the offspring had intermediate activity. This value is very similar to the 50% figure expected on the basis of autosomal codominant inheritance (chi(2) ([1]) = .059). Further experiments are required to determine whether this genetic polymorphism for an important drug metabolizing enzyme may represent one factor in individual variations in sensitivity to thiopurines.

[Absence of hydroxyindole O-methyltransferase, the terminal enzyme in melatonin synthesis, in the rat intestine]

The absence of hydroxinol-o-methyltrapherase, the terminal enzyme in the chain of melatonin biosynthesis in the gastrointestinal tract of rats was revealed by means of radioisotopic biochemical methods. A high endogenous methylation in the mucosa of the small intestine of rats with the participation of S-adenozylmethionine, differing from melatonin synthesis in the epiphysis, was revealed during this study.

Deficiency of S-adenosylmethionine-homocysteine methyltransferase activity in hepatoma cells

S-Adenosylmethionine-homocysteine methyltransferase, which catalyzes synthesis of methionine from homocysteine, with the use of S-adenosylmethionine as the methyl donor, is absent in tumor tissue such as rat ascites hepatoma and Morris hepatoma but is present in rat liver homogenate. Absence of the enzymatic activity in tumor cells is not due to the action of an inhibitor. S-Adenosylhomocysteine hydrolase, however, is present in both rat liver and hepatoma tissue.