- Email: [email protected]
The C677T mutation in the methylenetetrahydrofolate reductase gene contributes to hyperhomocysteinemia in patients taking anticonvulsants
Brain & Development 24 (2002) 223–226 www.elsevier.com/locate/braindev
Original article
The C677T mutation in the methylenetetrahydrofolate reductase gene contributes to hyperhomocysteinemia in patients taking anticonvulsants Hiroaki Ono*, Akiko Sakamoto, Nobuyuki Mizoguchi, Nobuo Sakura Department of Pediatrics, Hiroshima University School of Medicine, Kasumi 1-2-3, Minami-ku, Hiroshima 734-8551, Japan Received 9 April 2001; received in revised form 4 December 2001; accepted 13 December 2001
Abstract Hyperhomocysteinemia, a possible risk factor for vascular disease can result from folate deficiency due to anticonvulsant therapy. A reaction catalyzed by 5,10-methylenetetrahydrofolate reductase (MTHFR) supplies 5-methyltetrahydrofolate, needed to remethylate homocysteine to methionine. MTHFR gene mutation (C677T) also can lead to hyperhomocysteinemia. We examined interaction between anticonvulsant therapy, C677T homozygosity, serum folate concentration, and plasma total homocysteine (tHcy) concentration in 81 epileptic patients. Patients receiving monotherapy showed no difference in occurrence of hyperhomocysteinemia (tHcy . 90th percentile for controls) between homozygotes for C677T and heterozygotes or patients with no mutant MTHFR. No monotherapy patient was folate deficient (,3 ng/ml). Among patients receiving multidrug therapy, hyperhomocysteinemia in homozygotes for C677T occured significantly more often than in heterozygotes or patients with no mutant enzyme (88.9 vs. 21.1%). The same was true for folate deficiency (44.4 vs. 0%). The C677T mutation is closely related to hyperhomocysteinemia and folate deficiency in epileptic patients taking multiple anticonvulsants. q 2002 Elsevier Science B.V. All rights reserved. Keywords: Hyperhomocysteinemia; Folate deficiency; C677T mutation; Anticonvulsants
1. Introduction Homocysteine (Hcy) is a sulfur-containing amino acid formed during metabolism of methionine. Folate and vitamin B12 are required for remethylation of Hcy to methionine. For this reaction, 5-methyltetrahydrofolate serves as the methyl donor catalyzed by methionine synthase, while vitamin B12 acts as a coenzyme. The enzyme 5,10-methylenetetrahydrofolate reductase (MTHFR; EC 1.5.1.20) supplies the 5-methyltetrahydrofolate required for this reaction and is the key enzyme in methylation of Hcy to methionine. Deficiency or enzyme dysfunction involving vitamin B12 or folate can cause hyperhomocysteinemia, a risk factor for vascular disease [1]. Several investigators have reported increased total plasma Hcy (tHcy) concentrations and folate deficiency in patients taking anticonvulsants, as we reported previously [2–4]. Recent study has demonstrated a commonly occurring 677C ! T mutation (C677T) in the MTHFR gene [5], which reduces the activity of the reductase enzyme and * Corresponding author. Tel.: 181-082-257-5212; fax: 181-082-2575214. E-mail address: [email protected] (H. Ono).
renders it thermolabile. Homozygotes for the C677T mutation represent as many as 5–18% of various populations, and are predisposed to hyperhomocysteinemia only when plasma folate concentrations are low [6]. Both aniconvulsant treatment and the C677T mutation can cause folate deficiency and hyperhomocysteinemia. However, the degree to which folate deficiency and hyperhomocysteinemia can occur with anticonvulsant medications in homozygotes with C677T mutation has not been determined, except in one study of Yoo and Hong [7]. We investigated plasma tHcy and serum folate concentrations as well as C677T MTHFR zygosity in epileptic patients taking anticonvulsants. We found the C677T mutation to be closely associated with hyperhomocysteinemia and folate deficiency in patients treated with multiple anticonvulsants.
2. Subjects and methods Eighty-one outpatients with epilepsy who were regularly evaluated and treated in the pediatric neurology clinic at the Hiroshima University School of Medicine were enrolled. Ages ranged from 1 to 32 years (median, 13). Fifty-three patients were treated with monotherapy and 28 patients
0387-7604/02/$ - see front matter q 2002 Elsevier Science B.V. All rights reserved. PII: S 0387-760 4(02)00004-9
224
H. Ono et al. / Brain & Development 24 (2002) 223–226
Table 1 Hyperhomocysteinemic patients according to therapy and genotype a Therapy
Genotype 2 /2
Monotherapy Multidrug therapy
The Mann–Whitney U-test and Fisher’s exact probability test were used to analyze data. 3. Results
1 /2 or 1/1
No.
%
No.
%
0/7 8/9
0 88.9 b
5/46 4/19
10.9 21.1
a
The number of hyperhomocysteinemic patients is indicated as the numerator and the number of all patients in each group is indicated as the denominator. b P , 0:01 vs 1/2 or 1/1 group taking multidrug therapy (Fisher’s exact probability test).
were treated with multiple anticonvulsants. Agents used for monotherapy included valproic acid (VPA; 24 patients), carbamazepine (CBZ; 13 patients), phenobarbital (PB; eight patients), phenytoin (PHT, five patients), and zonisamide (ZNS, three patients). Various combinations of these agents were used in the muliple-drug group. The median duration of treatment was 3.4 years in the monotherapy group and 6.5 years in the multidrug therapy group. No patient had malnutrition, severe liver dysfunction, or renal dysfunction. Informed consent was obtained from all subjects or guardians before the study. Blood samples from epileptic patients were obtained a few hours after they had taken their morning doses of anticonvulsants. All samples were non-fasting, since in a preliminary experiment involving six healthy adults the plasma concentrations of tHcy were the same in samples obtained before and 4 h after breakfast (data not shown). After venous blood samples were obtained, genomic DNA was extracted from white blood cells while plasma or serum was separated immediately by centrifugation at 800 £ g for 10 min. Samples were stored at 2208C until analysis. Plasma tHcy concentrations in all subjects were measured by high-performance liquid chromatography using fluorescence detection, as described previously [2]. Folate and vitamin B12 were measured by a competitive protein-binding radioassay utilizing a Corning kit purchased from CibaCorning Diagnostics (Medfield, MA, USA). The status of the C677T MTHFR gene was investigated by amplification by a polymerase chain reaction and digestion of the products by the endonuclease Hinf I [5]. The mutant allele was designated ‘ 2 ’ and the wildtype, ‘ 1 ’; normal (1/1), heterozygous (1/2), and homozygous (2/2) genotypes were identified. Plasma tHcy concentrations were considered elevated when they exceeded 90th-percentile values for 81 healthy control subjects (7.6 mM in patients aged 1–14 years, and 10.5 mM in patients aged .15 years as reported previously) [2]. Folate concentrations ,3 ng/ml were considered deficient, as reported previously [2]. Since normal range of serum folate concentration was 3– 8 ng/ml, patients showing less than 5 ng/ml was considered as subjects with low-normal serum folate concentrations.
In the monotherapy group, no significant difference was seen in the fraction of patients who were hyperhomocysteinemic between patients with the 2/2 genotype and those with 1/2 or 1/1 genotypes. No 2/2 patients showed hyperhomocysteinemia and no folate-deficient patients were found in any genotype group. Among seven 2/2 patients receiving monotherapy, only one had a serum folate concentration ,5 ng/ml. In the group taking multiple drugs, occurrence of both hyperhomocysteinemia and folate deficiency among patients with the 2/2 genotype was significantly higher than among subjects with genotypes 1/2 or 1/1. Of eight hyperhomocysteinemic 2/2 patients undergoing multidrug therapy, seven had serum folate concentrations ,5 ng/ml (Tables 1–3). Serum vitamin B12 levels were within the normal range in all patients (230– 1200 pg/ml). No significant difference in duration of anticonvulsant treatment was seen between patients with the 2/2 genotype and those with the 1/2 or 1/1 genotype in either monotherapy or multidrug therapy group. No significant difference in fraction of 2/2 patients showing high serum concentration of anticonvulsant was seen between monotherapy and multidrug therapy group (Table 3). Generally, patients did not experience thromboembolic events during treatment. In one patient cerebral infarction was diagnosed by magnetic resonance imaging, but the plasma tHcy concentration was not abnormally high (3.3 mM); the patient was heterozygous for the C677T mutation. No significant difference in seizure frequency was appreciated between hyperhomocysteinemic and nonhyperhomocysteinemic patients. 4. Discussion This study demonstrated that the C677T mutation is associated with hyperhomocysteinemia and folate deficiency in patients taking multiple anticonvulsant drugs. On the other Table 2 Fractions of folate-deficient patients taking multiple-drug therapy, by genotype a Genotype 2 /2 No. 4/9
1 /2 or 1/1 % 44.4
b
No.
%
0/19
0
a The number of folate-deficient patients is indicated as the numerator and the number of all patients in each group is indicated as the denominator. b P , 0:01 vs 1/2 or 1/1 group (Fisher’s exact probability test).
H. Ono et al. / Brain & Development 24 (2002) 223–226
225
Table 3 Serum FA and plasma tHcy concentrations in homozygotes for C677T mutation taking anticonvulsants a No.
Therapy
Age
Duration of treatment (year)
FA (ng/ml)
tHcy (mM)
Serum concentration of anticonvulsant (mg/ml)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
CBZ PB ZNS VPA PB VPA VPA PB PRM PHT CBZ PB VPA PRM CBZ ZNS CZP PHT PB VPA CBZ VPA PB ZNS VPA VPA CZP VPA CZP
29 4 3 12 3 12 7 20 30 11 21 15 13 13 17 8
20.6 3.2 2.3 0.7 1.3 1.6 1.7 18 13 8.8 11 4.3 10 12 7 6
4.8 5.7 6.9 7 7.4 8 9.8 2.4 2.5 2.8 2.9 3.5 4.2 4.3 6.8 7.3
9 5.3 5.1 5 2.6 7.1 3.9 57.9 15.3 9.1 29.1 23.6 8.7 12.2 20.2 4.9
2.9 6.1 39.8 114 11.6 30.6 65.8 PB 26.2, PRM 8.0, PHT 4.2 CBZ 3.6, PB 3.5 VPA 84.0, PRM 5.7, PB 13.8 CBZ 9.1, ZSA 31.2, CZP 0.025 PHT 9.9, PB 16.7, VPA 66.2 CBZ 5.7,VPA 59.8 PB 22.9, ZSA 39.0, VPA 79.2 VPA 105.0, CZP 0.023 VPA 75.6, CZP 0.006
a
FA, serum folate concentration; tHcy, plasma total homocysteine concentration; CBZ, carbamazepine; PB, phenobarbital; ZNS, zonisamide; VPA, vaiproate; PRM, primidone; PHT, phenytoin; CZP, clonazepam. Normal range of serum concentration of anticonvulsant (mg/ml): CBZ 4–12; VPA 50– 100; PB 10–30; PRM 5–12; PHT 10–20; ZNS 10–30; CZP 0.025–0.075. Elevated levels of tHCy: .7.6 mM (1–14 years); .10.5 mM (.15 years). Deficient levels of FA: ,3 ng/ml.
hand, monotherapy in 2/2 patients was not associated with hyperhomocysteinemia or a folate deficiency. Of nine 2/2 patients undergoing multidrug therapy, five was treated with CBZ or PHT. On the other hand, among seven 2/2 patients undergoing monodrug therapy, only one patient was treated with CBZ and no patient was treated with PHT. Yoo and Hong have shown that epileptic patients with the C677T mutation had a high risk for hyperhomocysteinemia when treated with CBZ or PHT [7]. Thus, our result might be caused by the difference of number of patients taking anticonvulsant therapy including CBZ or PHT. The other three patients among multidrug therapy group showed hyperhomocysteinmia and C677T mutation and they were treated with anticonvulsants other than CBZ or PHT. This data might be inconsistent data with that by Yoo and Hong. However, since the number of 2/2 patients taking various multiple anticonvulsant drugs was very small, study of additional patients will be needed to clarify issues concerning correlations between hyperhomocysteinemia, folate deficency, C677T mutations, and multidrug therapy other than CBZ or PHT. Generally, the relationship between the C677T mutation and increased Hcy is strongly influenced by folate status [8]. Most homozygous individuals with low-normal plasma folate levels have increased tHcy, while, those with highnormal plasma folate levels have normal tHcy levels [8]. Seven of eight 2/2 patients with hyperhomocysteinemia receiving multidrug therapy showed serum folate levels ,5 ng/ml, which means low-normal serum folate levels. All patients except case 12 had received long-term treatment for .5 years. However, most 2/2 patients in the monotherapy group had serum folate concentrations
.5 ng/ml. The exception was patient 1, who had a 20year history of CBZ therapy. No 2/2 patient receiving monotherapy showed hyperhomocysteinemia. These findings suggested that low serum folate concentrations induced by prolonged multidrug therapy and impaired MTHFR activity in 2/2 patients might cause hyperhomocysteinemia. Mild to moderate homocysteinemia is a risk factor for vascular disease [1]. Boushey et al. have reported that a tHcy value above the 90th percentile for control subjects was associated with increased risk of atherosclerotic vascular disease [9]. Several prospective cohort studies have confirmed a significant association between hyperhomocysteinemia and myocardial infarction or stroke [1]. Most of our patients showed no clinical symptoms related to hyperhomocysteinemia including thrombo-embolism. However, to prevent vascular complications, we would recommend folic acid supplementation in patients taking anticonvulsants who have hyperhomocysteinemia. This is especially true for 2/2 patients receiving long-term therapy with multiple drugs. Several investigators have reported that seizures are induced by administration of high doses of exogenous homocysteine in experimental animals [10,11]. Further, some 21% of homocystinuric patients with cystathionine b-synthase deficiency who were not treated for their metabolic defect beginning in early infancy have experienced seizures [12]. In our study no patient showed plasma tHcy elevations as marked in homocystinuric patients, and seizure frequency in hyperhomocysteinemic patients did not differ from those in our other patients. This result suggests that mild to moderate hyperhomocysteinemia is
226
H. Ono et al. / Brain & Development 24 (2002) 223–226
not necessarily associated with increased seizure frequency in epileptic patients. Since our patients had a variety of etiologies of epilepsy and received varied anticonvulsant treatments, factors other than hyperhomocysteinemia are likely to have affected their seizure frequency. However, no patient with a plasma tHcy concentration exceeding 20 mM showed optimal seizure control; seizure frequencies ranged from 2 per year to 10 per month. If a trial of folic acid administration in these patients could decrease both plasma tHcy concentrations and seizure frequency, a relationship between hyperhomocysteinemia and seizure occurrence would be suggested. In conclusion, our results showed that hyperhomocysteinemia and folate deficiency can occur during anticonvulsant therapy, especially in patients homozygous for the C677T mutation who are treated with multiple drugs. Careful observation of these patients is required to prevent complications that might result from hyperhomocysteinemia and folate deficiency. References [1] Hankey GJ, Eikelboom JW. Homocysteine and vascular disease. Lancet 1999;354:407–413. [2] Ono H, Sakamoto A, Eguchi T, Sakura N, Noumura S, Fujita N, et al. Plasma total homocysteine concentrations in epileptic patients taking anticonvulsants. Metabolism 1997;46:959–962.
[3] James GK, Jones MW, Pudek MR. Homocyst(e)ine levels in patients on phenytoin therapy. Clin Chem 1997;30:647–649. [4] Schwaninger M, Ringleb P, Winter R, Kohl B, Fiehn W, Rieser PA, et al. Elevated plasma concentrations of homocysteine in antiepileptic drug treatment. Epilepsia 1999;40:345–350. [5] Frosst P, Blom HJ, Milos R, Goyette P, Sheppard CA, Matthews RG, et al. A candidate genetic risk factor for vascular disease: a common mutation in methylenetetrahydrofolate reductase. Nat Genet 1995;10:111–113. [6] Rozen R. Molecular genetics of methylenetetrahydrofolate reductase deficiency. J Inherit Metab Dis 1996;19:589–594. [7] Yoo JH, Hong SB. A common mutation in the methylenetetrahydrofolate reductase gene is a determinant of hyperhomocysteinemia in epileptic patients receiving anticonvulsants. Metabolism 1999;48: 1047–1051. [8] Blom HJ. Mutated 5,10-methylenetetrahydrofolate reductase and moderate hyperhomocysteinenaemia. Eur J Pediatr 1998;157(Suppl 2):S131–S134. [9] Boushey CJ, Beresford SA, Omenn GS, Motulsky AG. A quantitative assessment of plasma homocysteine as a risk factor for vascular disease: probable benefits of increasing folic acid intake. J Am Med Assoc 1995;274:1049–1057. [10] Marangos PJ, Loftus T, Wiesner J, Lowe T, Rossu E, Browne CE, et al. Adenosinergic modulation of homocysteine-induced seizures in mice. Epilepsia 1990;31:239–246. [11] Mudd SH, Levy HL, Skovby F. Disorders of transsulfuration. In: Charles RS, Arthur LB, William SS, David V, editors. The metabolic and molecular bases of inherited disease, 7th ed. New York, NY: McGraw-Hill, 1995. pp. 1279–1327. [12] Mudd SH, Skovby F, Levy HL, Pettigrew FD, Wlekeu B, Pyritz RE, et al. The natural history of homocystinuria due to cystathionine bsynthase deficiency. Am J Hum Genet 1985;37:1–31.
Original article
The C677T mutation in the methylenetetrahydrofolate reductase gene contributes to hyperhomocysteinemia in patients taking anticonvulsants Hiroaki Ono*, Akiko Sakamoto, Nobuyuki Mizoguchi, Nobuo Sakura Department of Pediatrics, Hiroshima University School of Medicine, Kasumi 1-2-3, Minami-ku, Hiroshima 734-8551, Japan Received 9 April 2001; received in revised form 4 December 2001; accepted 13 December 2001
Abstract Hyperhomocysteinemia, a possible risk factor for vascular disease can result from folate deficiency due to anticonvulsant therapy. A reaction catalyzed by 5,10-methylenetetrahydrofolate reductase (MTHFR) supplies 5-methyltetrahydrofolate, needed to remethylate homocysteine to methionine. MTHFR gene mutation (C677T) also can lead to hyperhomocysteinemia. We examined interaction between anticonvulsant therapy, C677T homozygosity, serum folate concentration, and plasma total homocysteine (tHcy) concentration in 81 epileptic patients. Patients receiving monotherapy showed no difference in occurrence of hyperhomocysteinemia (tHcy . 90th percentile for controls) between homozygotes for C677T and heterozygotes or patients with no mutant MTHFR. No monotherapy patient was folate deficient (,3 ng/ml). Among patients receiving multidrug therapy, hyperhomocysteinemia in homozygotes for C677T occured significantly more often than in heterozygotes or patients with no mutant enzyme (88.9 vs. 21.1%). The same was true for folate deficiency (44.4 vs. 0%). The C677T mutation is closely related to hyperhomocysteinemia and folate deficiency in epileptic patients taking multiple anticonvulsants. q 2002 Elsevier Science B.V. All rights reserved. Keywords: Hyperhomocysteinemia; Folate deficiency; C677T mutation; Anticonvulsants
1. Introduction Homocysteine (Hcy) is a sulfur-containing amino acid formed during metabolism of methionine. Folate and vitamin B12 are required for remethylation of Hcy to methionine. For this reaction, 5-methyltetrahydrofolate serves as the methyl donor catalyzed by methionine synthase, while vitamin B12 acts as a coenzyme. The enzyme 5,10-methylenetetrahydrofolate reductase (MTHFR; EC 1.5.1.20) supplies the 5-methyltetrahydrofolate required for this reaction and is the key enzyme in methylation of Hcy to methionine. Deficiency or enzyme dysfunction involving vitamin B12 or folate can cause hyperhomocysteinemia, a risk factor for vascular disease [1]. Several investigators have reported increased total plasma Hcy (tHcy) concentrations and folate deficiency in patients taking anticonvulsants, as we reported previously [2–4]. Recent study has demonstrated a commonly occurring 677C ! T mutation (C677T) in the MTHFR gene [5], which reduces the activity of the reductase enzyme and * Corresponding author. Tel.: 181-082-257-5212; fax: 181-082-2575214. E-mail address: [email protected] (H. Ono).
renders it thermolabile. Homozygotes for the C677T mutation represent as many as 5–18% of various populations, and are predisposed to hyperhomocysteinemia only when plasma folate concentrations are low [6]. Both aniconvulsant treatment and the C677T mutation can cause folate deficiency and hyperhomocysteinemia. However, the degree to which folate deficiency and hyperhomocysteinemia can occur with anticonvulsant medications in homozygotes with C677T mutation has not been determined, except in one study of Yoo and Hong [7]. We investigated plasma tHcy and serum folate concentrations as well as C677T MTHFR zygosity in epileptic patients taking anticonvulsants. We found the C677T mutation to be closely associated with hyperhomocysteinemia and folate deficiency in patients treated with multiple anticonvulsants.
2. Subjects and methods Eighty-one outpatients with epilepsy who were regularly evaluated and treated in the pediatric neurology clinic at the Hiroshima University School of Medicine were enrolled. Ages ranged from 1 to 32 years (median, 13). Fifty-three patients were treated with monotherapy and 28 patients
0387-7604/02/$ - see front matter q 2002 Elsevier Science B.V. All rights reserved. PII: S 0387-760 4(02)00004-9
224
H. Ono et al. / Brain & Development 24 (2002) 223–226
Table 1 Hyperhomocysteinemic patients according to therapy and genotype a Therapy
Genotype 2 /2
Monotherapy Multidrug therapy
The Mann–Whitney U-test and Fisher’s exact probability test were used to analyze data. 3. Results
1 /2 or 1/1
No.
%
No.
%
0/7 8/9
0 88.9 b
5/46 4/19
10.9 21.1
a
The number of hyperhomocysteinemic patients is indicated as the numerator and the number of all patients in each group is indicated as the denominator. b P , 0:01 vs 1/2 or 1/1 group taking multidrug therapy (Fisher’s exact probability test).
were treated with multiple anticonvulsants. Agents used for monotherapy included valproic acid (VPA; 24 patients), carbamazepine (CBZ; 13 patients), phenobarbital (PB; eight patients), phenytoin (PHT, five patients), and zonisamide (ZNS, three patients). Various combinations of these agents were used in the muliple-drug group. The median duration of treatment was 3.4 years in the monotherapy group and 6.5 years in the multidrug therapy group. No patient had malnutrition, severe liver dysfunction, or renal dysfunction. Informed consent was obtained from all subjects or guardians before the study. Blood samples from epileptic patients were obtained a few hours after they had taken their morning doses of anticonvulsants. All samples were non-fasting, since in a preliminary experiment involving six healthy adults the plasma concentrations of tHcy were the same in samples obtained before and 4 h after breakfast (data not shown). After venous blood samples were obtained, genomic DNA was extracted from white blood cells while plasma or serum was separated immediately by centrifugation at 800 £ g for 10 min. Samples were stored at 2208C until analysis. Plasma tHcy concentrations in all subjects were measured by high-performance liquid chromatography using fluorescence detection, as described previously [2]. Folate and vitamin B12 were measured by a competitive protein-binding radioassay utilizing a Corning kit purchased from CibaCorning Diagnostics (Medfield, MA, USA). The status of the C677T MTHFR gene was investigated by amplification by a polymerase chain reaction and digestion of the products by the endonuclease Hinf I [5]. The mutant allele was designated ‘ 2 ’ and the wildtype, ‘ 1 ’; normal (1/1), heterozygous (1/2), and homozygous (2/2) genotypes were identified. Plasma tHcy concentrations were considered elevated when they exceeded 90th-percentile values for 81 healthy control subjects (7.6 mM in patients aged 1–14 years, and 10.5 mM in patients aged .15 years as reported previously) [2]. Folate concentrations ,3 ng/ml were considered deficient, as reported previously [2]. Since normal range of serum folate concentration was 3– 8 ng/ml, patients showing less than 5 ng/ml was considered as subjects with low-normal serum folate concentrations.
In the monotherapy group, no significant difference was seen in the fraction of patients who were hyperhomocysteinemic between patients with the 2/2 genotype and those with 1/2 or 1/1 genotypes. No 2/2 patients showed hyperhomocysteinemia and no folate-deficient patients were found in any genotype group. Among seven 2/2 patients receiving monotherapy, only one had a serum folate concentration ,5 ng/ml. In the group taking multiple drugs, occurrence of both hyperhomocysteinemia and folate deficiency among patients with the 2/2 genotype was significantly higher than among subjects with genotypes 1/2 or 1/1. Of eight hyperhomocysteinemic 2/2 patients undergoing multidrug therapy, seven had serum folate concentrations ,5 ng/ml (Tables 1–3). Serum vitamin B12 levels were within the normal range in all patients (230– 1200 pg/ml). No significant difference in duration of anticonvulsant treatment was seen between patients with the 2/2 genotype and those with the 1/2 or 1/1 genotype in either monotherapy or multidrug therapy group. No significant difference in fraction of 2/2 patients showing high serum concentration of anticonvulsant was seen between monotherapy and multidrug therapy group (Table 3). Generally, patients did not experience thromboembolic events during treatment. In one patient cerebral infarction was diagnosed by magnetic resonance imaging, but the plasma tHcy concentration was not abnormally high (3.3 mM); the patient was heterozygous for the C677T mutation. No significant difference in seizure frequency was appreciated between hyperhomocysteinemic and nonhyperhomocysteinemic patients. 4. Discussion This study demonstrated that the C677T mutation is associated with hyperhomocysteinemia and folate deficiency in patients taking multiple anticonvulsant drugs. On the other Table 2 Fractions of folate-deficient patients taking multiple-drug therapy, by genotype a Genotype 2 /2 No. 4/9
1 /2 or 1/1 % 44.4
b
No.
%
0/19
0
a The number of folate-deficient patients is indicated as the numerator and the number of all patients in each group is indicated as the denominator. b P , 0:01 vs 1/2 or 1/1 group (Fisher’s exact probability test).
H. Ono et al. / Brain & Development 24 (2002) 223–226
225
Table 3 Serum FA and plasma tHcy concentrations in homozygotes for C677T mutation taking anticonvulsants a No.
Therapy
Age
Duration of treatment (year)
FA (ng/ml)
tHcy (mM)
Serum concentration of anticonvulsant (mg/ml)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
CBZ PB ZNS VPA PB VPA VPA PB PRM PHT CBZ PB VPA PRM CBZ ZNS CZP PHT PB VPA CBZ VPA PB ZNS VPA VPA CZP VPA CZP
29 4 3 12 3 12 7 20 30 11 21 15 13 13 17 8
20.6 3.2 2.3 0.7 1.3 1.6 1.7 18 13 8.8 11 4.3 10 12 7 6
4.8 5.7 6.9 7 7.4 8 9.8 2.4 2.5 2.8 2.9 3.5 4.2 4.3 6.8 7.3
9 5.3 5.1 5 2.6 7.1 3.9 57.9 15.3 9.1 29.1 23.6 8.7 12.2 20.2 4.9
2.9 6.1 39.8 114 11.6 30.6 65.8 PB 26.2, PRM 8.0, PHT 4.2 CBZ 3.6, PB 3.5 VPA 84.0, PRM 5.7, PB 13.8 CBZ 9.1, ZSA 31.2, CZP 0.025 PHT 9.9, PB 16.7, VPA 66.2 CBZ 5.7,VPA 59.8 PB 22.9, ZSA 39.0, VPA 79.2 VPA 105.0, CZP 0.023 VPA 75.6, CZP 0.006
a
FA, serum folate concentration; tHcy, plasma total homocysteine concentration; CBZ, carbamazepine; PB, phenobarbital; ZNS, zonisamide; VPA, vaiproate; PRM, primidone; PHT, phenytoin; CZP, clonazepam. Normal range of serum concentration of anticonvulsant (mg/ml): CBZ 4–12; VPA 50– 100; PB 10–30; PRM 5–12; PHT 10–20; ZNS 10–30; CZP 0.025–0.075. Elevated levels of tHCy: .7.6 mM (1–14 years); .10.5 mM (.15 years). Deficient levels of FA: ,3 ng/ml.
hand, monotherapy in 2/2 patients was not associated with hyperhomocysteinemia or a folate deficiency. Of nine 2/2 patients undergoing multidrug therapy, five was treated with CBZ or PHT. On the other hand, among seven 2/2 patients undergoing monodrug therapy, only one patient was treated with CBZ and no patient was treated with PHT. Yoo and Hong have shown that epileptic patients with the C677T mutation had a high risk for hyperhomocysteinemia when treated with CBZ or PHT [7]. Thus, our result might be caused by the difference of number of patients taking anticonvulsant therapy including CBZ or PHT. The other three patients among multidrug therapy group showed hyperhomocysteinmia and C677T mutation and they were treated with anticonvulsants other than CBZ or PHT. This data might be inconsistent data with that by Yoo and Hong. However, since the number of 2/2 patients taking various multiple anticonvulsant drugs was very small, study of additional patients will be needed to clarify issues concerning correlations between hyperhomocysteinemia, folate deficency, C677T mutations, and multidrug therapy other than CBZ or PHT. Generally, the relationship between the C677T mutation and increased Hcy is strongly influenced by folate status [8]. Most homozygous individuals with low-normal plasma folate levels have increased tHcy, while, those with highnormal plasma folate levels have normal tHcy levels [8]. Seven of eight 2/2 patients with hyperhomocysteinemia receiving multidrug therapy showed serum folate levels ,5 ng/ml, which means low-normal serum folate levels. All patients except case 12 had received long-term treatment for .5 years. However, most 2/2 patients in the monotherapy group had serum folate concentrations
.5 ng/ml. The exception was patient 1, who had a 20year history of CBZ therapy. No 2/2 patient receiving monotherapy showed hyperhomocysteinemia. These findings suggested that low serum folate concentrations induced by prolonged multidrug therapy and impaired MTHFR activity in 2/2 patients might cause hyperhomocysteinemia. Mild to moderate homocysteinemia is a risk factor for vascular disease [1]. Boushey et al. have reported that a tHcy value above the 90th percentile for control subjects was associated with increased risk of atherosclerotic vascular disease [9]. Several prospective cohort studies have confirmed a significant association between hyperhomocysteinemia and myocardial infarction or stroke [1]. Most of our patients showed no clinical symptoms related to hyperhomocysteinemia including thrombo-embolism. However, to prevent vascular complications, we would recommend folic acid supplementation in patients taking anticonvulsants who have hyperhomocysteinemia. This is especially true for 2/2 patients receiving long-term therapy with multiple drugs. Several investigators have reported that seizures are induced by administration of high doses of exogenous homocysteine in experimental animals [10,11]. Further, some 21% of homocystinuric patients with cystathionine b-synthase deficiency who were not treated for their metabolic defect beginning in early infancy have experienced seizures [12]. In our study no patient showed plasma tHcy elevations as marked in homocystinuric patients, and seizure frequency in hyperhomocysteinemic patients did not differ from those in our other patients. This result suggests that mild to moderate hyperhomocysteinemia is
226
H. Ono et al. / Brain & Development 24 (2002) 223–226
not necessarily associated with increased seizure frequency in epileptic patients. Since our patients had a variety of etiologies of epilepsy and received varied anticonvulsant treatments, factors other than hyperhomocysteinemia are likely to have affected their seizure frequency. However, no patient with a plasma tHcy concentration exceeding 20 mM showed optimal seizure control; seizure frequencies ranged from 2 per year to 10 per month. If a trial of folic acid administration in these patients could decrease both plasma tHcy concentrations and seizure frequency, a relationship between hyperhomocysteinemia and seizure occurrence would be suggested. In conclusion, our results showed that hyperhomocysteinemia and folate deficiency can occur during anticonvulsant therapy, especially in patients homozygous for the C677T mutation who are treated with multiple drugs. Careful observation of these patients is required to prevent complications that might result from hyperhomocysteinemia and folate deficiency. References [1] Hankey GJ, Eikelboom JW. Homocysteine and vascular disease. Lancet 1999;354:407–413. [2] Ono H, Sakamoto A, Eguchi T, Sakura N, Noumura S, Fujita N, et al. Plasma total homocysteine concentrations in epileptic patients taking anticonvulsants. Metabolism 1997;46:959–962.
[3] James GK, Jones MW, Pudek MR. Homocyst(e)ine levels in patients on phenytoin therapy. Clin Chem 1997;30:647–649. [4] Schwaninger M, Ringleb P, Winter R, Kohl B, Fiehn W, Rieser PA, et al. Elevated plasma concentrations of homocysteine in antiepileptic drug treatment. Epilepsia 1999;40:345–350. [5] Frosst P, Blom HJ, Milos R, Goyette P, Sheppard CA, Matthews RG, et al. A candidate genetic risk factor for vascular disease: a common mutation in methylenetetrahydrofolate reductase. Nat Genet 1995;10:111–113. [6] Rozen R. Molecular genetics of methylenetetrahydrofolate reductase deficiency. J Inherit Metab Dis 1996;19:589–594. [7] Yoo JH, Hong SB. A common mutation in the methylenetetrahydrofolate reductase gene is a determinant of hyperhomocysteinemia in epileptic patients receiving anticonvulsants. Metabolism 1999;48: 1047–1051. [8] Blom HJ. Mutated 5,10-methylenetetrahydrofolate reductase and moderate hyperhomocysteinenaemia. Eur J Pediatr 1998;157(Suppl 2):S131–S134. [9] Boushey CJ, Beresford SA, Omenn GS, Motulsky AG. A quantitative assessment of plasma homocysteine as a risk factor for vascular disease: probable benefits of increasing folic acid intake. J Am Med Assoc 1995;274:1049–1057. [10] Marangos PJ, Loftus T, Wiesner J, Lowe T, Rossu E, Browne CE, et al. Adenosinergic modulation of homocysteine-induced seizures in mice. Epilepsia 1990;31:239–246. [11] Mudd SH, Levy HL, Skovby F. Disorders of transsulfuration. In: Charles RS, Arthur LB, William SS, David V, editors. The metabolic and molecular bases of inherited disease, 7th ed. New York, NY: McGraw-Hill, 1995. pp. 1279–1327. [12] Mudd SH, Skovby F, Levy HL, Pettigrew FD, Wlekeu B, Pyritz RE, et al. The natural history of homocystinuria due to cystathionine bsynthase deficiency. Am J Hum Genet 1985;37:1–31.