|
A), which changes the conserved methionine at position 581 of the enzyme to isoleucine; this mutation is coinherited with two other, common MTHFR polymorphisms (677CT and 1298AC), each of which is associated with depressed enzyme function.3,4 We propose that a nitrous oxide–induced defect of methionine synthase superimposed on an inherited defect of MTHFR caused the patient's death. Case Report
Details of the patient's clinical course and biochemical and pathological findings were reported by Beckman et al. in 1987.2 In brief, the child appeared normal until three months of age, when a mass in the left leg was noted. It was not known before the patient's surgery that his father and one of his uncles had serum levels of total homocysteine above 20.0 µmol per liter and above 30.0 µmol per liter, respectively (normal range, 5.4 to 13.9). The proband's sibling, who was receiving lifelong therapy with high-dose vitamin B supplements, had a homocysteine level of 4.3 µmol per liter. Neither the father nor the sibling had received nitrous oxide. On preoperative assessment for excisional biopsy of the mass, the patient's physical status was deemed class I according to the American Society of Anesthesiologists criteria (class I denotes good health and class V critical illness). After premedication with atropine and induction of anesthesia with sodium thiopental and succinylcholine, the child's trachea was intubated; anesthesia was maintained with 0.75 percent halothane and 60 percent nitrous oxide in oxygen for 45 minutes.
The surgical biopsy revealed an infantile fibrosarcoma, and resection of the mass was performed on the fourth day after the biopsy. After induction of anesthesia with halothane, the child was anesthetized for 270 minutes with 0.75 percent halothane and 60 percent nitrous oxide. At the conclusion of the operation, his trachea was extubated, and he was transferred while awake to the intensive care unit. He was discharged on the seventh postoperative day in apparently good health.
Seventeen days later (25 days after the resection), the patient was admitted because of seizures and episodes of apnea. On examination, the infant was severely hypotonic, without reflexes and with ataxic ventilation. Cranial computed tomography showed generalized atrophy of the brain, with enlarged prepontine and medullary cisterns. The urine was positive for homocystine (1.30 µmol per milligram of creatinine [normal value, 0]) but negative for organic acids and methylmalonic acid. The plasma homocystine level was elevated, at 0.6 mg per deciliter (normal value, <0.01), and the methionine level was low, at 0.06 mg per deciliter (normal mean [±SD] value, 0.48±0.18); the vitamin B12 level was normal, at 403 pg per milliliter (297 pmol per liter) (normal range, 150 to 800 pg per milliliter [111 to 590 pmol per liter]). The serum folate level, measured by radioimmunoassay, was 3.8 ng per milliliter (8.5 nmol per liter) (normal range, 2.5 to 15.0 ng per milliliter [5.6 to 33.4 nmol per liter]), and the level of folate in the cerebrospinal fluid was 26.0 ng per milliliter (57.9 nmol per liter) (normal range, 10.6 to 85.0 ng per milliliter [23.6 to 189.2 nmol per liter]).
The patient died at 130 days of age (46 days after the resection) after a respiratory arrest. An autopsy showed asymmetric cerebral atrophy and severe demyelination, with astrogliosis and oligodendroglial-c ell depletion in the midbrain, medulla, and cerebellum. The values for MTHFR activity in cultured fibroblasts reported post mortem were 1.22 and 0.8 nmol of formaldehyde produced per milligram of protein per hour (normal mean value, 5.04±1.36) with and without flavinadenine dinucleotide, respectively. The simultaneous control values were 6.4 and 5.4 nmol of formaldehyde produced per hour per milligram of protein with and without flavinadenine dinucleotide, respectively.5
Methods
The investigations were carried out with approval of the University of Wisconsin's institutional review board. Written informed consent was obtained from all the participants.
Fibroblast Culture and MTHFR Activity
Fibroblasts were cultured from skin-punch biopsy specimens obtained from both parents and from the patient's stored samples. MTHFR activity was measured at confluence, as previously described.6 All the assays were performed in duplicate, with simultaneous assay of a normal control.
Preparation and Sequence Analysis of Genomic DNA
Genomic DNA was isolated from the cultured fibroblasts from the patient and both parents and from either blood or buccal cells from other relatives. Each of the 11 MTHFR exons was amplified from genomic DNA by the polymerase chain reaction (PCR) with the use of newly designed intronic primers.7,8 (The sequences of the primers are listed in Supplementary Appendix 1 with the full text of this article at http://www.nejm.org.) The patient's and both parents' PCR products were bidirectionally sequenced. A novel mutation in the patient's DNA at nucleotide 1755 (exon 10) and two previously described frequent polymorphisms at positions 677 (exon 4) and 1298 (exon 7) in the MTHFR gene were analyzed in the genomic DNA from the parents and other relatives with the use of the restriction enzymes NlaIII, HinfI, and MboII, as previously described.3,4 Family members were also screened as previously described for common polymorphisms in the genes encoding enzymes that regulate folate and homocysteine metabolism; these polymorphisms have been implicated in the pathogenesis of neural-tube defects, other congenital anomalies, and cardiovascular and neoplastic disease.9 The polymorphisms include those that encode methionine synthase (MTR; the polymorphism results in the substitution of glycine for aspartic acid at residue 919),10 methionine synthase reductase (MTRR; the polymorphism results in the substitution of methionine for isoleucine at residue 22),11 and cystathionine 
-synthase (CBS; the polymorphism is a 68-bp duplication).12
RNA Analysis
To evaluate the expression of an intact copy of the predominant 7.2-kb MTHFR isoform,13 RNA was isolated from the patient's cultured fibroblasts. A 2206-bp product containing the entire coding region was amplified by PCR from the complementary DNA (cDNA) transcript and sequenced in full. The 7.2-kb cDNA product was amplified as seven overlapping fragments ranging from 1.0 to 2.2 kb in size, as verified by gel electrophoresis. (The primers used to sequence the cDNA transcript and to amplify the cDNA as overlapping fragments are listed in Supplementary Appendixes 2 and 3, respectively, with the full text of this article at http://www.nejm.org.) Bands corresponding to the expected fragment sizes were excised, and the first 300 bases of the 5' and 3' ends were sequenced to allow positive identification of each fragment. Fragments from the patient and an unrelated control were then compared.
Results
Enzyme Activity in Fibroblasts
The patient's MTHFR activity in two replicates was 0.76 and 0.03 nmol of formaldehyde per milligram of protein per hour (normal mean value, 13.3±4.6 with the use of the current method of measurement6), with a simultaneous normal control of 11.52 nmol of formaldehyde per milligram of protein per hour. MTHFR activity in the father and mother (1.8 and 6.1 nmol of formaldehyde per milligram of protein per hour, respectively) was reduced, with a control level of 9.5 nmol of formaldehyde per milligram of protein per hour.
Genomic DNA-Sequence Analysis
The patient was found to be heterozygous for a novel mutation, 1755GA in exon 10, which causes a substitution of isoleucine for methionine at residue 581 (M581I)14 (GenBank accession number, NM_005957). Restriction-enzyme analysis confirmed the presence of the 1755GA mutation in the heterozygous patient, his father, his brother, one uncle, and one aunt, but not in 100 control chromosomes. The patient was also heterozygous for a 677CT mutation in exon 4 (resulting in a substitution of valine for alanine at residue 222) and a 1298AC mutation in exon 7 (resulting in a substitution of alanine for glutamic acid at residue 429). In addition to being heterozygous for 1755GA, the father was homozygous (TT) for the 677CT mutation and homozygous (AA) at 1298A (Figure 2). The mother was heterozygous for both common polymorphisms and homozygous (wild type) at 1755G. The sibling's haplotype was identical to that of the patient in all coding regions. The novel mutation at 1755GA was therefore transmitted to the patient from a paternal chromosome, in cis configuration with the 677CT mutation. Two of the father's four siblings had haplotypes identical to the father's haplotype and were heterozygous for the 1755GA mutation and homozygous for the 677CT mutation (Table 1).
|
|
We also performed genomic analysis of the genes encoding methionine synthase, methionine synthase reductase, and cystathionine -synthase. Genotypes at these loci for all members of the pedigree are provided in Table 1.
RNA Analysis
No size differences were observed among the seven MTHFR cDNA fragments, indicating that the patient's fibroblasts expressed an intact MTHFR transcript. The 2.2-kb product contained the entire coding region of the transcript and was used to sequence a region beginning 50 bp in the 5' direction from the translational start site and ending 150 bp downstream of the stop codon. This product was of the expected length, and no alternate splicing variants were detected. The entire product was sequenced and compared with the published sequence14 (GenBank accession number, NM_005957). The presence of the heterozygous common polymorphisms 677CT and 1298AC, as well as the heterozygote substitution 1775GA, was confirmed.
Discussion
The inactivation of methionine synthase by nitrous oxide has been demonstrated with purified enzyme,16 in cultured cells,17,18 in animal models,19 and in humans.20,21,22 The mean half-time of inactivation is 46 minutes. Residual methionine synthase activity more than 200 minutes after the start of nitrous oxide administration approaches zero.21 Mice, pigs, and rats exposed to nitrous oxide have delayed recovery of enzyme activity for periods of four days or more.19,23,24,25 Recovery in cultured cells indicates that nitrous oxide–mediated inhibition is irreversible, with de novo synthesis of the enzyme required to restore activity.26
The untoward consequences of nitrous oxide anesthesia in our patient are reminiscent of two recent case reports. In the first, an eight-month-old child had acute neurologic deterioration six days after an 80-minute period of anesthesia with nitrous oxide.27 In the second, a four-month-old child was admitted because of hypotonia, dehydration, and acidosis three weeks after surgery that had involved a 180-minute period of anesthesia with nitrous oxide.28 Both children were found to have severe dietary cobalamin deficiency. These instances of methionine synthase inhibition have a time course and clinical features similar to those observed in our patient but were nonlethal, perhaps because they were elicited after only a single exposure to nitrous oxide. Moreover, our patient had an inborn error of metabolism in an essential precursor in a metabolic pathway, rather than an acquired deficiency, and nitrous oxide was delivered on two occasions only a few days apart.
Severe MTHFR deficiency is an autosomal recessive disorder characterized by progressive hypotonia, convulsions, and psychomotor retardation. The clinical presentation may be subtle, with the disorder manifested as developmental disability in the setting of moderate homocystinuria and hyperhomocysteinemia and low-to-normal levels of plasma methionine.29 Twenty-nine mutations in MTHFR are associated with severe deficiency, with a resulting activity level that is usually 0 to 30 percent of control activity.7,8,14,30,31,32,33 Most patients are heterozygous for multiple MTHFR substitutions; a small minority are homozygous for mutations at this locus. The 1755GA substitution identified in our patient occurs in a phylogenetically conserved region of the MTHFR protein (as assessed with BLASTP software, version 2.2.1). This region, which is thought to be essential for functional protein folding,34 is a "hot spot" for mutations leading to MTHFR deficiency (1711CT, 1727CT, 1762AT, and 1768GA).7,8 The heterozygous presence of the 1755GA substitution in the patient's father, brother, one uncle, and one aunt and its absence in 100 independent control chromosomes suggest that it is not a benign variant.
Compound heterozygosity for the common MTHFR alleles 677CT and 1298AC, as seen in the patient, his mother, and his brother, causes elevations in the plasma homocysteine level4 that are associated with a 50 to 60 percent decrement in enzyme activity.35 In the absence of coding mutations elsewhere in the MTHFR gene or evidence of a mutant splice variant, our patient's deficient enzyme activity may be attributable to compound heterozygosity for the novel 1755GA mutation, with the prevalent 677CT polymorphism on the same (paternal) chromosome and the 1298AC mutation on the maternal chromosome. It has recently been shown that when mutations causing severe MTHFR deficiency are expressed in cis configuration with the common 677CT variant, the resultant phenotype is markedly aggravated.34
Every year, approximately 45 million persons in North America undergo anesthesia, and nitrous oxide constitutes a major component in about half these procedures.36 Because of the growing use of nitrous oxide,37,38,39,40,41 patients with known mutations associated with mild or severe abnormalities in folate-cycle enzymes are increasingly likely to receive nitrous oxide. On the strength of the current findings, we believe that patients with a diagnosis of severe MTHFR deficiency should not receive nitrous oxide as anesthesia. In the case of emergency procedures, patients whose clinical presentation fits that of severe MTHFR deficiency, even if the disorder has not been diagnosed, should also not receive nitrous oxide. In the case of elective procedures, patients whose clinical presentation fits that of severe MTHFR deficiency should be evaluated, and the diagnosis should be ruled out before anesthesia with nitrous oxide is contemplated.
Supported by grants from the Doris Duke Foundation and the University of Wisconsin Department of Anesthesiology Research and Development Fund (to Dr. Hogan) and the Canadian Institutes of Health Research (to Dr. Rosenblatt).
We are indebted to Singh Sekhon, Ph.D. (University of Wisconsin, Madison), for the fibroblast cultures and to Robert H. Allen, M.D. (Metabolite Laboratories, Denver), for the assays of total homocysteine.
Source Information
From the Departments of Anesthesiology (R.R.S., K.H.) and Medical Genetics (R.L.), University of Wisconsin Medical School, Madison; and the Departments of Human Genetics, Medicine, Pediatrics, and Biology, McGill University, Montreal (D.S.R.).
Address reprint requests to Dr. Hogan at the Department of Anesthesiology, B6/319 Clinical Sciences Center, 600 Highland Ave., Madison, WI 53792, or at khogan{at}facstaff.wisc.edu.
References
T can further reduce activity when expressed in cis with severe mutations for human methylenetetrahydrofolate reductase. Hum Mutat 2000;16:132-138. [Medline]
| |||||||||||||||||||||||||||||||||||||||||||||||||||||
Related Letters:
Nitrous Oxide and 5,10-Methylenetetrahydrofolate Reductase
Burman J. F., Kaufman J. L., Hogan K., Erbe R. W.
Extract |
Full Text |
PDF
N Engl J Med 2003;
349:1479-1480, Oct 9, 2003.
Correspondence
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
