Isolated Sulfite Oxidase Deficiency: A Case Report With a Novel Mutation and Review of the Literature
http://www.100md.com
《小儿科》
Genetics Training Program Harvard Combined Medicine/Pediatrics Program, Harvard Medical School, Boston, Massachusetts
Pediatric Neurology Unit Amino Acid Disorders Laboratory, Department of Neurology
Nutrition Services Division of Neuroradiology, Massachusetts General Hospital, Boston, Massachusetts
ABSTRACT
Isolated sulfite oxidase deficiency is a rare but devastating neurologic disease that usually presents in early infancy with seizures and alterations in muscle tone. Only 21 cases have been reported in the literature. We report a case of a newborn infant boy with isolated sulfite oxidase deficiency who presented with generalized seizures on his fourth day of life. Plasma total homocysteine was not detectable. Urinary sulfite, thiosulfate, and S-sulfocysteine levels were elevated. The patient began a low-methionine and low-cysteine diet and was treated with thiamine and dextromethorphan. However, he became increasingly microcephalic and was severely developmentally delayed. Mutation analysis of the sulfite oxidase gene revealed that the patient was homozygous for a novel 4-base pair deletion, and both of his parents were found to be heterozygous carriers of the same deletion. We reviewed the clinical, biochemical, neuroradiologic, and neuropathologic features in all published cases of isolated sulfite oxidase deficiency. Seizures or abnormal movements were prominent features in all cases. Developmental delays were reported in 17 cases. Ectopia lentis was detected in 9 cases. Clinical improvement with dietary therapy was seen in only 2 patients, both of whom presented after the age of 6 months and had relatively mild developmental delays. Plasma or urinary S-sulfocysteine levels were elevated in all cases. Urinary sulfite was detected in all except 1 case. Cerebral atrophy and cystic encephalomalacia were observed with neuroradiologic imaging and were noted in all 3 postmortem reports of isolated sulfite oxidase deficiency. The main alternative in the differential diagnosis of isolated sulfite oxidase deficiency is molybdenum cofactor deficiency.
Key Words: sulfite sulfite oxidase SUOX gene molybdenum cofactor inborn error of metabolism
Abbreviations: CVS, chorionic villus sampling EEG, electroencephalographic MMP, matrix metalloproteinase NMDA, N-methyl-D-aspartate TIMP, tissue inhibitor of metalloproteinases
Sulfite oxidase deficiency is a neurometabolic disease that results in severe developmental delay and premature death. Sulfite oxidase (EC 1.8.3.1) is 1 of 3 enzymes in humans that requires molybdenum as a cofactor, the other 2 being xanthine oxidase and aldehyde oxidase.
Sulfite oxidase oxidizes potentially toxic sulfites to nontoxic sulfates in the final step of the sulfur amino acid metabolism pathway. Xanthine oxidase catalyzes the hydroxylation of hypoxanthine to xanthine and thence to uric acid. Aldehyde oxidase is involved in the hydroxylation of hypoxanthine to xanthine but not to uric acid. Xanthine oxidase deficiency (also known as xanthinuria) may either lead to the deposition of xanthine in kidneys and muscle, resulting in renal calculi and myopathy, or be totally asymptomatic.1 Although isolated aldehyde oxidase deficiency has not been reported for humans, it is thought to be relatively benign.1 Molybdenum cofactor deficiency would therefore be the main differential diagnosis that should be considered.
A number of patients with sulfite oxidase deficiency have now been reported in the literature, all of whom had relatively dismal outcomes. In this article, we present a case of a newborn with sulfite oxidase deficiency and a previously unreported mutation in the sulfite oxidase gene. We also review the metabolism of sulfur amino acids, as well as all previously published cases of sulfite oxidase deficiency.
CASE REPORT
A term infant boy was born to a 21-year-old primigravid mother from Colombia, after a pregnancy that was complicated by 1 episode each of urinary tract infection and pyelonephritis, both of which were treated with orally administered antibiotics only. The patient required minimal resuscitation; his Apgar scores were 5 at 1 minute and 9 at 5 minutes. The birth weight was 4.06 kg (75th to 90th percentile), the length was 54.5 cm (97th percentile), and the head circumference was 37.0 cm (75th percentile).
The patient’s mother was the youngest of 18 siblings, and 3 of her brothers died in early childhood, as a result of uncertain causes. The patient’s father was 1 of 11 children, and 2 of his brothers also died in early childhood, as a result of unknown causes. The parents were not known to be consanguineous.
Toward the end of his first 24 hours of life, the patient became tachypneic and irritable, and he had coarse breath sounds. Chest radiographs suggested right-sided pneumonia, and intravenous ampicillin and gentamicin therapy was initiated for presumed pneumonia. The patient fed very poorly the next day; by his third day of life, he had respiratory distress and stridor during feeding. He was hypotonic, with episodes of opisthotonos, and had a high-pitched cry. On his fourth day of life, he had a few episodes of apnea associated with desaturation. Subsequently, he experienced a seizure, with bicycling movements in his right foot and rhythmic tonic-clonic activity of his right upper extremity.
Physical examination on the patient’s fourth day of life was significant only for generalized hypotonia with symmetric deep tendon reflexes and the presence of the Babinski reflex. However, the patient became hypertonic and hyperreflexic a few hours later. No dysmorphic features were observed.
Blood and cerebrospinal fluid cultures yielded negative results. Electroencephalographic (EEG) recordings showed the presence of burst suppression with frequent electrographic seizure activity, mostly over the frontotemporal regions with shifting predominance. Head computed tomography performed on the fourth day of life showed loss of gray-white matter differentiation in the frontal, parietal, and occipital regions bilaterally. The right caudate head and putamen were hypodense. Brain MRI performed on the fifth day of life revealed widespread decreased diffusion throughout the entire cortex, subcortical white matter and basal ganglia. Cystic changes in the cortical and subcortical white matter were noted at 3.5 months of life (Fig 1).
Initial metabolic investigations were unrevealing, with normal bicarbonate, ammonia, and serum uric acid levels, and unremarkable plasma amino acid and urinary organic acid profiles. Of note, his methionine level was normal, and neither free cystine nor S-sulfocysteine was detected in a plasma amino acid screen.
Urinalysis performed on the sixth day of life revealed the presence of sulfites (40 mg/L) with a sulfite dipstick (EM Quant Sulfite Test Strips, EMD Chemicals, Gibbstown, NJ) and increased urinary thiosulfate levels (1.2 μmol/mg creatinine; normal: <0.4 μmol/mg creatinine). Subsequently, the urinary sulfocysteine concentration was found to be markedly elevated at 1446 μmol/g creatinine at 2 weeks of life (normal: <24 μmol/g creatinine). Plasma total homocysteine and free cystine were not detectable. Urinary xanthine and hypoxanthine levels were normal. These laboratory findings were consistent with the diagnosis of isolated sulfite oxidase deficiency.
At 12 days of age, the patient began a diet consisting of a mixture of methionine- and cysteine-free formula (XMet, XCys Analog; SHS North America, Gaithersburg, MD) with normal infant formula. Sulfites were shown to destroy thiamine in in vitro and animal experiments.2,3 Thiamine was therefore added to the patient’s regimen when he was 3 weeks of age, in the hope that the potential toxicity of sulfites could be partially ameliorated with supplementation of his thiamine stores. In addition, dextromethorphan was added to his therapeutic regimen to block possible excitotoxicity of sulfocysteine at the N-methyl-D-aspartate (NMDA) receptor. While receiving phenobarbital, the patient did not experience any additional generalized tonic-clonic seizures beyond the neonatal period, but he experienced occasional myoclonic jerks. An EEG study performed at 3 months of age showed predominantly left-sided spike and sharp wave complexes, but there were no electrographic seizures.
As indicated in the growth charts (Fig 2), the patient developed progressive microcephaly (<3rd percentile), and his development was severely delayed. At 13 months of age, he was very hypertonic in all 4 limbs, with brisk deep tendon reflexes, but his head control and truncal tone were very poor. He exhibited occasional opisthotonic posturing and repeated startle responses to loud noises. Developmentally, he was unable to roll over or sit up. He was not reaching out for objects, and there was no meaningful vocalization. An ophthalmologic examination performed at 5.5 months of age did not reveal any evidence of ectopia lentis, but a follow-up examination performed at 11 months of age showed bilateral, mild, nasal subluxation of the lenses.
B, head circumference and weight for length from birth to 13 months of age.
Genomic DNA was isolated from peripheral leukocytes, and the entire coding region of the SUOX gene (exons 2 and 3) was amplified in overlapping segments by polymerase chain reaction, with primers as described previously.4 Each segment was sequenced with an Applied Biosystems Taq DyeDeoxy Terminator cycle sequencing kit in an Applied Biosystems 3700 Prism automated sequencer (Applied Biosystems, Foster City, CA). The sequences obtained from the samples were compared with the GenBank sequence (accession number AY056018) with National Center for Biotechnology Information Blast 2 Sequences software.
The patient was homozygous for a novel 4-base pair deletion (c.2037–2040delTAGA) in exon 3 that led to a frameshift and resulted in a prematurely truncated protein. Both of the patient’s parents were heterozygous carriers for the same deletion (Fig 3).
DISCUSSION
Metabolism of Sulfur-Containing Amino Acids
The catabolism of sulfur-containing amino acids, namely, methionine and cysteine, contributes to the bulk of the sulfite load in the body. The metabolic pathway is shown in Fig 4. 1,5
The human sulfite oxidase enzyme is located in the mitochondrial intermembranous space and is involved in the transfer of electrons from sulfites into the electron transport chain via cytochrome c.6 The sulfite oxidase gene is expressed most strongly in the liver, kidney, skeletal muscle, heart, and placenta, as well as in the cerebral cortex.7
Biochemical Diagnosis of Sulfite Oxidase Deficiency
Sulfite oxidase requires a molybdenum cofactor. Deficiencies of the enzyme and the molybdenum cofactor share the following biochemical hallmarks: the presence of urinary sulfites, elevation of urinary thiosulfate levels, increased plasma and urinary S-sulfocysteine levels, and reduced plasma total homocysteine and cystine levels. However, in molybdenum cofactor deficiency, serum and urinary uric acid levels are reduced and urinary xanthine and hypoxanthine levels are increased because of a secondary deficiency in xanthine oxidase. In addition, urinary urothione, a breakdown product of the molybdenum cofactor, is absent in molybdenum cofactor deficiency but present in isolated sulfite oxidase deficiency.
Urinary sulfite is very unstable even when the urine is stored at 4°C. False-negative results can occur after storage for only 4 hours.8 Urine samples should therefore be collected on ice and tested for sulfites as soon as possible. If urinary sulfite is tested with sulfite dipsticks (EM Quant Sulfite Test Strips), then false-negative results may also occur if the pH of the urine is <6. False-positive results may occur in the presence of drugs containing a free, reactive, aliphatic sulfhydryl group, such as N-acetylcysteine, mercaptamine, and dimercaprol.9
Urinary thiosulfates and S-sulfocysteine are more stable. Urinary thiosulfate is thought to be stable for at least 3 days at room temperature and for at least 18 months when urine is frozen at –20°C.10 However, antibiotics such as cefotaxime, cefuroxime, ampicillin, and benzylpenicillin may produce false-positive results in the measurement of urinary thiosulfates.11,12 Urinary thiosulfate levels were measured using a spectrophotometric method,10 and urinary S-sulfocysteine levels were measured by tandem mass spectrometry at the Pediatric Biochemical Genetics Laboratory, Duke University Medical Center (Durham, NC).13
Plasma total homocysteine was found to be undetectable (<1.0 μmol/L) in a few cases of sulfite oxidase deficiency,14,15 although it is not known whether this finding is seen in other sulfur amino acid metabolic disorders. The lower limit of normal (ie, 2.5th percentile) for total homocysteine levels among newborns and infants varies from 2.02 μmol/L in the Canadian population to 3.8 μmol/L in the Norwegian population.16,17 Although total homocysteine levels increase with age, the lower limit of normal in the 2-month to 10-year age group in a Spanish population was found to be 3.3 μmol/L.18 Unlike the measurement of urinary thiosulfate and S-sulfocysteine levels, assessment of plasma total homocysteine levels involves a readily available test. We therefore propose that levels should be measured for every infant with intractable seizures or abnormal movements of uncertain cause and a total homocysteine level of <2.0 μmol/L should prompt additional diagnostic tests for sulfite oxidase deficiency.
Review of Published Cases
The first case of isolated sulfite oxidase deficiency was reported by Irreverre et al19 in 1967. In that report, the authors described a 30-month-old boy who was opisthotonic at birth, had a poor suck, and was hypertonic in all 4 limbs by 1 month of age, with frequent episodes of opisthotonic posturing. He never developed any motor or verbal skills. Bilateral subluxation of his lenses was observed at 1 year of age.
Sulfite, S-sulfocysteine, and thiosulfate were identified in the urine. Urinary inorganic sulfate levels were reduced. Serum uric acid levels were normal. These findings led the authors to hypothesize that the underlying metabolic defect was in the sulfite oxidase enzyme; this was subsequently confirmed by measuring sulfite oxidase activity in postmortem samples of liver, brain, and kidney from the patient.20
Since the initial case, 20 other cases of isolated sulfite oxidase deficiency have been reported. In addition, a 4.5-year-old boy had previously been reported to have isolated sulfite oxidase deficiency, on the basis of undetectable sulfite oxidase activity in skin fibroblasts and increased serum uric acid levels.21 However, urinary xanthine and hypoxanthine levels were subsequently found to be elevated, urinary urothione was undetectable, and urinary uric acid levels were normal. The authors postulated that this patient had a partial molybdenum cofactor deficiency with low levels of xanthine oxidase and sulfite oxidase activities that were undetectable with standard assays.22 Of the 21 cases that were published, urinary oxypurine (ie, xanthine and hypoxanthine) and urinary urothione levels were not measured in 5 cases.23–25 Although serum uric acid levels were reportedly normal in 4 of those 5 cases, the possibility of a partial molybdenum cofactor deficiency could not be excluded.21,22
The major clinical and biochemical findings for all reported cases are summarized in Tables 1 and 2. All of the patients were born at or near term, with birth weights ranging from 2.72 kg to 3.9 kg for the 8 patients for whom birth weights were reported.
Neurologic abnormalities were prominent in the presentation of all reported cases. Thirteen (62%) of 21 patients had seizures. Of the 8 patients for whom seizures were not reported, all had abnormal muscle tone and 6 had abnormal movements, including choreoathetosis and dystonia. The EEG results were reportedly normal for 3 (25%) of 12 patients, including 1 who presented initially with generalized convulsions.23,26,27 Significant global developmental delays were observed for 14 (78%) of 18 patients. One patient was able to attend kindergarten with good social adaptation.26 Although it is unproven, it is possible that that patient had some residual sulfite oxidase activity, leading to a milder phenotype. Two patients had nearly normal development, ie, a 4-year-old boy with a developmental quotient of 90 and a 3-year-old girl with a developmental quotient of 80.25 However, partial molybdenum coenzyme deficiency could not be excluded for those 2 patients. Ectopia lentis was reported for 9 (53%) of 17 patients; the age of detection ranged from 3 months to 3.5 years.
Although sulfite oxidase deficiency is a disease that primarily affects the central nervous system, as in our case described above, respiratory distress was part of the initial symptoms for 4 previously reported patients, all of whom presented in the first 24 hours of life.28–31
Among the 9 deceased patients, 4 died as a result of pneumonia, between the ages of 10 weeks and 4 years,23,29,32,33 1 as a result of an apneic crisis at 10 months,31 1 as a result of progressive neurologic deterioration at 12 months,34 1 as a result of probable sepsis at 31 months,19 and 2 as a result of unstated causes at 9 days and 19 months.28,29
Neuropathologic Findings
In the 3 postmortem reports that were published, all of the patients presented at birth; 1 died at the age of 19 months,29 whereas the other 2 died between the ages of 30 and 32 months.19,33 The brain weighed only 40% to 45% of the expected weight for age. The lateral and third ventricles were dilated. There was atrophy of the cerebral white matter, with small scattered cysts, as well as severe loss of axons and myelin with diffuse gliosis, although Rupar et al33 did not find any evidence of cortical demyelination in their patient.
Marked atrophy of the basal ganglia and thalami was noted. The internal capsule was affected in 2 cases29,35 but was reportedly spared in the other case.33 Myelin loss was also evident in the cerebellum. The midbrain, pons and medulla were reportedly normal except for mild atrophy of the pyramidal tracts. Similarly, the spinal cord was normal except for loss of myelination in the corticospinal tracts.
All of the pathologic lesions observed were of different ages, which raises the possibility that the damage to the central nervous system could have occurred in utero. In 1 case, the oldest lesions were surrounded by dense glial reactions, which suggests that they occurred only after the fifth month of gestation, when astrocytes had developed.35
Neuroradiologic Findings
Early in the course of the disease, cerebral edema without any evidence of ventricular dilation is seen on both computed tomographic and MRI scans.29,36 Ventriculomegaly can be seen on MRI scans as early as 12 days of age.36 Subsequently, cerebral and cerebellar atrophy, and cystic changes are observed, consistent with the neuropathologic findings described above. Other neuroradiologic changes that have been reported include hypoplasia of the corpus callosum, basal ganglia, and brainstem,30,31,33 and cystic changes and calcifications in the basal ganglia.4,29 We were unable to find any reports of diffusion-weighted imaging for patients with isolated sulfite oxidase deficiency.
Pathogenesis of Sulfite Oxidase Deficiency
The neuropathogenesis of sulfite oxidase deficiency remains unclear. It might be due to a lack of sulfates, an accumulation of sulfites, or an accumulation of the sulfur-containing compounds that form as a result of excessive sulfites.
The insulating capacity and stability of myelin are thought to be dependent on cerebrosides (also known as galactosylceramide) and sulfatides, which are sulfated cerebrosides. The synthesis of sulfatides requires sulfates; therefore, a deficiency in sulfates might lead to unstable myelin and consequent neurologic dysfunction and death. However, sulfatide analysis in a case of isolated sulfite oxidase deficiency was normal,35 and thus sulfate deficiency is unlikely to be responsible for the pathogenesis of this disorder.
In the postmortem examination of a 9-month-old boy with probable molybdenum cofactor deficiency in whom the neuropathologic lesions were similar to those seen in isolated sulfite oxidase deficiency, X-ray energy microanalysis revealed excess sulfur in the neuronal cell bodies and dendrites, and excess magnesium in the neuronal cell bodies. Because NMDA receptor activation of rat cortical neurons has previously been shown to cause an increase in intracellular magnesium concentrations, the authors postulated that neuronal injury might be attributable to either the direct toxicity of sulfites or excessive NMDA receptor activation by sulfur-containing amino acids, leading to increases in intracellular magnesium and calcium concentrations.37 However, the sulfur and magnesium concentrations in the neurons of patients with isolated sulfite oxidase deficiency have never been reported, and it is uncertain whether these compounds indeed contribute to the neuropathogenesis seen in isolated sulfite oxidase deficiency.
More recently, it was found that, when rat and mouse neuronal cell lines were exposed to sulfites in vitro, there was an increase in the production of reactive oxygen species and a reduction in intracellular ATP production.38 The authors also found that glutamate dehydrogenase in rat brain mitochondria was inhibited by sulfites; they hypothesized that this might result in an "energy crisis" at the level of the neurons, with secondary inhibition of the tricarboxylic acid cycle, leading to widespread neuronal ischemia and the neuropathologic lesions seen in sulfite oxidase deficiency.
The pathophysiology of ectopia lentis in sulfite oxidase deficiency is unknown. However, a reduction in the activity of tissue inhibitor of metalloproteinases (TIMP), with a concomitant increase in the proteolytic activity of matrix metalloproteinases (MMPs), was reported in a case of isolated ectopia lentis.39 Those authors suggested that an increase in the activity of MMPs might result in increased fibrillin degradation and hence dislocation of the lens.
The binding of TIMP to MMPs involves the formation of disulfide bonds.40 Because sulfites are thought to interact with disulfide bonds in vivo to form S-sulfonates,41 it is possible that an accumulation of sulfites might inhibit the binding of TIMP to MMPs, leading to excess MMP activity and lens subluxation. However, currently there is neither clinical nor experimental evidence for this hypothesis.
Genotype-Phenotype Correlation
The human sulfite oxidase gene (SUOX) is located on chromosome 12q13.13. At least 16 different pathogenic mutations and 1 polymorphism (2012C>G) have been reported for the gene. Most of the mutations are private, but the 479G>A mutation, resulting in R160Q substitution, occurs at a CpG hotspot in the gene and has been identified for at least 4 patients.42,43
Initial reports suggested that missense mutations might result in relatively milder phenotypes. However, at least 2 patients with missense mutations experienced severe developmental delay.43–45 In contrast, of the 6 patients whose clinical course and genotypes were reported, the 2 patients who had deletions in the sulfite oxidase gene that resulted in a truncated protein both died, one at 10 weeks of age32 and the other at 32 months of age.33
Therapeutic Interventions
The first attempt to treat isolated sulfite oxidase deficiency was reported in 1989.24 In the hope that D-penicillamine would induce the binding of sulfites to thiol to form S-sulfopenicillamine, the authors administered a single dose of D-penicillamine; however, urinary excretion of S-sulfocysteine and sulfites remained unchanged, and S-sulfopenicillamine was not detected in the urine. It was unclear whether the patient experienced any clinical improvement.
Touati et al25 treated 2 children with sulfite oxidase deficiency by using a low-protein diet with reduced intake of methionine, cystine, and taurine. The biochemical profile improved, with significant reductions in urinary thiosulfate and S-sulfocysteine levels, and both children grew well physically. Development was considered normal for one of the children and relatively mildly delayed for the other. However, the possibility that these 2 children had molybdenum cofactor deficiency could not be excluded, and molybdenum cofactor deficiency has been shown to have a better response to dietary interventions.21 In 2 other reports of dietary intervention in isolated sulfite oxidase deficiency with low-protein diets with limited methionine and cysteine, the patients became less irritable but there was no perceptible improvement in their neurologic function or development.43,45
In contrast, dietary restriction of methionine and supplementation of cysteine in a 5-month-old girl with molybdenum cofactor deficiency led to rapid clearance of urinary sulfites, improvement in neurodevelopment, and resumption of head growth 1 month after initiation of the diet. However, when dietary restrictions were abandoned 2 months later, urinary sulfites reappeared, the patient regressed developmentally, and her head stopped growing.46 The authors suggested that cysteine supplementation might be important because cysteine can form disulfide bonds with toxic sulfur-containing metabolites and promote their excretion through an alternative "salvage pathway."
Dextromethorphan inhibits NMDA receptor channels and voltage-operated calcium and sodium channels in the brain. Because S-sulfocysteine is structurally similar to glutamate and other neuroexcitatory acidic amino acids, a 3-year-old boy with molybdenum cofactor deficiency was treated with dextromethorphan.47 His seizures, which had been poorly controlled, stopped and his EEG findings showed significant improvement, although the impact of dextromethorphan on his neurodevelopment was not reported.
Prenatal Diagnosis
Prenatal diagnosis of sulfite oxidase deficiency was first demonstrated in 1996, when amniotic fluid obtained from a fetus with sulfite oxidase deficiency at 17 weeks of gestation revealed sulfocysteine, which is normally absent, and reduced levels of half-cystine, although the authors cautioned that sulfocysteine is relatively unstable and might lead to false-negative results.48 In 2002, Johnson et al32 undertook prenatal diagnosis with a combination of mutation analysis and enzymatic assay. The index case was homozygous for a 4-base pair deletion in the SUOX gene. Sulfite oxidase activity in the chorionic villus sampling (CVS) sample of a subsequent pregnancy was undetectable, and mutation analysis of DNA from the fetus, obtained through CVS, revealed that the fetus was homozygous for the same deletion. In the next pregnancy, the sulfite oxidase activity in the CVS sample was slightly reduced, and mutation analysis of DNA obtained through CVS confirmed that the fetus was a heterozygous carrier of the mutation. The latter fetus was carried to term, and a healthy child was born.
Although sulfite oxidase activity can also be measured in cultured amniocytes, the concentration of the enzyme in amniocytes is thought to be low, and a large number of cells would be required.49
CONCLUSIONS
Isolated sulfite oxidase deficiency is a devastating illness for which no effective long-term therapy has been found. Although many authors think that accumulation of sulfites or their metabolites is responsible for the pathogenesis of the disease, the possibility that a deficiency of inorganic sulfates might play a major etiologic role has not been explored adequately.
Sulfite oxidase and molybdenum cofactor deficiencies should be part of the differential diagnoses for all young infants who present with seizures or alterations in muscle tone of uncertain cause. Determination of plasma total homocysteine levels would be a good initial screening test, because it is commonly performed in many standard clinical chemistry laboratories and, to the best of our knowledge, sulfite oxidase and molybdenum cofactor deficiencies are the only conditions in which plasma total homocysteine is undetectable.14,15 If the total homocysteine level is abnormally low or if there is a strong clinical suspicion of sulfite oxidase deficiency, then the presence of urinary sulfites can be determined with sulfite dipstick testing of fresh urine, and urinary thiosulfate and S-sulfocysteine levels should be assayed. Serum uric acid and urinary oxypurine levels should also be measured, to distinguish isolated sulfite oxidase deficiency from molybdenum cofactor deficiency.
Although it has never been attempted in isolated sulfite oxidase deficiency, the encouraging response to dietary restriction of methionine with supplementation of cysteine in a case of molybdenum cofactor deficiency46 makes that a potentially attractive therapeutic option.
ACKNOWLEDGMENTS
This work was supported in part by the Mary L. Efron Fund at Massachusetts General Hospital.
We are indebted to Jean L. Johnson, PhD (Department of Biochemistry, Duke University), for providing us with the primers and sequences required for mutation analysis of the sulfite oxidase gene. We are grateful to Roseann Mandell, BA, for expert technical assistance and advice. We thank the staff of the NICU at Massachusetts General Hospital for their excellent care of the patient during his early days.
FOOTNOTES
Accepted Dec 7, 2004.
No conflict of interest declared.
REFERENCES
Johnson JL, Duran M. Molybdenum cofactor deficiency and isolated sulfite oxidase deficiency. In: Scriver CR, Sly WS, Childs B, et al, eds. The Metabolic and Molecular Bases of Inherited Disease. 8th ed. New York, NY: McGraw-Hill Professional; 2001:3163–3177
Williams RR, Waterman RE, Keresztesy JC, Buchman ER. Studies of crystalline vitamin B1. III. Cleavage of vitamin with sulfite. J Am Chem Soc. 1935;57 :536 –537
Til HP, Feron VJ, De Groot AP. The toxicity of sulphite. I. Long-term feeding and multigeneration studies in rats. Food Cosmet Toxicol. 1972;10 :291 –310
Garrett RM, Johnson JL, Graf TN, Feigenbaum A, Rajagopalan KV. Human sulfite oxidase R160Q: identification of the mutation in a sulfite oxidase-deficient patient and expression and characterization of the mutant enzyme. Proc Natl Acad Sci USA. 1998;95 :6394 –6398
Griffith OW. Mammalian sulfur amino acid metabolism: an overview. Methods Enzymol. 1987;143 :366 –376
Oshino N, Chance B. The properties of sulfite oxidation in perfused rat liver; interaction of sulfite oxidase with the mitochondrial respiratory chain. Arch Biochem Biophys. 1975;170 :514 –528
Woo WH, Yang H, Wong KP, Halliwell B. Sulphite oxidase gene expression in human brain and in other human and rat tissues. Biochem Biophys Res Commun. 2003;305 :619 –623
Kutter D, Humbel R. Screening for sulfite oxidase deficiency. Clin Chim Acta. 1969;24 :211 –214
Wadman SK, Cats BP, de Bree PK. Sulfite oxidase deficiency and the detection of urinary sulfite. Eur J Pediatr. 1983;141 :62 –63
Shih VE, Carney MM, Mandell R. A simple screening test for sulfite oxidase deficiency: detection of urinary thiosulfate by a modification of Sorbo’s method. Clin Chim Acta. 1979;95 :143 –145
Mann G, Kirk JM. Antibiotic interference in urinary thiosulphate measurements. J Inherit Metab Dis. 1994;17 :120 –121
Shih VE. Ampicillin interference with test for sulfite oxidase deficiency. Clin Chem. 1983;29 :211 –212
Millington DS, Stevens RD, Rajagopalan KV, Johnson JL. Quantification of S-sulphocysteine, a marker for sulphite oxidase deficiency, by tandem mass spectrometry . J Inherit Metab Dis. 2000;23 :77
Graf WD, Oleinik OE, Jack RM, Weiss AH, Johnson JL. Ahomocysteinemia in molybdenum cofactor deficiency. Neurology. 1998;51 :860 –862
Sass JO, Nakanishi T, Sato T, Shimizu A. New approaches towards laboratory diagnosis of isolated sulphite oxidase deficiency. Ann Clin Biochem. 2004;41 :157 –159
Infante-Rivard C, Rivard GE, Yotov WV, Théret Y. Perinatal reference intervals for plasma homocysteine and factors influencing its concentration. Clin Chem. 2002;48 :1100 –1102
Refsum H, Grindflek AW, Ueland PM, et al. Screening for serum total homocysteine in newborn children. Clin Chem. 2004;50 :1769 –1784
Vilaseca MA, Moyano D, Ferrer I, Artuch R. Total homocysteine in pediatric patients. Clin Chem. 1997;43 :690 –692
Irreverre F, Mudd SH, Heizer WD, Laster L. Sulfite oxidase deficiency: studies of a patient with mental retardation, dislocated ocular lenses, and abnormal urinary excretion of S-sulfo-L-cysteine, sulfite, and thiosulfate. Biochem Med. 1967;1 :187 –217
Mudd SH, Irreverre F, Laster L. Sulfite oxidase deficiency in man: demonstration of the enzymatic defect. Science. 1967;156 :1599 –1602
Shih VE, Abroms IF, Johnson JL, et al. Sulfite oxidase deficiency: biochemical and clinical investigations of a hereditary metabolic disorder in sulfur metabolism. N Engl J Med. 1977;297 :1022 –1028
Johnson JL, Wuebbens MM, Mandell R, Shih VE. Molybdenum cofactor deficiency in a patient previously characterized as deficient in sulfite oxidase. Biochem Med Metab Biol. 1988;40 :86 –93
Goh A, Lim KW. Sulphite oxidase deficiency: a report of two siblings. Singapore Med J. 1997;38 :391 –394
Tardy P, Parvy P, Charpentier C, Bonnefont JP, Saudubray JM, Kamoun P. Attempt at therapy in sulphite oxidase deficiency. J Inherit Metab Dis. 1989;12 :94 –95
Touati G, Rusthoven E, Depondt E, et al. Dietary therapy in two patients with a mild form of sulphite oxidase deficiency: evidence for clinical and biological improvement. J Inherit Metab Dis. 2000;23 :45 –53
Barbot C, Martins E, Vilarinho L, Dorche C, Cardoso ML. A mild form of infantile isolated sulphite oxidase deficiency. Neuropediatrics. 1995;26 :322 –324
van der Klei-van Moorsel JM, Smit LM, Brockstedt M, Jakobs C, Dorche C, Duran M. Infantile isolated sulphite oxidase deficiency: report of a case with negative sulphite test and normal sulphate excretion. Eur J Pediatr. 1991;150 :196 –197
Vianey-Liaud C, Desjacques P, Gaulme J, et al. A new case of isolated sulphite oxidase deficiency with rapid fatal outcome. J Inherit Metab Dis. 1988;11 :425 –426
Brown GK, Scholem RD, Croll HB, Wraith JE, McGill JJ. Sulfite oxidase deficiency: clinical, neuroradiologic, and biochemical features in two new patients. Neurology. 1989;39 :252 –257
Edwards MC, Johnson JL, Marriage B, et al. Isolated sulfite oxidase deficiency: review of two cases in one family. Ophthalmology. 1999;106 :1957 –1961
Schiaffino MC, Fantasia AR, Minniti G, Caruso U, Carnevale F, Cerone R. Isolated sulphite oxidase deficiency: clinical and biochemical features in an Italian patient. J Inherit Metab Dis. 2004;27 :101 –102
Johnson JL, Rajagopalan KV, Renier WO, van der Burgt I, Ruitenbeek W. Isolated sulfite oxidase deficiency: mutation analysis and DNA-based prenatal diagnosis. Prenat Diagn. 2002;22 :433 –436
Rupar CA, Gillett J, Gordon BA, et al. Isolated sulfite oxidase deficiency. Neuropediatrics. 1996;27 :299 –304
Amiel J, Gagey V, Rabier D, et al. Sulfite oxidase deficiency presenting as Leigh syndrome [in French]. Arch Pédiatr. 1994;1 :1023 –1027
Rosenblum WI. Neuropathologic changes in a case of sulfite oxidase deficiency. Neurology. 1968;18 :1187 –1196
Dublin AB, Hald JK, Wootton-Gorges SL. Isolated sulfite oxidase deficiency: MR imaging features. AJNR Am J Neuroradiol. 2002;23 :484 –485
Salman MS, Ackerley C, Senger C, Becker L. New insights into the neuropathogenesis of molybdenum cofactor deficiency. Can J Neurol Sci. 2002;29 :91 –96
Zhang X, Vincent AS, Halliwell B, Wong KP. A mechanism of sulfite neurotoxicity. J Biol Chem. 2004;279 :43035 –43045
Sachdev NH, Coroneo MT, Wakefield D, Hennessy MP. Isolated ectopia lentis: potential role of matrix metalloproteinases in fibrillin degradation. Arch Ophthalmol. 2004;122 :111 –114
Brew K, Dinakarpandian D, Nagase H. Tissue inhibitor of metalloproteinases: evolution, structure and function. Biochim Biophys Acta. 2000;1477 :267 –283
Gunnison AF, Benton AW. Sulfur dioxide: Sulfite. Interaction with mammalian serum and plasma. Arch Environ Health. 1971;22 :381 –388
Johnson JL, Coyne KE, Garrett RM, et al. Isolated sulfite oxidase deficiency: identification of 12 novel SUOX mutations in 10 patients. Hum Mutat. 2002;20 :74
Lee HF, Mak BS, Chi CS, Tsai CR, Chen CH, Shu SG. A novel mutation in neonatal isolated sulphite oxidase deficiency. Neuropediatrics. 2002;33 :174 –179
Lam CW, Li CK, Lai CK, et al. DNA-based diagnosis of isolated sulfite oxidase deficiency by denaturing high-performance liquid chromatography. Mol Genet Metab. 2002;75 :91 –95
Chan KY, Li CK, Lai CK, Ng SF, Chan AY. Infantile isolated sulphite oxidase deficiency in a Chinese family: a rare neurodegenerative disorder. Hong Kong Med J. 2002;8 :279 –282
Boles RG, Ment LR, Meyn MS, Horwich AL, Kratz LE, Rinaldo P. Short-term response to dietary therapy in molybdenum cofactor deficiency. Ann Neurol. 1993;34 :742 –744
Kurlemann G, Debus O, Schuierer G. Dextromethorphan in molybdenum cofactor deficiency. Eur J Pediatr. 1996;155 :422 –423
Rabier D, Chadefaux-Vekemans B, Oury JF, et al. Gestational age-related reference values for amniotic fluid amino acids: a useful tool for prenatal diagnosis of aminoacidopathies. Prenat Diagn. 1996;16 :623 –628
Johnson JL. Prenatal diagnosis of molybdenum cofactor deficiency and isolated sulfite oxidase deficiency. Prenat Diagn. 2003;23 :6 –8
Vilarinho L, Alves JR, Dorche C, Chadefaux B, Parvy P. Citrullinaemia and isolated sulphite oxidase deficiency in two siblings. J Inherit Metab Dis. 1994;17 :638 –639(Wen-Hann Tan, MRCPCH, Flo)
Pediatric Neurology Unit Amino Acid Disorders Laboratory, Department of Neurology
Nutrition Services Division of Neuroradiology, Massachusetts General Hospital, Boston, Massachusetts
ABSTRACT
Isolated sulfite oxidase deficiency is a rare but devastating neurologic disease that usually presents in early infancy with seizures and alterations in muscle tone. Only 21 cases have been reported in the literature. We report a case of a newborn infant boy with isolated sulfite oxidase deficiency who presented with generalized seizures on his fourth day of life. Plasma total homocysteine was not detectable. Urinary sulfite, thiosulfate, and S-sulfocysteine levels were elevated. The patient began a low-methionine and low-cysteine diet and was treated with thiamine and dextromethorphan. However, he became increasingly microcephalic and was severely developmentally delayed. Mutation analysis of the sulfite oxidase gene revealed that the patient was homozygous for a novel 4-base pair deletion, and both of his parents were found to be heterozygous carriers of the same deletion. We reviewed the clinical, biochemical, neuroradiologic, and neuropathologic features in all published cases of isolated sulfite oxidase deficiency. Seizures or abnormal movements were prominent features in all cases. Developmental delays were reported in 17 cases. Ectopia lentis was detected in 9 cases. Clinical improvement with dietary therapy was seen in only 2 patients, both of whom presented after the age of 6 months and had relatively mild developmental delays. Plasma or urinary S-sulfocysteine levels were elevated in all cases. Urinary sulfite was detected in all except 1 case. Cerebral atrophy and cystic encephalomalacia were observed with neuroradiologic imaging and were noted in all 3 postmortem reports of isolated sulfite oxidase deficiency. The main alternative in the differential diagnosis of isolated sulfite oxidase deficiency is molybdenum cofactor deficiency.
Key Words: sulfite sulfite oxidase SUOX gene molybdenum cofactor inborn error of metabolism
Abbreviations: CVS, chorionic villus sampling EEG, electroencephalographic MMP, matrix metalloproteinase NMDA, N-methyl-D-aspartate TIMP, tissue inhibitor of metalloproteinases
Sulfite oxidase deficiency is a neurometabolic disease that results in severe developmental delay and premature death. Sulfite oxidase (EC 1.8.3.1) is 1 of 3 enzymes in humans that requires molybdenum as a cofactor, the other 2 being xanthine oxidase and aldehyde oxidase.
Sulfite oxidase oxidizes potentially toxic sulfites to nontoxic sulfates in the final step of the sulfur amino acid metabolism pathway. Xanthine oxidase catalyzes the hydroxylation of hypoxanthine to xanthine and thence to uric acid. Aldehyde oxidase is involved in the hydroxylation of hypoxanthine to xanthine but not to uric acid. Xanthine oxidase deficiency (also known as xanthinuria) may either lead to the deposition of xanthine in kidneys and muscle, resulting in renal calculi and myopathy, or be totally asymptomatic.1 Although isolated aldehyde oxidase deficiency has not been reported for humans, it is thought to be relatively benign.1 Molybdenum cofactor deficiency would therefore be the main differential diagnosis that should be considered.
A number of patients with sulfite oxidase deficiency have now been reported in the literature, all of whom had relatively dismal outcomes. In this article, we present a case of a newborn with sulfite oxidase deficiency and a previously unreported mutation in the sulfite oxidase gene. We also review the metabolism of sulfur amino acids, as well as all previously published cases of sulfite oxidase deficiency.
CASE REPORT
A term infant boy was born to a 21-year-old primigravid mother from Colombia, after a pregnancy that was complicated by 1 episode each of urinary tract infection and pyelonephritis, both of which were treated with orally administered antibiotics only. The patient required minimal resuscitation; his Apgar scores were 5 at 1 minute and 9 at 5 minutes. The birth weight was 4.06 kg (75th to 90th percentile), the length was 54.5 cm (97th percentile), and the head circumference was 37.0 cm (75th percentile).
The patient’s mother was the youngest of 18 siblings, and 3 of her brothers died in early childhood, as a result of uncertain causes. The patient’s father was 1 of 11 children, and 2 of his brothers also died in early childhood, as a result of unknown causes. The parents were not known to be consanguineous.
Toward the end of his first 24 hours of life, the patient became tachypneic and irritable, and he had coarse breath sounds. Chest radiographs suggested right-sided pneumonia, and intravenous ampicillin and gentamicin therapy was initiated for presumed pneumonia. The patient fed very poorly the next day; by his third day of life, he had respiratory distress and stridor during feeding. He was hypotonic, with episodes of opisthotonos, and had a high-pitched cry. On his fourth day of life, he had a few episodes of apnea associated with desaturation. Subsequently, he experienced a seizure, with bicycling movements in his right foot and rhythmic tonic-clonic activity of his right upper extremity.
Physical examination on the patient’s fourth day of life was significant only for generalized hypotonia with symmetric deep tendon reflexes and the presence of the Babinski reflex. However, the patient became hypertonic and hyperreflexic a few hours later. No dysmorphic features were observed.
Blood and cerebrospinal fluid cultures yielded negative results. Electroencephalographic (EEG) recordings showed the presence of burst suppression with frequent electrographic seizure activity, mostly over the frontotemporal regions with shifting predominance. Head computed tomography performed on the fourth day of life showed loss of gray-white matter differentiation in the frontal, parietal, and occipital regions bilaterally. The right caudate head and putamen were hypodense. Brain MRI performed on the fifth day of life revealed widespread decreased diffusion throughout the entire cortex, subcortical white matter and basal ganglia. Cystic changes in the cortical and subcortical white matter were noted at 3.5 months of life (Fig 1).
Initial metabolic investigations were unrevealing, with normal bicarbonate, ammonia, and serum uric acid levels, and unremarkable plasma amino acid and urinary organic acid profiles. Of note, his methionine level was normal, and neither free cystine nor S-sulfocysteine was detected in a plasma amino acid screen.
Urinalysis performed on the sixth day of life revealed the presence of sulfites (40 mg/L) with a sulfite dipstick (EM Quant Sulfite Test Strips, EMD Chemicals, Gibbstown, NJ) and increased urinary thiosulfate levels (1.2 μmol/mg creatinine; normal: <0.4 μmol/mg creatinine). Subsequently, the urinary sulfocysteine concentration was found to be markedly elevated at 1446 μmol/g creatinine at 2 weeks of life (normal: <24 μmol/g creatinine). Plasma total homocysteine and free cystine were not detectable. Urinary xanthine and hypoxanthine levels were normal. These laboratory findings were consistent with the diagnosis of isolated sulfite oxidase deficiency.
At 12 days of age, the patient began a diet consisting of a mixture of methionine- and cysteine-free formula (XMet, XCys Analog; SHS North America, Gaithersburg, MD) with normal infant formula. Sulfites were shown to destroy thiamine in in vitro and animal experiments.2,3 Thiamine was therefore added to the patient’s regimen when he was 3 weeks of age, in the hope that the potential toxicity of sulfites could be partially ameliorated with supplementation of his thiamine stores. In addition, dextromethorphan was added to his therapeutic regimen to block possible excitotoxicity of sulfocysteine at the N-methyl-D-aspartate (NMDA) receptor. While receiving phenobarbital, the patient did not experience any additional generalized tonic-clonic seizures beyond the neonatal period, but he experienced occasional myoclonic jerks. An EEG study performed at 3 months of age showed predominantly left-sided spike and sharp wave complexes, but there were no electrographic seizures.
As indicated in the growth charts (Fig 2), the patient developed progressive microcephaly (<3rd percentile), and his development was severely delayed. At 13 months of age, he was very hypertonic in all 4 limbs, with brisk deep tendon reflexes, but his head control and truncal tone were very poor. He exhibited occasional opisthotonic posturing and repeated startle responses to loud noises. Developmentally, he was unable to roll over or sit up. He was not reaching out for objects, and there was no meaningful vocalization. An ophthalmologic examination performed at 5.5 months of age did not reveal any evidence of ectopia lentis, but a follow-up examination performed at 11 months of age showed bilateral, mild, nasal subluxation of the lenses.
B, head circumference and weight for length from birth to 13 months of age.
Genomic DNA was isolated from peripheral leukocytes, and the entire coding region of the SUOX gene (exons 2 and 3) was amplified in overlapping segments by polymerase chain reaction, with primers as described previously.4 Each segment was sequenced with an Applied Biosystems Taq DyeDeoxy Terminator cycle sequencing kit in an Applied Biosystems 3700 Prism automated sequencer (Applied Biosystems, Foster City, CA). The sequences obtained from the samples were compared with the GenBank sequence (accession number AY056018) with National Center for Biotechnology Information Blast 2 Sequences software.
The patient was homozygous for a novel 4-base pair deletion (c.2037–2040delTAGA) in exon 3 that led to a frameshift and resulted in a prematurely truncated protein. Both of the patient’s parents were heterozygous carriers for the same deletion (Fig 3).
DISCUSSION
Metabolism of Sulfur-Containing Amino Acids
The catabolism of sulfur-containing amino acids, namely, methionine and cysteine, contributes to the bulk of the sulfite load in the body. The metabolic pathway is shown in Fig 4. 1,5
The human sulfite oxidase enzyme is located in the mitochondrial intermembranous space and is involved in the transfer of electrons from sulfites into the electron transport chain via cytochrome c.6 The sulfite oxidase gene is expressed most strongly in the liver, kidney, skeletal muscle, heart, and placenta, as well as in the cerebral cortex.7
Biochemical Diagnosis of Sulfite Oxidase Deficiency
Sulfite oxidase requires a molybdenum cofactor. Deficiencies of the enzyme and the molybdenum cofactor share the following biochemical hallmarks: the presence of urinary sulfites, elevation of urinary thiosulfate levels, increased plasma and urinary S-sulfocysteine levels, and reduced plasma total homocysteine and cystine levels. However, in molybdenum cofactor deficiency, serum and urinary uric acid levels are reduced and urinary xanthine and hypoxanthine levels are increased because of a secondary deficiency in xanthine oxidase. In addition, urinary urothione, a breakdown product of the molybdenum cofactor, is absent in molybdenum cofactor deficiency but present in isolated sulfite oxidase deficiency.
Urinary sulfite is very unstable even when the urine is stored at 4°C. False-negative results can occur after storage for only 4 hours.8 Urine samples should therefore be collected on ice and tested for sulfites as soon as possible. If urinary sulfite is tested with sulfite dipsticks (EM Quant Sulfite Test Strips), then false-negative results may also occur if the pH of the urine is <6. False-positive results may occur in the presence of drugs containing a free, reactive, aliphatic sulfhydryl group, such as N-acetylcysteine, mercaptamine, and dimercaprol.9
Urinary thiosulfates and S-sulfocysteine are more stable. Urinary thiosulfate is thought to be stable for at least 3 days at room temperature and for at least 18 months when urine is frozen at –20°C.10 However, antibiotics such as cefotaxime, cefuroxime, ampicillin, and benzylpenicillin may produce false-positive results in the measurement of urinary thiosulfates.11,12 Urinary thiosulfate levels were measured using a spectrophotometric method,10 and urinary S-sulfocysteine levels were measured by tandem mass spectrometry at the Pediatric Biochemical Genetics Laboratory, Duke University Medical Center (Durham, NC).13
Plasma total homocysteine was found to be undetectable (<1.0 μmol/L) in a few cases of sulfite oxidase deficiency,14,15 although it is not known whether this finding is seen in other sulfur amino acid metabolic disorders. The lower limit of normal (ie, 2.5th percentile) for total homocysteine levels among newborns and infants varies from 2.02 μmol/L in the Canadian population to 3.8 μmol/L in the Norwegian population.16,17 Although total homocysteine levels increase with age, the lower limit of normal in the 2-month to 10-year age group in a Spanish population was found to be 3.3 μmol/L.18 Unlike the measurement of urinary thiosulfate and S-sulfocysteine levels, assessment of plasma total homocysteine levels involves a readily available test. We therefore propose that levels should be measured for every infant with intractable seizures or abnormal movements of uncertain cause and a total homocysteine level of <2.0 μmol/L should prompt additional diagnostic tests for sulfite oxidase deficiency.
Review of Published Cases
The first case of isolated sulfite oxidase deficiency was reported by Irreverre et al19 in 1967. In that report, the authors described a 30-month-old boy who was opisthotonic at birth, had a poor suck, and was hypertonic in all 4 limbs by 1 month of age, with frequent episodes of opisthotonic posturing. He never developed any motor or verbal skills. Bilateral subluxation of his lenses was observed at 1 year of age.
Sulfite, S-sulfocysteine, and thiosulfate were identified in the urine. Urinary inorganic sulfate levels were reduced. Serum uric acid levels were normal. These findings led the authors to hypothesize that the underlying metabolic defect was in the sulfite oxidase enzyme; this was subsequently confirmed by measuring sulfite oxidase activity in postmortem samples of liver, brain, and kidney from the patient.20
Since the initial case, 20 other cases of isolated sulfite oxidase deficiency have been reported. In addition, a 4.5-year-old boy had previously been reported to have isolated sulfite oxidase deficiency, on the basis of undetectable sulfite oxidase activity in skin fibroblasts and increased serum uric acid levels.21 However, urinary xanthine and hypoxanthine levels were subsequently found to be elevated, urinary urothione was undetectable, and urinary uric acid levels were normal. The authors postulated that this patient had a partial molybdenum cofactor deficiency with low levels of xanthine oxidase and sulfite oxidase activities that were undetectable with standard assays.22 Of the 21 cases that were published, urinary oxypurine (ie, xanthine and hypoxanthine) and urinary urothione levels were not measured in 5 cases.23–25 Although serum uric acid levels were reportedly normal in 4 of those 5 cases, the possibility of a partial molybdenum cofactor deficiency could not be excluded.21,22
The major clinical and biochemical findings for all reported cases are summarized in Tables 1 and 2. All of the patients were born at or near term, with birth weights ranging from 2.72 kg to 3.9 kg for the 8 patients for whom birth weights were reported.
Neurologic abnormalities were prominent in the presentation of all reported cases. Thirteen (62%) of 21 patients had seizures. Of the 8 patients for whom seizures were not reported, all had abnormal muscle tone and 6 had abnormal movements, including choreoathetosis and dystonia. The EEG results were reportedly normal for 3 (25%) of 12 patients, including 1 who presented initially with generalized convulsions.23,26,27 Significant global developmental delays were observed for 14 (78%) of 18 patients. One patient was able to attend kindergarten with good social adaptation.26 Although it is unproven, it is possible that that patient had some residual sulfite oxidase activity, leading to a milder phenotype. Two patients had nearly normal development, ie, a 4-year-old boy with a developmental quotient of 90 and a 3-year-old girl with a developmental quotient of 80.25 However, partial molybdenum coenzyme deficiency could not be excluded for those 2 patients. Ectopia lentis was reported for 9 (53%) of 17 patients; the age of detection ranged from 3 months to 3.5 years.
Although sulfite oxidase deficiency is a disease that primarily affects the central nervous system, as in our case described above, respiratory distress was part of the initial symptoms for 4 previously reported patients, all of whom presented in the first 24 hours of life.28–31
Among the 9 deceased patients, 4 died as a result of pneumonia, between the ages of 10 weeks and 4 years,23,29,32,33 1 as a result of an apneic crisis at 10 months,31 1 as a result of progressive neurologic deterioration at 12 months,34 1 as a result of probable sepsis at 31 months,19 and 2 as a result of unstated causes at 9 days and 19 months.28,29
Neuropathologic Findings
In the 3 postmortem reports that were published, all of the patients presented at birth; 1 died at the age of 19 months,29 whereas the other 2 died between the ages of 30 and 32 months.19,33 The brain weighed only 40% to 45% of the expected weight for age. The lateral and third ventricles were dilated. There was atrophy of the cerebral white matter, with small scattered cysts, as well as severe loss of axons and myelin with diffuse gliosis, although Rupar et al33 did not find any evidence of cortical demyelination in their patient.
Marked atrophy of the basal ganglia and thalami was noted. The internal capsule was affected in 2 cases29,35 but was reportedly spared in the other case.33 Myelin loss was also evident in the cerebellum. The midbrain, pons and medulla were reportedly normal except for mild atrophy of the pyramidal tracts. Similarly, the spinal cord was normal except for loss of myelination in the corticospinal tracts.
All of the pathologic lesions observed were of different ages, which raises the possibility that the damage to the central nervous system could have occurred in utero. In 1 case, the oldest lesions were surrounded by dense glial reactions, which suggests that they occurred only after the fifth month of gestation, when astrocytes had developed.35
Neuroradiologic Findings
Early in the course of the disease, cerebral edema without any evidence of ventricular dilation is seen on both computed tomographic and MRI scans.29,36 Ventriculomegaly can be seen on MRI scans as early as 12 days of age.36 Subsequently, cerebral and cerebellar atrophy, and cystic changes are observed, consistent with the neuropathologic findings described above. Other neuroradiologic changes that have been reported include hypoplasia of the corpus callosum, basal ganglia, and brainstem,30,31,33 and cystic changes and calcifications in the basal ganglia.4,29 We were unable to find any reports of diffusion-weighted imaging for patients with isolated sulfite oxidase deficiency.
Pathogenesis of Sulfite Oxidase Deficiency
The neuropathogenesis of sulfite oxidase deficiency remains unclear. It might be due to a lack of sulfates, an accumulation of sulfites, or an accumulation of the sulfur-containing compounds that form as a result of excessive sulfites.
The insulating capacity and stability of myelin are thought to be dependent on cerebrosides (also known as galactosylceramide) and sulfatides, which are sulfated cerebrosides. The synthesis of sulfatides requires sulfates; therefore, a deficiency in sulfates might lead to unstable myelin and consequent neurologic dysfunction and death. However, sulfatide analysis in a case of isolated sulfite oxidase deficiency was normal,35 and thus sulfate deficiency is unlikely to be responsible for the pathogenesis of this disorder.
In the postmortem examination of a 9-month-old boy with probable molybdenum cofactor deficiency in whom the neuropathologic lesions were similar to those seen in isolated sulfite oxidase deficiency, X-ray energy microanalysis revealed excess sulfur in the neuronal cell bodies and dendrites, and excess magnesium in the neuronal cell bodies. Because NMDA receptor activation of rat cortical neurons has previously been shown to cause an increase in intracellular magnesium concentrations, the authors postulated that neuronal injury might be attributable to either the direct toxicity of sulfites or excessive NMDA receptor activation by sulfur-containing amino acids, leading to increases in intracellular magnesium and calcium concentrations.37 However, the sulfur and magnesium concentrations in the neurons of patients with isolated sulfite oxidase deficiency have never been reported, and it is uncertain whether these compounds indeed contribute to the neuropathogenesis seen in isolated sulfite oxidase deficiency.
More recently, it was found that, when rat and mouse neuronal cell lines were exposed to sulfites in vitro, there was an increase in the production of reactive oxygen species and a reduction in intracellular ATP production.38 The authors also found that glutamate dehydrogenase in rat brain mitochondria was inhibited by sulfites; they hypothesized that this might result in an "energy crisis" at the level of the neurons, with secondary inhibition of the tricarboxylic acid cycle, leading to widespread neuronal ischemia and the neuropathologic lesions seen in sulfite oxidase deficiency.
The pathophysiology of ectopia lentis in sulfite oxidase deficiency is unknown. However, a reduction in the activity of tissue inhibitor of metalloproteinases (TIMP), with a concomitant increase in the proteolytic activity of matrix metalloproteinases (MMPs), was reported in a case of isolated ectopia lentis.39 Those authors suggested that an increase in the activity of MMPs might result in increased fibrillin degradation and hence dislocation of the lens.
The binding of TIMP to MMPs involves the formation of disulfide bonds.40 Because sulfites are thought to interact with disulfide bonds in vivo to form S-sulfonates,41 it is possible that an accumulation of sulfites might inhibit the binding of TIMP to MMPs, leading to excess MMP activity and lens subluxation. However, currently there is neither clinical nor experimental evidence for this hypothesis.
Genotype-Phenotype Correlation
The human sulfite oxidase gene (SUOX) is located on chromosome 12q13.13. At least 16 different pathogenic mutations and 1 polymorphism (2012C>G) have been reported for the gene. Most of the mutations are private, but the 479G>A mutation, resulting in R160Q substitution, occurs at a CpG hotspot in the gene and has been identified for at least 4 patients.42,43
Initial reports suggested that missense mutations might result in relatively milder phenotypes. However, at least 2 patients with missense mutations experienced severe developmental delay.43–45 In contrast, of the 6 patients whose clinical course and genotypes were reported, the 2 patients who had deletions in the sulfite oxidase gene that resulted in a truncated protein both died, one at 10 weeks of age32 and the other at 32 months of age.33
Therapeutic Interventions
The first attempt to treat isolated sulfite oxidase deficiency was reported in 1989.24 In the hope that D-penicillamine would induce the binding of sulfites to thiol to form S-sulfopenicillamine, the authors administered a single dose of D-penicillamine; however, urinary excretion of S-sulfocysteine and sulfites remained unchanged, and S-sulfopenicillamine was not detected in the urine. It was unclear whether the patient experienced any clinical improvement.
Touati et al25 treated 2 children with sulfite oxidase deficiency by using a low-protein diet with reduced intake of methionine, cystine, and taurine. The biochemical profile improved, with significant reductions in urinary thiosulfate and S-sulfocysteine levels, and both children grew well physically. Development was considered normal for one of the children and relatively mildly delayed for the other. However, the possibility that these 2 children had molybdenum cofactor deficiency could not be excluded, and molybdenum cofactor deficiency has been shown to have a better response to dietary interventions.21 In 2 other reports of dietary intervention in isolated sulfite oxidase deficiency with low-protein diets with limited methionine and cysteine, the patients became less irritable but there was no perceptible improvement in their neurologic function or development.43,45
In contrast, dietary restriction of methionine and supplementation of cysteine in a 5-month-old girl with molybdenum cofactor deficiency led to rapid clearance of urinary sulfites, improvement in neurodevelopment, and resumption of head growth 1 month after initiation of the diet. However, when dietary restrictions were abandoned 2 months later, urinary sulfites reappeared, the patient regressed developmentally, and her head stopped growing.46 The authors suggested that cysteine supplementation might be important because cysteine can form disulfide bonds with toxic sulfur-containing metabolites and promote their excretion through an alternative "salvage pathway."
Dextromethorphan inhibits NMDA receptor channels and voltage-operated calcium and sodium channels in the brain. Because S-sulfocysteine is structurally similar to glutamate and other neuroexcitatory acidic amino acids, a 3-year-old boy with molybdenum cofactor deficiency was treated with dextromethorphan.47 His seizures, which had been poorly controlled, stopped and his EEG findings showed significant improvement, although the impact of dextromethorphan on his neurodevelopment was not reported.
Prenatal Diagnosis
Prenatal diagnosis of sulfite oxidase deficiency was first demonstrated in 1996, when amniotic fluid obtained from a fetus with sulfite oxidase deficiency at 17 weeks of gestation revealed sulfocysteine, which is normally absent, and reduced levels of half-cystine, although the authors cautioned that sulfocysteine is relatively unstable and might lead to false-negative results.48 In 2002, Johnson et al32 undertook prenatal diagnosis with a combination of mutation analysis and enzymatic assay. The index case was homozygous for a 4-base pair deletion in the SUOX gene. Sulfite oxidase activity in the chorionic villus sampling (CVS) sample of a subsequent pregnancy was undetectable, and mutation analysis of DNA from the fetus, obtained through CVS, revealed that the fetus was homozygous for the same deletion. In the next pregnancy, the sulfite oxidase activity in the CVS sample was slightly reduced, and mutation analysis of DNA obtained through CVS confirmed that the fetus was a heterozygous carrier of the mutation. The latter fetus was carried to term, and a healthy child was born.
Although sulfite oxidase activity can also be measured in cultured amniocytes, the concentration of the enzyme in amniocytes is thought to be low, and a large number of cells would be required.49
CONCLUSIONS
Isolated sulfite oxidase deficiency is a devastating illness for which no effective long-term therapy has been found. Although many authors think that accumulation of sulfites or their metabolites is responsible for the pathogenesis of the disease, the possibility that a deficiency of inorganic sulfates might play a major etiologic role has not been explored adequately.
Sulfite oxidase and molybdenum cofactor deficiencies should be part of the differential diagnoses for all young infants who present with seizures or alterations in muscle tone of uncertain cause. Determination of plasma total homocysteine levels would be a good initial screening test, because it is commonly performed in many standard clinical chemistry laboratories and, to the best of our knowledge, sulfite oxidase and molybdenum cofactor deficiencies are the only conditions in which plasma total homocysteine is undetectable.14,15 If the total homocysteine level is abnormally low or if there is a strong clinical suspicion of sulfite oxidase deficiency, then the presence of urinary sulfites can be determined with sulfite dipstick testing of fresh urine, and urinary thiosulfate and S-sulfocysteine levels should be assayed. Serum uric acid and urinary oxypurine levels should also be measured, to distinguish isolated sulfite oxidase deficiency from molybdenum cofactor deficiency.
Although it has never been attempted in isolated sulfite oxidase deficiency, the encouraging response to dietary restriction of methionine with supplementation of cysteine in a case of molybdenum cofactor deficiency46 makes that a potentially attractive therapeutic option.
ACKNOWLEDGMENTS
This work was supported in part by the Mary L. Efron Fund at Massachusetts General Hospital.
We are indebted to Jean L. Johnson, PhD (Department of Biochemistry, Duke University), for providing us with the primers and sequences required for mutation analysis of the sulfite oxidase gene. We are grateful to Roseann Mandell, BA, for expert technical assistance and advice. We thank the staff of the NICU at Massachusetts General Hospital for their excellent care of the patient during his early days.
FOOTNOTES
Accepted Dec 7, 2004.
No conflict of interest declared.
REFERENCES
Johnson JL, Duran M. Molybdenum cofactor deficiency and isolated sulfite oxidase deficiency. In: Scriver CR, Sly WS, Childs B, et al, eds. The Metabolic and Molecular Bases of Inherited Disease. 8th ed. New York, NY: McGraw-Hill Professional; 2001:3163–3177
Williams RR, Waterman RE, Keresztesy JC, Buchman ER. Studies of crystalline vitamin B1. III. Cleavage of vitamin with sulfite. J Am Chem Soc. 1935;57 :536 –537
Til HP, Feron VJ, De Groot AP. The toxicity of sulphite. I. Long-term feeding and multigeneration studies in rats. Food Cosmet Toxicol. 1972;10 :291 –310
Garrett RM, Johnson JL, Graf TN, Feigenbaum A, Rajagopalan KV. Human sulfite oxidase R160Q: identification of the mutation in a sulfite oxidase-deficient patient and expression and characterization of the mutant enzyme. Proc Natl Acad Sci USA. 1998;95 :6394 –6398
Griffith OW. Mammalian sulfur amino acid metabolism: an overview. Methods Enzymol. 1987;143 :366 –376
Oshino N, Chance B. The properties of sulfite oxidation in perfused rat liver; interaction of sulfite oxidase with the mitochondrial respiratory chain. Arch Biochem Biophys. 1975;170 :514 –528
Woo WH, Yang H, Wong KP, Halliwell B. Sulphite oxidase gene expression in human brain and in other human and rat tissues. Biochem Biophys Res Commun. 2003;305 :619 –623
Kutter D, Humbel R. Screening for sulfite oxidase deficiency. Clin Chim Acta. 1969;24 :211 –214
Wadman SK, Cats BP, de Bree PK. Sulfite oxidase deficiency and the detection of urinary sulfite. Eur J Pediatr. 1983;141 :62 –63
Shih VE, Carney MM, Mandell R. A simple screening test for sulfite oxidase deficiency: detection of urinary thiosulfate by a modification of Sorbo’s method. Clin Chim Acta. 1979;95 :143 –145
Mann G, Kirk JM. Antibiotic interference in urinary thiosulphate measurements. J Inherit Metab Dis. 1994;17 :120 –121
Shih VE. Ampicillin interference with test for sulfite oxidase deficiency. Clin Chem. 1983;29 :211 –212
Millington DS, Stevens RD, Rajagopalan KV, Johnson JL. Quantification of S-sulphocysteine, a marker for sulphite oxidase deficiency, by tandem mass spectrometry . J Inherit Metab Dis. 2000;23 :77
Graf WD, Oleinik OE, Jack RM, Weiss AH, Johnson JL. Ahomocysteinemia in molybdenum cofactor deficiency. Neurology. 1998;51 :860 –862
Sass JO, Nakanishi T, Sato T, Shimizu A. New approaches towards laboratory diagnosis of isolated sulphite oxidase deficiency. Ann Clin Biochem. 2004;41 :157 –159
Infante-Rivard C, Rivard GE, Yotov WV, Théret Y. Perinatal reference intervals for plasma homocysteine and factors influencing its concentration. Clin Chem. 2002;48 :1100 –1102
Refsum H, Grindflek AW, Ueland PM, et al. Screening for serum total homocysteine in newborn children. Clin Chem. 2004;50 :1769 –1784
Vilaseca MA, Moyano D, Ferrer I, Artuch R. Total homocysteine in pediatric patients. Clin Chem. 1997;43 :690 –692
Irreverre F, Mudd SH, Heizer WD, Laster L. Sulfite oxidase deficiency: studies of a patient with mental retardation, dislocated ocular lenses, and abnormal urinary excretion of S-sulfo-L-cysteine, sulfite, and thiosulfate. Biochem Med. 1967;1 :187 –217
Mudd SH, Irreverre F, Laster L. Sulfite oxidase deficiency in man: demonstration of the enzymatic defect. Science. 1967;156 :1599 –1602
Shih VE, Abroms IF, Johnson JL, et al. Sulfite oxidase deficiency: biochemical and clinical investigations of a hereditary metabolic disorder in sulfur metabolism. N Engl J Med. 1977;297 :1022 –1028
Johnson JL, Wuebbens MM, Mandell R, Shih VE. Molybdenum cofactor deficiency in a patient previously characterized as deficient in sulfite oxidase. Biochem Med Metab Biol. 1988;40 :86 –93
Goh A, Lim KW. Sulphite oxidase deficiency: a report of two siblings. Singapore Med J. 1997;38 :391 –394
Tardy P, Parvy P, Charpentier C, Bonnefont JP, Saudubray JM, Kamoun P. Attempt at therapy in sulphite oxidase deficiency. J Inherit Metab Dis. 1989;12 :94 –95
Touati G, Rusthoven E, Depondt E, et al. Dietary therapy in two patients with a mild form of sulphite oxidase deficiency: evidence for clinical and biological improvement. J Inherit Metab Dis. 2000;23 :45 –53
Barbot C, Martins E, Vilarinho L, Dorche C, Cardoso ML. A mild form of infantile isolated sulphite oxidase deficiency. Neuropediatrics. 1995;26 :322 –324
van der Klei-van Moorsel JM, Smit LM, Brockstedt M, Jakobs C, Dorche C, Duran M. Infantile isolated sulphite oxidase deficiency: report of a case with negative sulphite test and normal sulphate excretion. Eur J Pediatr. 1991;150 :196 –197
Vianey-Liaud C, Desjacques P, Gaulme J, et al. A new case of isolated sulphite oxidase deficiency with rapid fatal outcome. J Inherit Metab Dis. 1988;11 :425 –426
Brown GK, Scholem RD, Croll HB, Wraith JE, McGill JJ. Sulfite oxidase deficiency: clinical, neuroradiologic, and biochemical features in two new patients. Neurology. 1989;39 :252 –257
Edwards MC, Johnson JL, Marriage B, et al. Isolated sulfite oxidase deficiency: review of two cases in one family. Ophthalmology. 1999;106 :1957 –1961
Schiaffino MC, Fantasia AR, Minniti G, Caruso U, Carnevale F, Cerone R. Isolated sulphite oxidase deficiency: clinical and biochemical features in an Italian patient. J Inherit Metab Dis. 2004;27 :101 –102
Johnson JL, Rajagopalan KV, Renier WO, van der Burgt I, Ruitenbeek W. Isolated sulfite oxidase deficiency: mutation analysis and DNA-based prenatal diagnosis. Prenat Diagn. 2002;22 :433 –436
Rupar CA, Gillett J, Gordon BA, et al. Isolated sulfite oxidase deficiency. Neuropediatrics. 1996;27 :299 –304
Amiel J, Gagey V, Rabier D, et al. Sulfite oxidase deficiency presenting as Leigh syndrome [in French]. Arch Pédiatr. 1994;1 :1023 –1027
Rosenblum WI. Neuropathologic changes in a case of sulfite oxidase deficiency. Neurology. 1968;18 :1187 –1196
Dublin AB, Hald JK, Wootton-Gorges SL. Isolated sulfite oxidase deficiency: MR imaging features. AJNR Am J Neuroradiol. 2002;23 :484 –485
Salman MS, Ackerley C, Senger C, Becker L. New insights into the neuropathogenesis of molybdenum cofactor deficiency. Can J Neurol Sci. 2002;29 :91 –96
Zhang X, Vincent AS, Halliwell B, Wong KP. A mechanism of sulfite neurotoxicity. J Biol Chem. 2004;279 :43035 –43045
Sachdev NH, Coroneo MT, Wakefield D, Hennessy MP. Isolated ectopia lentis: potential role of matrix metalloproteinases in fibrillin degradation. Arch Ophthalmol. 2004;122 :111 –114
Brew K, Dinakarpandian D, Nagase H. Tissue inhibitor of metalloproteinases: evolution, structure and function. Biochim Biophys Acta. 2000;1477 :267 –283
Gunnison AF, Benton AW. Sulfur dioxide: Sulfite. Interaction with mammalian serum and plasma. Arch Environ Health. 1971;22 :381 –388
Johnson JL, Coyne KE, Garrett RM, et al. Isolated sulfite oxidase deficiency: identification of 12 novel SUOX mutations in 10 patients. Hum Mutat. 2002;20 :74
Lee HF, Mak BS, Chi CS, Tsai CR, Chen CH, Shu SG. A novel mutation in neonatal isolated sulphite oxidase deficiency. Neuropediatrics. 2002;33 :174 –179
Lam CW, Li CK, Lai CK, et al. DNA-based diagnosis of isolated sulfite oxidase deficiency by denaturing high-performance liquid chromatography. Mol Genet Metab. 2002;75 :91 –95
Chan KY, Li CK, Lai CK, Ng SF, Chan AY. Infantile isolated sulphite oxidase deficiency in a Chinese family: a rare neurodegenerative disorder. Hong Kong Med J. 2002;8 :279 –282
Boles RG, Ment LR, Meyn MS, Horwich AL, Kratz LE, Rinaldo P. Short-term response to dietary therapy in molybdenum cofactor deficiency. Ann Neurol. 1993;34 :742 –744
Kurlemann G, Debus O, Schuierer G. Dextromethorphan in molybdenum cofactor deficiency. Eur J Pediatr. 1996;155 :422 –423
Rabier D, Chadefaux-Vekemans B, Oury JF, et al. Gestational age-related reference values for amniotic fluid amino acids: a useful tool for prenatal diagnosis of aminoacidopathies. Prenat Diagn. 1996;16 :623 –628
Johnson JL. Prenatal diagnosis of molybdenum cofactor deficiency and isolated sulfite oxidase deficiency. Prenat Diagn. 2003;23 :6 –8
Vilarinho L, Alves JR, Dorche C, Chadefaux B, Parvy P. Citrullinaemia and isolated sulphite oxidase deficiency in two siblings. J Inherit Metab Dis. 1994;17 :638 –639(Wen-Hann Tan, MRCPCH, Flo)