Eur J Pediatr (1984) 142 : 147-150 Betaine in the treatment of homocysfinuria due to 5,10-methylenetetrahydrofolate reductase deficiency* U. Wendel and H. J. Bremer University Children's Hospital C, Diisseldorf, Federal Republic of Germany Abstract. In a 3-year-old mentally retard- ed girl with homocystinuria due to 5,10- methylenetetrahydrofolate reductase deficiency among different therapeutic approaches only treatment with betaine (15-20g/day) resulted in a satisfactory biochemical response. Betaine improved homocysteine remethylation and thus lowered plasma homocystine to trace amounts and normalized the previously very low plasma methionine concentra- tion. This biochemical response was associated with a clinical improvement although she remained mentally retard- ed. Key words: Homocystinuria - 5,10- methylenetetrahydrofolate reductase de- ficiency - Betaine - Homocysteine remethylation Introduction In 5,10-methylenetetrahydrofolate re- ductase (MTHFR) deficiency homocys- teine remethylation to methionine is im- paired due to lack of the endogenously formed methyl donor, 5-methyltetrahy- drofolate (5-MTHF). As a consequence patients show homocystinuria and ho- mocystinemia of moderate degree and decreased plasma and tissue concentra- tions of methionine. All patients, about 20, diagnosed up to now [for references see 2, 14] showed neurological dysfunc- tion of variable type and severity, even- tually manifesting as progressive neona- tal leukoencephalomyopathy [1]. The etiology of the neurological dys- function remains unclear. Reduced * In part presented as poster at the 21st annual symposium of the Society for the Study of Inborn Errors of Metabolism (SSIEM), 6-9th September, 1983 in Lyon Offprint requests to: Prof. Dr. U. Wendel, Uni- versity Children's Hospital C, University of Dfisseldorf, Moorenstr. 5, D-4000 Dtissel- doff, FRG capacity of methionine biosynthesis in the patients most accurately reflects the clinical severity [2] and may be of major pathogenetic importance, causing sub- strate deficiency for S-adenosyl-L- methionine (SAMe) formation. SAMe is the methyl donor for many methyl trans- fer reactions in the body, being involved in neurotransmitter, carnitine, phospha- tidylcholine, and subsequently also in myelin synthesis. Cerebral thromboem- bolism being related to homocysteine accumulation or deficient brain folates may also be involved in the neurological damage. In most cases no effective therapy has been found. Two patients responded to folates [6], another two patients with onset of the disease in early infancy responded to therapy with methionine, vitamins B6 and B12 and folinic acid [8] or folates, methionine and carnitine [1], respectively. In our patient these measures did not lead to biochemical improvement. How- ever, it was possible to improve homo- cysteine remethylation by stimulating the betaine-dependent pathway with oral supplementation of betaine mon0hy- drate. Case report C.M., the second child of a nonconsan- guineous Greek couple, came to our attention at 2 years of age because of marked psychomotor retardation. Gesta- tion and delivery were normal (birth weight 3250 g, length 52 cm, head circum- ference 33 cm). According to her mother she had been doing well until the age of 5 months when psychomotor retardation became obvious. From months 4-6 she was hospitalized because of congenital subluxation of the hip. At 2 years of age she was microcephat- ic (head circumference: 43 cm). Internal findings were normal. She was hypotonic and restless, showed athetoid move- ments of her arms and myotonicjerks of the lower extremities. She could not sit without support. Tendon reflexes were easily obtainable. She grasped objects and put them into her mouth. Social con- tact was poor. Drooling, grimacing and stereotype smacking movements of the mouth were observed. She could not speak, however, periodically she screamed without motive. Using the Denver Developmental Screening Test, she was found to function at a level of 6 months. Routine laboratory evaluations, in- cluding cerebrospinal fluid analysis, were unremarkable. A CAT scan re- vealed mild internal and external hydro- cephalus. EEG showed some dysrhyth- mia, seizures were never observed. Nerve conduction velocity was normal as were ophthalmological examinations. She had mild homocystinuria and cystathioninuria. Urinary homocystine excretion was 20-40gmol/day, that of cystathionine was twice as much. The plasma concentrations of the relevant amino acids were: homocystine: 15-25 lamol/1, mixed homocysteine-cysteine disulfide: 25-35 lamol/1, cystine: 251amol/1, cystathionine: traces, and methionine: 4-5 lamol/l. Methylmalonic acid was not detectable by gas chromato- graphy. Serum folate (3 ng/ml) and free camitine concentration (18.41amol/1) were low, cobalamin was normal. An oral methionine loading test of 100 mg L-methionine per kg body weight [3, 13] showed a normal disappearance of methionine from the plasma. 5,10-methylenetetrahydrofolate re- ductase activity measured in extracts of lymphocytes and cultured skin fibro- blasts was l~ss~than 2% of control values. Detailed data have been published [17]. Therapy was started at the age of 274 years. The different regimens are shown in Table 1 and were changed every 4-6 weeks. Family history. The first-born sister was microcephalic and severely mentally re- tarded. She had an almost identical course of disease. Since the age of 11/2 years she had regressed rapidly and had three episodes of deep coma and respira- tory failure. At 21/2 years she died in coma. No diagnosis had been estab- lished. Postmortem examination re- vealed multiple thromboses in various organs, including the transverse sinus
`
`Lilly Ex. 2097
`Sandoz v. Lilly IPR2016-00318
`
`

`
`t48 Table 1. Plasma amino acid concentrations before and during therapy Therapy Methionine Half cystine (gmol/1) Homocystine (1) None 4-5 25 15-25 (2) Vitamin B 6 (240 mg/day) 4 25 23 (3) Folic acid (20rag/day) 5 24 24 Folic acid (20 rag/day) 10-89 a 48-60 t4-19 Methionine (1 g/day) Folic acid (80 mg/day) 10-28 a 45 12-17 Methionine (1g/day) (4) Folinic acid (60mg/day) 21-115 a 45-66 12-18 Methionine (1 g/day) Folinic acid (60 rag/day) 8-112 a 35-70 11-15 Vitamin B6 (240 rag/day) Methionine (1 g/day) (5) Betaine (6g/day) 12-19 58-69 3 Folinic acid (15 rag/day) (6) Betaine (12g/day) 19-22 55 Trace Folinic acid (15 mg/day) (7) Betaine (15g/day) 30 55 Trace Folinic acid (15 mg/day) (8) Betaine (20g/day) 23-45 65-90 Trace Folinic acid (15 mg/day) Body weight was 12.5-13.5 kg Permanent treatment with antiplatelet drugs (dipyridamole 100 mg/day and acetyl-salicylic acid 375 rag/day) until administration of 15 g betaine/day Wide fluctuations at different intervals after methionine intake. Traces of cystathionine were always present. The concentration of the mixed homocysteine- cysteine disulfide remained rather constant (18-42 gmol/1), also when on betaine and small cerebral vessels 1. Widespread necroses in the cerebral cortex and me- dulla and areas of demyelination in the brainstem and spinal cord out of propor- tion to the vascular changes had been found. These findings are compatible with homocystinuria due to MTHFR-de- ficiency. Identical findings in a patient with MTHFR deficiency had been re- ported by Kanwar et al. [9]. Intermediate levels of MTHFR-activ- ity were observed in the parents and in one brother prenatally and after birth [171. Laboratory methods. Plasma samples were obtained from heparinized venous blood by immediate centrifugation. The precipitation of proteins with 5% sulfosa- licylic acid was directly performed there- 1 Neuropathological examinations were done by Prof. J. Pfeiffer, Institut ffir Him- forschung, and Prof. W. Schlote, Patholo- gisches Institut der Universiffit Ttibingen after. The supernatant obtained after pre- cipitation was stored at -20~ until analysis. Urine was collected on ice and kept deep-frozen until analysis. Quanti- tative amino acid analyses were per- formed on an LKB 4400 amino acid ana- lyzer using lithium citrate buffers and standard programs. Results The child was on various therapeutic regimens for 18 months. The effects on the sulfur-containing free plasma amino acids are shown on Table 1. There was no significant response to vitamin B 6 and folates, given alone or in combination. A methionine supplement of 1 g/clay, given in four doses, resulted in a high but un- stable rise of the plasma methionine levels, fluctuating according to the methionine intake. The concentrations of homocystine and the mixed cysteine- h5 35 g .~ 3o 2s o 20 o / o / o/ / /o / / / 5 / i 6 / / / d / o ;2 ,'5 ~; Betaine (g/day) Fig. 1. Methionine concentration in plasma as a function of the intake of betaine monohy- drate. Methionine concentrations were meas- ured every 4-6 weeks after raising the betaine dosage. The values represent fasting plasma concentrations. Reaching 20g betaine/day plasma samples were taken monthly at vari- able intervals to betalne intake. The values represent the mean (cid:127) S.D. of 14 plasma samples homocysteine disulfide remained mainly unchanged, while the plasma cystine concentration was normalized. A marked improvement was only not- ed after the administration of betaine monohydrate (6g/day): plasma homo- cystine dropped to trace amounts. The stepwise increase of betaine, every 4-6 weeks, up to 20 g/day, resulted in a dose- related rise of plasma methionine into the normal range (Fig. 1). The methion- ine levels were stable over the day without wide oscillations. Though ho- mocystine disappeared from blood the level of the mixed cysteine-homocyste- ine disulfide remained constant. Cystine concentration was normal. Serine and glycine, decomposition products ofbeta- ine, did not accumulate in plasma. The urinary output of homocystine and cystathionine varied considerably during all therapeutic regimens. Al- though urinary homocystine decreased when the child was on betaine it was nevertheless still excreted. During betaine therapy the child's motor function improved. She became more alert and interested in her sur- roundings and responded to her mother. She learned to crawl to objects of inter- est, to stand alone with a little support and to walk with support. The muscular tone improved. Still, she remained se- verely mentally retarded and could not speak, however, she stopped grimacing and screaming became infrequent. The
`
`Lilly Ex. 2097
`Sandoz v. Lilly IPR2016-00318
`
`

`
`149 increments of somatic and skull growth became normal. At the age of 4 years length was 96cm, weight lY2kg, and head circumference 55.0 cm. During therapy total serum folates were high (> 100 ng/ml), serum cobala- min remained normal without supple- ments and free serum carnitine remained low (11 lamol/1). For more than halfa year the maximum dose of 20 g betaine mono- hydrate per day was administered to the girl now weighing 13 kg without apparent harmful effects and without signs of a disturbed liver function. Discussion Using a regimen, comprising vitamin B6, folates and methionine, as proposed by Harpey et al. [8] our patient did not show a satisfactory response, however, we achieved good biochemical control by high-dose betaine supplement. Adminis- tration of as much as 20 g betaine mono- hydrate per day led to an increase and normalization of the plasma methionine concentration with only minor fluctua- tions and to a reduction of plasma homo- cystine to trace levels. Associated with the use of betaine the child improved cli- nically, however, still remaining men- tally retarded. Recycling of homocysteine is neces- sary for maintaining intracellular methionine levels [12]. Two pathways exist for homocysteine remethylation to methionine [11]. While the folate-de- pendent pathway (5-MTHF-homocyste- ine methyltransferase-EC 2.1.1.13), being ubiquitously distributed and being the more important one in humans, is impaired in MTHFR deficiency, the flux through the betaine-dependent pathway (betaine-homocysteine methyltransfer- ase-EC 2.1.1.5) can apparently be enhanced by supplementation of the methyl donor betaine. Low betaine doses have been shown to have no biochemical effect [3,10]. High doses, however, proved to be effective in stimulating the betaine dependent me- thylating pathway: patients with homo- cystinuria due to cystathionine-fl-synthe- tase deficiency, when on 6-10g betaine per day, showed a substantial reduction in plasma homocystine concentrations combined with a further increase in the already highly elevated methionine blood levels and sometimes a striking clinical improvement [16, 18]. Obviously there is some variability in methionine response to betaine in different patients. In disorders of homocysteine reme- thylation, as in MTHFR deficiency, beta- ine is acting against two biochemical dis- turbances: homocystinemia and hypo- methioninemia. Moderate degrees of homocystinemia bear the risk of induc- ing fatal thromboembolic complications in the brain [9] as was demonstrated here by the autopsy of the patient's sister. They should therefore be prevented. Methionine deficiency of the brain with secondary reduction ofneurotrans- mitter and myelin synthesis might be balanced more adequately by betaine treatment than by methionine supple- mentation. Using betaine, the tissues containing betaine-homocysteine me- thyltransferase activity, such as liver, kid- ney and brain [7], could meet their me- thionine requirements by in situ syn- thesis and in addition by uptake from the blood [19], after normalization of blood methionine concentrations. The persisting homocystinuria and the constantly elevated levels of cysteine- homocysteine disulfide in blood and urine, indicate a continous production of homocysteine in the body during betaine treatment. Total correction of the homo- cysteine-methionine remethylation by betaine could not be expected since the betaine methylating enzyme is not pres- ent in every organ. Replenishment of methionine in tissues when the patient is on betaine [5, 7] increases the methion- ine metabolism and in consequence leads to enhanced production of homo- cysteine and to increased plasma cystine levels. The still reduced urinary excre- tion ofhomocystine, however, denotes a highly stimulated flux of homocysteine through the transmethylation pathway. Recently betaine was added to vita- min Bn treatment in a child with homo- cystinuria due to an abnormal cobalamin metabolism [18], causing some biochemi- cal response. The relationship of the deficient brain folates to the neurological damage re- mains speculative. Although 5-MTHF is the fraction of folate in blood and tissues its only function is to participate in methionine biosynthesis [4, 151. Cystathioninuria in this patient, not being influenced by vitamin B6, was the result of an unbalanced formation and utilization of cystathionine caused by an increased homocysteine concentration. Cystathioninuria was reported only once in another patient with a disorder of homocysteine remethylation. MTHFR deficiency has turned out to be heterogenous with respect to the degree of neurological dysfunction, re- sidual enzyme activity probably enzyme characteristics, residual capacity of methionine biosynthesis, and response to therapy. In patients with unrespon- siveness to folates it seems to be reason- able to try betaine in high doses as early as possible. For further treatment of this patient a permanent supplement of 12g daily of betaine would be sufficient to obtain a reasonable plasma methionine level with only a trace of homocystine and a normal cystine content. Continous folinic acid treatment has not so far been proven effective. References 1. Allen RJ, Wong P, Rothenberg SP, Di Mauro S, Headington JT (1980) Progres- sive neonatal leukoencephalomyopathy due to absent methylenetetrahydrofolate reductase, responsive to treatment. Ann Neurol 8:211 2. Boss GR, Erbe RW (1981) Decreased rates of methionine synthesis by methylene tetrahydrofolate reductase-deficient fi- broblasts and lymphoblasts. J Clin Invest 67 : 1659-1664 3. Brenton DP, Cusworth DC, Gaull GE (1965) Homocystinuria: metabolic studies on 3 patients. J Pediatr 67 : 58-68 4. Erbe RW (1979) Genetic aspects of folate metabolism. Adv Hum Genet 9 : 293-354 5. Finkelstein JD, Harris B J, Kyle WE (1972) Methionine metabolism in mammals:ki- netic study of betaine-homocysteine me- thyltransferase. Arch Biochem Biophys 153 : 320-324 6. Freeman JM, Finkelstein JD, Mudd SH (1975) Folate-responsive homocystinuria and "schizophrenia". A defect in methyla- tion due to deficient 5,10-methylenetetra- hydrofolate reductase activity. N Engl J Med 292 : 491-496 7. Gaull GE, yon Berg W, R/iih~i NCR, Stur- man JA (1973) Development of methyl- transferase activities of human fetal tis- sues. Pediatr Res 7 : 527-533 8. Harpey JP, Rosenblatt DS, Cooper BA, Le Mo~l G, Roy C, Lafourcade J (1981) Homocystinuria caused by 5,10-methyl- enetetrahydrofolate reductase deficiency: A case in an infant responding to methio- nine, folinic acid, pyridoxine, and vitamin Bn therapy. J Pediatr 98:275-278 9. Kanwar YS, Manalignd JR, Wong PWK (1976) Morphologic studies ih a patient with homocystinuria due to 5,10 methyl- enetetrahydrofolate reductase deficiency. Pediatr Res 10 : 598-609 10. Levy HL, Mudd SH, Schulman JD, Drey- fus PM, Abeles RH (1970) A derangement in Bn metabolism associated with homo- cystinemia, cystathioninemia, hypome- thioninemia, and methylmalonic aciduria. Am J Med 48 : 390-397 11. Mudd SH, Levy HL (1983) Disorders of transsulfuration, In: Stanbury JB, Wyn- gaarden JB, Fredricksen DS, Goldstein JL, Brown MS (eds) The metabolic bases of inherited disease, 5th edn. Mac Graw Hill, pp 522-559
`
`Lilly Ex. 2097
`Sandoz v. Lilly IPR2016-00318
`
`

`
`150 12. Mudd SH, Poole JR (1975) Labile methyl balances for normal humans on various dietary regimens. Metabolism 24:721- 735 13. Perry TL, Hansen S, Mac Dougall L, War- rington PD (1967) Sulfur containing ami- no acids in the plasma and urine ofhomo- cystinuries. Clin Chim Acta 15 : 409-420 14. Rosenblatt DS, Cooper BA, Lue-Shing S, Wong PWK, Berlow S, Narisawa K, Baumgartner R (1979) Folate distribution in cultured human cells. Studies on 5,10- CH~-H4PteGlu reductase deficiency. J Clin Invest 63 : 1019-1025 Note added in proof 15. Rowe PB (1983) Inherited disorders of folate metabolism. In: Stanbury JB, Wyn- gaarden JB, Fredricksen DS, Goldstein JL, Brown MS (eds) The metabolic bases of inherited disease, 5th edn. Mac Graw Hill, pp 498-521 16. Smolin LA, Benevenga NY, Berlow S (1981) The use ofbetaine for the treatment of homocystinuria. J Pediatr 99:467-472 17. Wendel U, Claussen U, Diekmann E (1983) Prenatal diagnosis for methylene- tetrahydrofolate reductase deficiency. J Pediatr 102 : 938-940 18. Wilcken DEL, Wilcken B, Dudman NPB, Tyrrell PA (1983) Homocystinuria - the effects of betaine in the treatment of pa- tients not responsive to pyridoxine. 309 : 448-453 19. Zeisel SH, Wurtman RJ (1979) Dietary intake of methionine: influence on brain S-adenosylmethionine. In: Usdin E, Bor- chardt ET, Creveling CR (eds) Trans- methylation. Elsevier/North Holland, pp 5%68 Received September 29, 1983 / Accepted December 20, 1983 After another six months of betaine treatment (15 g betaine/day) the child's motor function had persistently improved and she is now able to walk without support. Recently we startet betaine treatment (15 g betaine-monohydrate/day) in a 15-month-old severely mentally retarded child with MTHFR- deficiency. Plasma homocystine dropped immediately from 40 lamol/1 to traces and plasma methionine increased from 14 to 34 lamol/1. One month of treatment resulted in a surprisingly profound improvement of the child's psychomotor function. Letters to the editor Eur J Pediatr (1984) 142 : 150 Fatal infantile cardiac glycogenosis without acid maltase deficiency presenting as congenital hydrops J. Atkin 1, J. W. Snow Jr. 2, H. Zellwcger 1, and W. J. Rhead 1 1 Department of Pediatrics, 2 Department of Pathology, University of Iowa Hospitals and Clinics, Iowa City, Iowa 52242, USA Sir,-We discuss here an unusual case of congenital hydrops with cardiomyopa- thy. The infant, born at 37 weeks gesta- tion, was a grossly hydropic male with ascites, enlarged tongue, and heart mur- mur. Workup for immune hydrops was negative. The electrocardiogram was abnormal, showing a Wolf-Parkinson- White pattern. Considerable cardiome- galy was found radiologically. The heart condition worsened and the infant suc- cumbed due to cardiorespiratory arrest on day 18. Autopsy performed 16 h after death revealed an enlarged and hyper- trophic heart. The weight was 80g (nor- mal 23 __ 14 g). Thickness of the right ven- tricular wall was 1.2 cm (normal 0.4-0.6). Thickness of the left ventricular wall was 1.5 cm (normal 0.3-0.7). Prominent gly- cogen deposits were found in the heart myocardium and, to a lesser extent, in the skeletal muscle (light microscopy). Although postmortem autolysis made it impossible to identify clearly membrane- Offprint requests to: Joan F. Atkin, M.D., Department of Pediatrics, Division of Medi- cal Genetics, University of Virginia School of Medicine, Charlottesville, VA22908, Phone - (804) 924-2665 bound glycogen deposits, fragments of lysosomal membranes were found adja- cent to glycogen deposits in the heart (electron microscopy). The glycogen content of the heart was 7.8% (normal less than 1.80/0) and the glycogen struc- ture was normal. Glycogen content of the liver was normal histologically, as well as chemically; muscle was not available for biochemical glycogen determination. Enzyme studies (Dr. B. Brown, St. Louis; Dr. D. Wenger, Denver; Dr. R. Howell, Houston) on frozen liver and heart, cul- tured lung and diaphragm fibroblasts yielded normal levels of branching enzyme, debranching enzyme, glucose- 6-phosphatase, phosphorylase, and alpha-glucosidase at pH 4 and pH 6.6, using maltose and glycogen as substrates. Cardiomuscular and muscular glyco- genosis resembling glycogenosis II (Pompe's Disease) with normal acid mal- tase (alpha-glucosidase) has been de- scribed on several occasions [1, 2, 3]. De Barsy et al. [2] described an 8-year-old boy with muscle weakness and a normal heart. In that case, acid maltase was nor- mal in the muscle, but deficient in leuko- cytes and fibroblasts. Four other boys [1, 3] were similar to our patient, yet in con- trast to our case, they presented in their teens with proximal muscle weakness and a hypertrophic cardiomyopathy. While cardiomyopathy is typical for glycogenosis II in infancy, the heart is usually not involved in childhood and adolescent variants of glycogenosis II. Our case is the first observation of car- diomuscular glycogenosis without a de- monstrable enzyme deficiency is an infant. It is notable that all hitherto reported patients with this condition were boys, which raises the possibility of X-linked recessive inheritance, although classic glycogenosis II with acid maltase deficiency is transmitted as an autosomal recessive condition. However, this case may have other implications, as well. It may represent a hitherto unreported cause of congenital hydrops; thus, car- diomuscular glycogenosis might be in- cluded in the differential diagnosis of nonautoimmune hydrops congenitus. References 1. Danon M, Oh S J, DiMauro S, Manaligod JR, Eastwood A, Naidu S, Schliselfeld L (1981) Lysosomal glycogen storage disease with normal acid maltase. Neurology 31: 51-57 2. deBarsy T, Ferriere G, Fernandez-Alvarez E (.1979) Uncommon case of type II glyco- genosis. Acta Neuropathol 47:245-247 3. Riggs JE, Schochet SS, Gutmann L, Shanske S, Neal WA, DiMauro S (1983) Lysosomal glycogen storage disease with- out acid maltase deficiency. Neurology 33 : 873-878 Received January 16, 1983 / Accepted February 6, 1984
`
`Lilly Ex. 2097
`Sandoz v. Lilly IPR2016-00318