Acute Basal Ganglia Infarction in Propionic Acidemia Richard H. Haas, MD; Deborah L. Marsden, MD; Sylvia Capistrano-Estrada, MD; Ronald Hamilton, MD; Marjorie R. Grafe, MD; Wade Wong, MD; William L. Nyhan, MD, PhD ABSTRACT An 8-year-old girl with propionic acidemia had acute and rapidly fatal symmetric necrosis of the caudate, globus pallidus, and putamen. Clinical presentation was with acute aphasia, generalized hypotonia, and muscle weakness. There was no evidence of metabolic decompensation, and analysis of the organic acids of the urine indicated good metabolic control. Organic acids in the cerebrospinal fluid were unremarkable. These observations indicate that the pathophysiology of "metabolic stroke" is more complicated than previously thought. (J Child Neurol 1995;10:18-22). Propionic acidemia, like methylmalonic acidemia, is a defect in the catabolic pathway for the amino acids valine, isoleucine, threonine, and methionine (Figure 1).1I These disorders usually present in the neonatal period with severe metabolic ketoacidosis and dehydration. Associated hypotension and diminished perfusion of vital organs is one of a variety of mechanisms productive of injury to the brain in organic acidemia. Hypoglycemia and neonatal hyperammonemia are other factors that may be followed by global developmental delay. &dquo;Metabolic stroke&dquo; involving the basal ganglia was first reported as a complication of methylmalonic acidemia.’,’ The patients presented acutely with dystonia and were found to have high T2 signal abnormalities on magnetic resonance imaging (MRI) or areas of decreased attenuation on computed tomographic (CT) scans, typically in the globus pallidus. Each of these episodes accompanied typical acute metabolic decompensation with ketosis and acidosis. It is the purpose of this report to describe a patient with propionic acidemia who developed fatal necrosis of the basal ganglia despite evidence of excellent metabolic control at the time of the appearance of the lesions. This experience suggests that factors other than acute metabolic decom- pensation account for this strokelike picture. CASE HISTORY E.N. was bom to a 33-year-old mother who had six previous chil- dren. On the 2nd day of life, the infant became gravely ill and was transferred to the University of California San Diego Medical Received Nov 29, 1993. Received revised May 5, 1994. Accepted for publication May 20, 1994. From the Departments of Neurosciences (Dr Haas), Pediatrics (Drs Marsden, Capistrano-Estrada, and Nyhan), Radiology (Dr Wong), and Pathology (Drs Hamilton and Grafe), University of California San Diego, San Diego, CA. Address correspondence to Dr Richard H. Haas, Division of Pediatric Neurology, University of California, San Diego, 9500 Gilman Drive, Department 0935, La Jolla, CA 92093-0935. 18 Center, where she was found to be acidotic, hypoglycemic, and hyperammonemic. The concentration of ammonia in the blood was 700 ug/dL. She developed seizures and apnea and required intubation and artificial ventilation. The initial level of 3-hydroxypropionic acid in urine was 2308 mmol/mol creatinine, and the methylcitrate was 557 mmol/mol creatinine. A diagnosis of propionic acidemia was made. The activity of propionyl coenzyme A (CoA) carboxylase in cultured fibroblasts was 25 pmol/min/mg protein, a value that was 4% of the control mean. She was initially treated with eight exchange transfusions and parenteral fluid therapy, and dietary therapy was ultimately begun. Satisfactory weight gain and nitrogen balance were recorded in hospital with a protein intake of 0.75 g/kg; at this intake level, methylcitric acid excretion was 111 mmol/mol creatinine. Carnitine was begun in a dose of 60 mg/kg. She had a number of admissions to the hospital for study or dietary adjustment but was not admitted for illness until age 17 months, when she developed Shigella dysentery, and 6 months later for pneumonia and otitis media. On each occasion, she had large amounts of ketones in the urine, but the bicarbonate was 16 to 19 mEq/L. Camitine treatment during this period ranged from 60 to 200 mg/kg. The plasma free carnitine was low, with free levels of 18 to 24 umol/L, and the total urinary camitine esters ranged from 100 to 1000 umol/L. She developed ketonuria with bilateral maxillary sinusitis 4 at years of age, which was successfully treated with antibiotics on an outpatient basis. She was admitted to the hospital at age 41/z years because of ketosis and acidosis associated with respiratory infection and ethmoid sinusitis. She was admitted with ketoacidosis requiring intravenous fluids containing bicarbonate at 5 years of age, having been managed without admission for a number of episodes associated with respiratory infection over the previous year. She continued to have recurrent episodes of ketonuria, although none required admission to the hospital. At the age of 7 years 9 months she was admitted with an intercurrent viral illness and persistent abdominal pain, vomiting, and ketonuria. Urine organic acid analysis at that time revealed metabolic decompen- Downloaded from jcn.sagepub.com at UNIV ARIZONA LIBRARY on July 20, 2015 19 well. Her organic acid analysis revealed very high levels: 3- Figure 1. Metabolic pathways and the propionyl coenzyme A carboxylase reaction. sation with a 3-hydroxypropionic acid level of 1694, propionyl- glycine 1019, tiglylglycine 365, and methylcitric acid 733 mmol/mol creatinine (see Table 1 for normal values). She was treated with carnitine intravenously and parenteral alimentation with a specially formulated amino acid solution containing none of the toxic precursors.4 She responded well and was discharged after 4 days, by which time her 3-hydroxypropionic acid had decreased to 91 and methylcitric acid to 320 nimol/mol creatinine. Onset of Neurologic Symptoms She was next admitted 3 months later with an episode in which presumptive diagnosis of viral encephalitis was made. She was encephalopathic, with lethargy and confusion. There were 20 white blood cells/mm3 in the cerebrospinal fluid (four segmented neutrophils, eight lymphocytes, and eight macrophages); glucose and protein levels were normal. Cultures were negative. Organic acid analysis revealed a 3-hydroxypropionic acid level of 1402, propionylglycine 361, tiglylglycine 869, and methylcitric acid 395 mmol/mol creatinine. She was again treated with carnitine intravenously and the intravenous amino acid formula, with a good response. Four days after admission, the 3-hydroxypropionic acid level was 111, propionylglycine 0, tiglylglycine 100, and methylcitrate 465 mmol/mol creatinine. MRI scan of the brain on the day of admission showed increased T2-weighted signal bilaterally in the caudate nucleus, globus pallidus, and putamen (Figure 2). Repeat MRI 2 months later showed almost complete resa hydroxypropionic acid 1997, propionylglycine 1520, tiglylglycine 49, and methylcitric acid 1548 mmol/mol creatinine. She was started on oral neomycin, 500 mg four times daily. Repeat organic acid analysis 1 week later showed considerable improvement in organic acid levels: 3-hydroxypropionic acid 88, propionylglycine 37, tiglylglycine 0, and methylcitric acid 286 mmol/mol creatinine. At 8 years 10 months, she was again admitted after a 3-day history of abdominal pain, vomiting, and ketonuria. There were no neurologic findings. Serum amylase level was elevated at 538 U/L, and the lipase level was 4514 U/L. A diagnosis of acute pancreatitis was made. On this occasion, metabolites were elevated (Table 1). 3-Hydroxypropionic acid level was 866, tiglylglycine 1540, propionylglycine 963, and methylcitric acid 412 mmol/mol creatinine. Treatment was again with parenteral fluids containing the amino acid formulation, as well as oral neomycin and carnitine intravenously. She had a quick metabolic response. Levels of 3-hydroxypropionic acid, propionylglycine, tiglylglycine, and methylcitric acid decreased within 2 days to 219, 292, 59, and 5311 mmol/mol creatinine, respectively. She was discharged after 9 days to receive her usual diet of 1 g/kg of protein per day. She was again admitted 4 months later (23 days before her death) with pancreatitis (amylase 1510 U/L and lipase 11348 U/L), although metabolic control was excellent. Levels of metabolites were 3-hydroxypropionic acid 205, propionylglycine 294, tiglylglycine 41, and methylcitric acid 93 mmol/mol creatinine. The pancreatitis resolved over 19 days with intravenous man- agement consisting of carnitine initially, with 10% glucose and lipid, and in the last 5 days, the special amino acid formulation. She appeared well and discharge was planned. However, she developed an acute change of mental status with disorientation, rapidly developing aphasia, and generalized hypotonia and weakness. Organic acid analysis continued to reveal low levels of metabolites (3-hydroxypropionic acid 113, propionylglycine 169, tiglylglycine 0, and methylcitric acid 72 mmol/mol creatinine (Table 1). CT and MRI scans of the brain revealed evidence of bilateral infarction of the caudate, putamen and globus pallidus (Figure 3). Initial head CT scan showed decreased attenuation throughout both caudates and putamina with contrast enhancement in the putamina. An MRI scan 24 hours later showed gadolinium enhancement. She was treated with dexamethasone and hyperbaric oxygenation, which was followed by clinical improvement manifested by improved muscle tone and power and some responsiveness. However, after a second hyperbaric treatment, she developed large bilateral hemorrhagic Table 1. Organic Acids of Urine olution of these changes. The next admission was at 8 years 4 months of age because of lethargy, vomiting, and ketonuria. Organic analysis again showed evidence of acute metabolic decompensation with enormous elevation of organic acids: the 3-hydroxypropionic acid level was 2138, propionylglycine 820, tiglylglycine 1705, and methylcitric acid 2104 mmol/mol creatine. On this occasion, she was treated with oral neomycin, intravenous carnitine, and fluids, with good response. Enterovirus was isolated. A CT scan revealed cerebral atrophy when compared to previous studies. She was admitted again 2 weeks later for a dietary evaluation. She was mildly lethargic but had no ketonuria and was otherwise clinically *All values mmol/mol creatinine. Downloaded from jcn.sagepub.com at UNIV ARIZONA LIBRARY on July 20, 2015 20 lesions of the basal ganglia (Figure 3) and died on the 26th day of hospitalization at the age of 91/z years. Studies of coagulation after the hemorrhage revealed a platelet count of 438 x lO3/mm3, a prothrombin time of 13.1 seconds (control, 11.7 seconds), a partial thromboplastin time of 45.8 seconds (control, 28.1 seconds), and fibrinogen concentration of 273 mg/dL. Quantification of the organic acids of the urine during the terminal illness are compared to those during severe metabolic decompensation at the time of the episode of pancreatitis 3 months earlier in Table 1. Organic acids of the cerebrospinal fluid 2 days before death showed only slight elevation of 3hydroxypropionic acid of 22 pmol/L (normal, 0 to 16 umol/L) and of methylcitric acid of 1 umol/L (normal, 0 limol/L). Levels of lactic and pyruvic acid in the cerebrospinal fluid were normal. Total organic acid concentration in the cerebrospinal fluid was 51 mEq/L, and that in the plasma was 25 mEq/L. The levels of methylcitric acid, lactic acid, and pyruvic acid were higher in the cerebrospinal fluid than in the plasma, whereas propionylglycine was raised in plasma but not cerebrospinal fluid. Neuropathologic study (see Hamilton et al5) confirmed that lesions were limited to the gray matter of the basal ganglia. The lesions did not follow a vascular distribution but remained within the anatomic borders of these nuclei. The hippocampi and cerebellum were normal, suggesting that generalized hypoxic/ ischemic injury was not an etiologic factor. DISCUSSION Late complications of organic acidemia are being recognized increasingly. The patient reported experienced two of these, acute recurrent pancreatitis and acute necrosis of the basal ganglia. Pancreatitis has only recently been recognized 6,7 as a serious, recurrent, and sometimes fatal complication of a wide variety of organic acidemias, including methylmalonic acidemia. Of nine patients in a series, five had methylmalonic acidemia and two had isovaleric acidemia. In two of the methylmalonic acidemia patients, pancreatitis was fatal. It has not previously been reported in propionic acidemia. The pathogenesis of this complication is not clear, but in our experience, patients have been those with relatively poor degrees of compliance or metabolic control and high levels of organic acids in body fluids. The current patient was in a typical ketoacidotic episode when she developed pancreatitis for the first time. Later recurrences were associated with enormous accumulations of the organic acids typical of propionic acidemia. These observations suggest that one or more of these organic acids is responsible for the genesis of the pancreatitis. Acute infarction of the basal ganglia is an even more devastating complication of organic acidemia. It has not previously been reported in propionic acidemia. The initial report by Korf et a12 was of a 19-month-old infant with methylmalonic acidemia who developed dystonia after an episode of metabolic decompensation and encephalopathy and developed bilateral lucency of the globus pallidus. Four patients with methylmalonic acidemia were reported3 who developed an acute extrapyramidal syndrome at times of acute ketoacidosis and had bilateral globus pallidus lesions on CT scan, with variable involvement of the Figure 2. A, MRI scan of brain (TRITE 3000 ms/80 ms; 1 nex). High T2 signal was found in caudate nuclei, globi pallidi, and putamina during an episode originally thought to be encephalitis 13 months before the terminal illness. B, Follow-up MRI 2 months after A. Lesions of basal ganglia had resolved except for minimal residual T2 signal in the medial globi pallidi. = internal capsules. Each of these episodes accompanied typical acute metabolic decompensation with ketosis and acidosis. This was called metabolic stroke. The patient with propionic acidemia in our report differs in that she had not had ketoacidosis for some time, and a period of intensive therapy with parenteral carnitine and alimentation had led to unusually low levels of all of the relevant organic acid intermediates. It remains possible that the earlier high levels that presumably led to the pancreatitis somehow set the stage for the development of this catastrophic complication. Although the coexistence of pancreatitis at the time of the terminal catastrophic disease may be a predisposing factor for the fulminant course, hemorrhagic infarction did not accompany any of the previous episodes of pancreatitis and has not been noted in other patients with pancreatitis and organic acidemia.77 Neuropathologic evidence of disease in the basal ganglia has been reported8 in propionic acidemia. A marbled appearance was identified in the caudate, along with loss of large neurons and gliosis in a 14-year-old patient who had athetosis. The putamen was also affected, but the globus pallidus and thalamus were normal. In a 4year-old patient with propionic acidemia, the distribution of changes, including cell loss and gliosis, was in the pallidus as well as the caudate and putamen. The classic neuropathology of propionic acidemia has been spongiform cerebral and cerebellar white-matter change.9,10 Widespread hypoxic/ischemic injury may be seen in infants dying in the initial ketoacidotic episode. An adult onset of chorea and dementia has been described in pro- pionic acidemia.l As noted above, in previously described patients with organic acidemia and damage to the basal ganglia, there has been no evidence of hemorrhage. The hemorrhagic areas in this patient were essentially confined to the graymatter nuclei of the basal ganglia, ignoring vascular Downloaded from jcn.sagepub.com at UNIV ARIZONA LIBRARY on July 20, 2015 21 Figure 3. A and B, CT scan of the brain 48 hours before death showed decreased attenuation throughout the basal ganglia, edema of adja- cent white matter, and central contrast enhancement within these nuclei (B). C, Ti-weighted image after gadolinium (TR/TE = 550 ms/20 ms; 2 nex) taken 24 hours after the first hyperbaric treatment (which was also 6 hours before death) reveals enhancement in the nuclei. The previ- hyperintensities in the basal ganglia (not shown) remained unchanged. D, Non-contrast-enhanced CT scan after terminal collapse. Unilateral hemorrhage was seen in the basal ganglia. ous T2 boundaries and sparing white matter. It is unlikely that such a hemorrhagic infarction could be the result of an unrecognized hypotensive or anoxic/ischemic insult, perhaps with reperfusion. The lack of vascular distribution of the lesions argues strongly against such a scenario. The evidence of hemorrhage occurred after treatment with steroid and hyperbaric oxygen, and this could be relevant to hemorrhage in an already infarcted area, even though we have found no reports of hyperbaric treatment producing intracranial hemorrhage. No coagulation abnormalities were identified in the patient. A breakdown of endothelial cell integrity may have led to hemorrhage. The pathogenesis of acute infarction of the basal ganglia in our propionic acidemic patient is not clear. These structures of the brain appear to be particularly sensitive to metabolic disturbances that affect oxidative metabolism. In Leigh syndrome and methanol toxicity, the typical lesions are in the putamen. In organic acidemias, the globus pallidus is usually affected first. Changes in the basal ganglia on neuroimaging are nonspecific and have been seen in other metabolic diseases, such as Wilson’s disease, glutaric aciduria type 2, metachromatic leukodystrophy, and mitochondrial encephalomyopathy, lactic acidosis, and strokelike episodes (MELAS).12 In some of these conditions, the normal high concentrations of iron and trace metals in the basal ganglia may play a role in the pathophysiology of injury through the mechanism of freeradical damage. 13 Acute processes such as hypoxia,l4 hypoglycemia, hemolytic uremic syndrome, and in some cases, encephalitis can be manifested by basal ganglial changes. The latter was considered during our patient’s illness at 8 years of age, which was accompanied by cerebrospinal fluid pleocytosis. Mitochondrial dysfunction occurs during acute ketoacidotic illness in propionic acidemia and methylmalonic acidemia. However, in our patient, the terminal acute lesions of the basal ganglia occurred at a time of extraordinary systemic metabolic stability. Studies of the cerebrospinal fluid not only showed low levels of the characteristic metabolites at the time of the acute changes in the basal ganglia, but there was also little accumulation of lactic acid. Certainly, widespread ischemia of the brain was not occurring. In patients with defects of propionate metabolism, the accumulation of propionyl coenzyme A leads to the presence in body lipids of large amounts of odd-numbered long-chain fatty acids.l5,ls Abnormal composition of brain lipids could result in structural and functional changes in the central nervous system, perhaps accounting for some of the white-matter disease characteristic of the neuropathology. Structural membrane abnormalities in the basal ganglia could cause increased susceptibility to toxic metabolite injury. It is possible that one or more of the metabolites that accumulate in propionic acidemia is responsible for the observed neurotoxicity. It does not seem likely that tiglylglycine was responsible, because levels in this patient tended to be lower than in many with this disease and were undetectable at the time of the final episode. Propionylglycine on the other hand was unusually high in this patient for some time over the last months of life. This may signify that accumulated propionyl coenzyme A is the toxic compound. Propionyl coenzyme A, the primary compound that accumulates in propionic acidemia and methylmalonic acidemia, would be expected to reduce the mitochondrial respiratory control ratio, on the basis of studies with isolated liver mitochondria incubated with propionate. 17,18 Propionyl coenzyme A also inhibits the citric acid cycle enzyme succinyl coenzyme A ligase.19 Accumulation of toxic metabolic intermediates may be expected to contribute to central nervous system damage occurring during episodes of metabolic decompensation, but it is difficult to invoke such a mechanism in the infarctions of the basal ganglia sustained by this patient because both urine and cerebrospinal fluid organic acid analyses indicated excellent metabolic control. Downloaded from jcn.sagepub.com at UNIV ARIZONA LIBRARY on July 20, 2015 22 Defective supplies of succinyl coenzyme A and acetyl coenzyme A may occur even in well-controlled patients with propionic acidemia and methylmalonic acidemia, and this would compromise citric acid cycle function and oxidative metabolism. The level of 2-oxoglutaric acid was elevated, and that of citric acid was relatively high in the urine collected 4 days before death in spite of relatively low levels of propionate precursors, but levels of fumaric acid, lactic acid, and pyruvic acid were relatively low (Table 1). The elevated 2-oxoglutaric acid is compatible with citric acid cycle dysfunction, as might occur with a propionyl coenzyme A-induced inhibition of succinyl coenzyme A ligase. Potentially neurotoxic compounds accumulating in propionic acidemia include fatty acyl coenzyme A derivatives of propionic acid itself. With the recognition of the specific substantia nigra and basal ganglia neurotoxicity produced by 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP),2° the theory of the particular sensitivity of the basal ganglia to neurotoxic agents has gained support. 1Methyl-4-phenyl-1,2,3,6-tetrahydropyridine works by selective uptake and conversion to the complex I electron transport chain inhibitor, 1-methyl-4-phenylpyridinium (MPP+ ).21 1-Methyl-4-phenylpyridinium itself is structurally very similar to the herbicide paraquat, and this has raised questions about the role of agricultural toxins in the genesis of human idiopathic Parkinson disease. 20-12 A number of plant and fungal neurotoxins are now known to have particular striatal toxicity. These include cycad neurotoxins, which are implicated in the western Pacific parkinsonism-dementia amyotrophic lateral sclerosis complex,23 and 3-nitropropionic acid, a plant and fungal neurotoxin thought to be responsible for acute encephalopathy and delayed dystonia in children who ingest mildewed sugar cane.24 It is possible that patients with defects in propionic acid metabolism and those with other mitochondrial defects, such as Leigh syndrome, might be particularly susceptible to striatal toxicity from such environmental toxins as well as intrinsic neurotoxic agents. Recent work with a rat model shows greater vulnerability of the striatum in older animals to the effect of 3-nitropropionic acid when compared to neonatal animals .21 If neurotoxins are an underlying mechanism of basal ganglia damage in propionic acidemia and methylmalonic acidemia, such a model might help to explain why basal ganglia injury is predominantly a feature of these organic acidemias outside the period of early infancy. There is no evidence to support an etiologic role for an exogenous or endogenous neurotoxin in the basal ganglia destruction of our patient, who was metabolically stable at the time of her death. However, as noted above, there are several endogenous potential toxic agents in propionic acidemia, and it seems reasonable to suggest that infarction may have a multifactorial basis, with toxic metabolites contributing, perhaps by prior glial and neuronal insults. References 1. Nyhan WL, Sakati NO: Diagnostic Recognition of Genetic Dis- 1986;20:364-366. 3. 4. 5. 6. 7. 8. Bethesda, Maryland. Heidenreich R, Natowicz M, Hainline BE, et al: Acute extrapyramidal syndrome in methylmalonic acidemia: "Metabolic stroke" involving the globus pallidus. J Pediatr 1988;113:1022-1027. Nyhan WL, Rice-Asaro M, Acosta P: Advances in the treatment of amino acid and organic acid disorders, in Desnick RJ (ed): Treatment of Genetic Diseases. New York, Churchill Livingstone, 1991, pp 45-67. Hamilton RL, Haas RH, Nyhan WL, et al: Neuropathology of propionic acidemia: A report of two patients with basal ganglia lesions. J Child Neurol 1995;10:25-30. Wilson WG, Cass MB, Sovik O, et al: A child with acute pancreatitis and recurrent hypoglycemia due to 3-hydroxy-3-methylglutaryl-CoA lyase deficiency. Eur J Pediatr 1984;142:289-291. Kahler SG, Sherwood WG, Wolff D, et al: Pancreatitis in patients with organic acidemias. J Pediatr 1994;124:239-243. Harding BN, Leonard JV, Erdohazi M: Propionic acidemia: A neuropathological study of two patients presenting in infancy. 1991; 17:133-138. Neuropathol Appl Neurobiol Shuman RM, Leech RW, Scott CR: The neuropathology of the nonketotic and ketotic hyperglycinemias: Three cases. Neurology 1978;28:139-146. 10. Steinman L, Clancy RR, Cann H, Urich H: The neuropathology of propionic acidemia. Dev Med Child Neurol 1983;25:87-94. 11. Sethi KD, Ray R, Roesel RA, et al: Adult-onset chorea and dementia with propionic acidemia. Neurology 1989;39:1343-1345. 12. Ho VB, Fitz CR, Chuang SH, Geyer CA: Bilateral basal ganglia lesions: Pediatric differential considerations. Radiographics 1993;13:269-292. 13. Fahn S: Biochemistry of the basal ganglia. Adv Neurol 1976;14: 9. 14. 15. 16. 17. 59-89. Barkovich AJ: MR and CT evaluation of profound neonatal and infantile asphyxia. AJNR 1992;13:959-972. Wendel U, Baumgartner R, van der Meer SB, Spaapen LJ: Accumulation of odd-numbered long-chain fatty acids in fetuses and neonates with inherited disorders of propionate metabolism. Pediatr Res 1991;29:403-405. Wendel U: Abnormality of odd-numbered long-chain fatty acids in erythrocyte membrane lipids from patients with disorders of propionate metabolism. Pediatr Res 1989;25:147-150. Matsuishi T, Stumpf DA, Chrislip K: The effect of malate on propionate mitochondrial toxicity. Biochem Med Metab Biol 1991;46:177-184. 18. 19. 20. 21. 22. 23. 24. Acknowledgment This work was aided by US Public Health Service Grant No. NS 22343 for the Center for the Study of the Neurological Basis of Language, from the National Institute of Neurological Disease and Stroke, National Institutes of Health, Philadelphia, Lea & Febiger, 1987, pp 36-41. Korf B, Wallman JK, Levy HL: Bilateral lucency of the globus pallidus complicating methylmalonic acidemia. Ann Neurol ease. 2. 25. Matsuishi T, Stumpf DA, Seliem M, et al: Propionate mitochondrial toxicity in liver and skeletal muscle: Acyl CoA levels. Biochem Med Metab Biol 1991;45:244-253. Stumpf DA, McAfee J, Parks JK, Eguren L: Propionate inhibition of succinate: CoA ligase (GDP) and the citric acid cycle in mitochondria. Pediatr Res 1980;14:1127-1131. Langston JW, Ballard P: Chronic parkinsonism in humans due to a product of meperidine-analog synthesis. Science 1983;219: 979-980. Singer TP, Ramsay RR, McKeown K, et al: Mechanism of the neurotoxicity of 1-methyl-4-phenylpyridinium (MPP ), the toxic + bioactivation product of 1-methyl-4-phenyl-1,2,3,6-tetrahy- dropyridine (MPTP). Toxicology 1988;49:17-23. Barbeau A, Roy M, Bernier G, et al: Ecogenetics of Parkinson’s disease: Prevalence and environmental aspects in rural areas. Can J Neurol Sci 1987; 14:36-41. Spencer PS, Kisby GE, Ludolph AC: Slow toxins, biologic markers, and long-latency neurodegenerative disease in the western Pacific region. Neurology 1991;41(Suppl 2):62-66. Ludolph AC, He F, Spencer PS, et al: 3-Nitropropionic acid— exogenous animal neurotoxin and possible human striatal toxin. Can J Neurol Sci 1991;18:492-498. Brouillet E, Jenkins BG, Hyman BT, et al: Age dependent vulnerability of the striatum to the mitochondrial toxin 3-nitropropionic acid. J Neurochem 1993;60:356-359. Downloaded from jcn.sagepub.com at UNIV ARIZONA LIBRARY on July 20, 2015