GENE THERAPY FOR MAPLE SYRUP URINE DISEASE

Abstract

Maple syrup urine disease (MSUD) is a rare autosomal recessive disease with an incidence that is caused by a defective activity of the branched-chain 2-keto acid dehydrogenase (BCKD) leading to accumulation of branched-chain amino acids (BCAA) leucine, isoleucine, valine and their corresponding alpha-ketoacids (BCKA) in tissues and body fluids. The inventors herein characterized the Bckdha.sup.−/− mouse and Bckdhb.sup.−/− mouse recapitulating the classical forms of MSUD. As a proof of concept, they developed a (liver-directed) AAV gene therapy based on the transfer of human BCKDHA (hBCKDHA) or BCKDHB (hBCKDHB) mediated by AAV8 during immediate neonatal period in Bckdha−/− or Bckdhb.sup.−/− mice. The inventors demonstrated that hBCKDHA gene transfer completely rescued the lethal early-onset phenotype of Bckdha−/− mice allowing long-term survival to age 12 months, at which they were systematically sacrificed, without overt phenotypic abnormalities. They also demonstrated that hBCKDHB gene transfer exhibited similar survival and a normal growth without overt phenotypic abnormalities at age 3 months, with a dramatic improvement of the biochemical phenotype. The present invention relates to a method of treating MSUD by gene therapy.

Claims

1. A recombinant nucleic acid molecule comprising a transgene encoding for the branched-chain keto acid decarboxylase alpha or beta subunit wherein the transgene is operatively linked to a promoter.

2. The recombinant nucleic acid molecule of claim 1 wherein the transgene comprises a nucleic acid sequence having at least 80% of identity with SEQ ID NO:1 or SEQ ID NO:2.

3. The recombinant nucleic acid molecule of claim 1 wherein the sequence of the transgene is codon-optimized

4. The recombinant nucleic acid molecule of claim 3 wherein the transgene comprises the nucleic acid sequence of SEQ ID NO:3, SEQ ID NO:4, SEQ ID NO:5 or SEQ ID NO:6.

5. The recombinant nucleic acid molecule of claim 1 wherein the promoter is selected to drive the expression of the transgene specifically in the liver.

6. The recombinant nucleic acid molecule of claim 5 wherein the promoter is the hAAT promoter that comprises the nucleic acid sequence of SEQ ID NO:7.

7. The recombinant nucleic acid molecule of claim 1 wherein the promoter is selected to drive the expression of the transgene not specifically in the liver.

8. The recombinant nucleic acid molecule of claim 7 wherein the promoter is the EF1a promoter that comprises the nucleic acid sequence of SEQ ID NO:8.

9. The recombinant nucleic acid molecule of claim 7 wherein the EF1a promoter further comprises an extra intronic sequence that will increase the expression of the transgene by the promoter.

10. The recombinant nucleic acid molecule of claim 9 wherein the extra intronic sequence consists of the nucleic acid sequence of SEQ ID NO:9.

11. The recombinant nucleic acid molecule of claim 1 comprising the Woodchuck Hepatitis Virus (WHP) Posttranscriptional Regulatory Element (WPRE) sequence of SEQ ID NO:10.

12. The recombinant nucleic acid molecule of claim 1 comprising a polyadenylation signal sequence inserted downstream to the transgene of SEQ ID NO:11.

13. The recombinant nucleic acid molecule of claim 1 comprising the inverted terminal repeats (ITRs) sequences of SEQ ID NO:12.

14. The recombinant nucleic acid molecule of claim 1 comprising a nucleic acid sequence selected from the group consisting of SEQ ID NO:13 to SEQ ID NO:24.

15. A recombinant AAV8 viral particle that comprises the recombinant nucleic acid molecule of claim 1.

16. A method of treating maple syrup urine disease in a subject in need thereof comprising administering to the subject a therapeutically effective amount of the recombinant AAV8 viral particle of claim 15.

17. The method of claim 16 wherein the recombinant AAV8 viral particle is administered to the subject intravenously.

Description

FIGURES

[0053] FIG. 1. Bckdha.sup.−/− mouse model recapitulates the severe human MSUD phenotype. a) Kaplan-Meier curves showing the survival probability during the first 30 days of life (Bckdha.sup.−/− N=15, Bckdha.sup.+/− N=44, Bckdha.sup.+/+ N=26); **** P<0.0001 for the comparison Bckdha.sup.−/− vs Bckdha.sup.+/+ and Bckdha.sup.−/− vs Bckdha.sup.+/−, log-rank Mantel-Cox test. b) Weight curves showing major growth delay for Bckdha.sup.−/− (Bckdha.sup.−/− N=1, Bckdha.sup.+/− N=2, Bckdha.sup.+/+ N=3); data are means±SD. c) Weights at 1 week; data are means±SD; **** P<0.0001 for the comparison Bckdha.sup.−/− vs Bckdha.sup.+/+ and Bckdha.sup.−/− vs Bckdha.sup.+/−, one-way analysis of variance (ANOVA) with Tukey's post hoc. d) Leucine concentrations in plasma at 3 days; data are means±SD; **** P<0.0001 for the comparison Bckdha.sup.−/− vs Bckdha.sup.+/+ and Bckdha.sup.+/− vs Bckdha.sup.+/−, one-way analysis of variance (ANOVA) with Tukey's post hoc. e) Alloisoleucine concentrations in plasma at 3 days; data are means±SD.

[0054] FIG. 2. Scheme of the optimised AAV expression cassettes coding for either human BCKDHA or BCKDHB with (a) the Human elongation factor-1 alpha promoter (EF-1 alpha) and (b) the Human alpha anti-trypsin promoter (hAAT).

[0055] FIG. 3. High dose gene therapy allows long-term rescue of severe MSUD phenotype of Bckdha−/− mice with EF1α hBCKDHA transgene. a) Weight curves for males (Bckdha.sup.−/− N=6, Bckdha−/− N=3, Bckdha.sup.+/+ N=3); data are means±SD. b) Weight curves for females (Bckdha.sup.−/− N=3, Bckdha.sup.+/− N=8, Bckdha.sup.+/+ N=6); data are means±SD. c) Leucine concentrations in plasma ((Bckdha.sup.−/− non injected N=3, Bckdha.sup.−/− injected N=5, Bckdha.sup.+/− injected N=5, Bckdha.sup.+/+ injected N=5); data are means±SD; Western blot analyses of Bckdha.sup.−/− mice sacrificed at 6 months (N=5) and 1 non injected Bckdha.sup.−/− mice died at 1 week and 1 non injected Bckdha.sup.+/+. Histograms represent 3 technical replicates per individual. Data are means and SEM. d) Liver. e) Heart. f) Brain. g) Muscle.

[0056] FIG. 4. Reducing EF1a hBCKDHA dosage allows partial albeit transient rescue of the MSUD phenotype in Bckdha−/− mice. a) Weight curves (Bckdha.sup.−/− injected with 10.sup.13 vg/kg and sacrificed <4 weeks N=2, Bckdha.sup.−/− injected with 10.sup.13 vg/kg and sacrificed at 4 weeks N=5, Bckdha.sup.−/− injected with 10.sup.14 vg/kg and sacrificed at 4 weeks N=2, Bckdha.sup.+/− N=18, Bckdha.sup.+/+ N=9); data are means±SD. b) Leucine concentrations in plasma at sacrifice (Bckdha.sup.−/− injected with 10.sup.13 vg/kg and sacrificed 4 weeks N=2, Bckdha.sup.−/− injected with 10.sup.13 vg/kg and sacrificed at 4 weeks N=5, Bckdha.sup.−/− injected with 10.sup.14 vg/kg and sacrificed at 4 weeks N=2, Bckdha.sup.+/− N=18, Bckdha.sup.+/+ N=9) or at 4 weeks (Bckdha.sup.−/− injected with 10.sup.13 vg/kg from the previous experiment N=3); data are means±SD. Western blot analyses of Bckdha.sup.−/− mice injected with 10.sup.13 vg/kg and sacrificed at 4 weeks (N=7), Bckdha.sup.−/− mice injected with 10.sup.14 vg/kg and sacrificed at 4 weeks (N=2) and 1 non injected Bckdha.sup.−/− mice died at 1 week and 1 non injected Bckdha.sup.+/+. Histograms represent 3 technical replicates per individual. Data are means and SEM. c) Liver. d) Heart. e) Brain. f) Muscle.

[0057] FIG. 5. Gene therapy with hAAT hBCKDHA transgene allows transient rescue of the MSUD phenotype in Bckdha.sup.−/− mice. Weight curves showing a growth arrest and weight loss after day 14 leading to sacrifice at day 19 or 21 (Bckdha.sup.−/− with initial normal growth N=3, Bckdha.sup.−/− with growth on a lower curve N=2, Bckdha.sup.+/− N=10, Bckdha.sup.+/+ N=6); data are means±SD.

[0058] FIG. 6. Bckdhb.sup.−/− mouse model recapitulates the severe human MSUD phenotype. a) Kaplan-Meier curves showing the survival probability during the first 16 days of life (Bckdhb.sup.−/− N=7, Bckdhb.sup.+/+ N=8, Bckdhb.sup.+/+ N=4). b) Leucine and c) alloisoleucine concentrations in plasma at 1 days (Bckdhb.sup.−/− N=7, Bckdhb.sup.+/− N=8, Bckdhb.sup.+/+ N=4). Data are means±SD.

[0059] FIG. 7. High dose gene therapy allows rescue of severe MSUD phenotype of Bckdhb.sup.−/− mice with EF1α hBCKDHB transgene. Weight curves for a) males (Bckdhb.sup.−/− N=5, Bckdhb.sup.+/− N=4) and b) females (Bckdhb.sup.−/− N=2, Bckdhb.sup.+/− N=4, Bckdhb.sup.+/+ N=1). c) Leucine concentrations in plasma at 28 and 56 days post injection (Bckdha.sup.−/− N=7, Bckdha.sup.+/− N=8, Bckdha.sup.+/+ N=1). Data are means±SD.

EXAMPLE

Example 1

[0060] Bckdha−/− Mice Recapitulate the Severe Human MSUD Phenotype

[0061] We generated Bckdha−/− mice by crossing commercial heterozygous Bckdha+/− males and females, which did not display any particular phenotype. Bckdha−/− mice showed a lethal early-onset phenotype. Fifty percent of mice died before P3, 50% around P7 with a maximum life expectancy of 12 days (FIG. 1a). Bckdha−/− mice had a major growth delay (FIG. 1b, c). Mice that survived for more than a week showed reduced activity and abnormal response in the hindlimb test. At the biochemical level, Bckdha−/− mice displayed a major increase of branched-chain amino acids (FIG. 1d) and accumulation of alloisoleucine, a pathognomonic marker of MSUD in humans (FIG. 1e) in blood. In wild-type (WT) individuals, western blot showed that Bckdha protein was not expressed in skeletal muscle (quadriceps) (FIG. 3g) and showed mild expression in brain (FIG. 3f). In Bckdha−/− mice, Bckdha protein was absent in liver and heart, confirming the null nature of the model (FIG. 3d, e).

[0062] Design and In Vitro Validation of a Viral Vector for the Treatment of MSUD

[0063] To maximize transgene expression in liver in vivo we developed optimised AAV expression cassettes coding for either human BCKDHA or BCKDHB. We generated 3 variants of each gene coding sequence (CDS): the wild-type version (WT) and 2 different codon-optimised versions, the first one denominated co1 is a classic optimisation to increase protein expression and the second one, denominated co2 has a reduced CpG content (FIG. 2a, b). This is due to the fact that the reduction or elimination of immunostimulatory CpG sequences in plasmid expression vectors prevents the stimulation of transgene product-specific immune responses without necessarily reducing transgene expression.

[0064] The capsid serotype of choice was AAV8 due to its tropism to the liver; two different promoters, one ubiquitous, the Human elongation factor-1 alpha promoter (EF-1 alpha) (FIG. 2a) and one liver specific, the Human alpha anti-trypsin promoter (hAAT) were chosen to compare the protein expression (FIG. 2b), vector genome copy number (VGCN) in different tissues along with mRNA expression.

[0065] Intravenous EF1a hBCKDHA Allows Long-Term and Sustainable Rescue of Severe MSUD phenotype of Bckdha−/− Mice

[0066] In order to establish a proof of concept of treatment efficacy, we performed systemic intra-temporal injection of hBCKDHA transgene under the control of the ubiquitous promoter EF1α encapsulated in AAV8 at 10.sup.14 vg/kg (further referred to as high dose) at P0, immediately after birth in mice pups. All the pups of the litters were injected, prior to get the genotype results. One pup died at P2 without corpse for genotyping and was not further included in the study. Genotypes were performed at P10. Nine Bckdha−/− pups from 3 litters were injected. Compared to their wild-type and heterozygous littermates they exhibited similar survival and a normal growth (FIG. 3a, b). At age 6 months these 9 pups were still alive without overt phenotypic abnormalities (FIG. 3a, b). The biochemical phenotype was dramatically improved (FIG. 3c). We sacrificed 5 Bckdha−/− mice at 6 months of age. At that age, hBCKDHA protein was detectable mainly in the liver and the heart and present albeit at lower levels in brain and skeletal muscle (FIG. 3d, e, f, g). The remaining 4 Bckdha−/− mice were still alive without overt phenotypic abnormalities at age 12 months.

[0067] Reducing EF1a hBCKDHA Dosage Allows Partial Though Transient Rescue of the MSUD Phenotype in Bckdha−/− Mice

[0068] We performed the same experiment reducing EF1α hBCKDHA in mice to 10.sup.13 vg/kg. Three litters were injected at P0 with EF1α hBCKDHA at 10.sup.13 vg/kg and two as control with EF1α hBCKDHA at 10.sup.14 vg/kg. In the litters injected at 10.sup.13 vg/kg, one pup died at P3 without corpse, one Bckdha.sup.−/− died at P1 probably due to injection failure and one Bckdha.sup.−/− died at P7 of traumatic urine sampling, leaving for the analysis 7 Bckdha.sup.−/−, 9 Bckdha.sup.+/− and 5 Bckdha.sup.+/+ mice. In the litters injected at 10.sup.14 vg/kg, one Bckdha.sup.−/− died at P2 probably due to injection failure and one Bckdha.sup.+/− died at P7 in a context of major growth retardation, leaving for the analysis 2 Bckdha.sup.−/−, 9 Bckdha.sup.+/− and 4 Bckdha.sup.+/+ mice. With the injections at 10.sup.13 vg/kg, we observed a partial and transient recue of the MSUD phenotype in Bckdha−/− mice (N=7) with important inter-individual variability. Five out of seven Bckdha−/− mice showed a normal growth without obvious neurological signs during the first 3 weeks but then stopped growing and developed neurological signs (chiefly ataxia with frequent falls), urging us to sacrifice them at age 4 weeks (FIG. 4a). Two out of seven Bckdha−/− mice showed a more severe evolution, with a growth arrest during the third week followed by weight loss and the development of neurological signs evolving towards a moribund state requiring an anticipated sacrifice before age 4 weeks (FIG. 4a). As expected, the Bckdha−/− mice injected with EF1α hBCKDHA at 10.sup.14 vg/kg displayed a normal growth without any neurological signs (FIG. 4a N=2 Bckdha−/− injected at 10.sup.14 vg/kg and sacrificed at 4 weeks). We observed a similar dose effect at the biochemical level: while the Bckdha−/− mice injected at 10.sup.14 vg/kg displayed normal-high plasma leucine concentrations, the Bckdha−/− mice injected at 10.sup.13 vg/kg and sacrificed at 4 weeks displayed a marked increase in leucine concentrations (FIG. 4b). The two Bckdha−/− mice injected at 10.sup.13 vg/kg and sacrificed before 4 weeks showed a massive increase in leucine concentrations consistent with their worse clinical state, suggesting a loss of treatment efficacy and a relapse of the disease. Western blot analysis showed an important increase of hBCKDHA expression in liver for the two dosages, an increase of hBCKDHA expression in heart with a dose effect and a detectable hBCKDHA expression only at the 10.sup.14 vg/kg dosage in the brain (FIG. 4c, d, e, f).

[0069] Intravenous hAAT hBCKDHA Allows Transient Rescue of the MSUD Phenotype in Bckdha.sup.−/− Mice

[0070] To evaluate the contribution of liver and extra-hepatic tissues to the whole-body BCKDHA enzyme activity responsible for the phenotypic rescue of mice treated with the EF1α hBCKDHA transgene at 10.sup.14 vg/kg, we tested a non-ubiquitous liver-specific promotor (hAAT) with a dosage of 10.sup.13 vg/kg that would be equivalent to 10.sup.14 vg/kg with EF1α in terms of “liver” targeting. We performed systemic intra-temporal injections at P0, immediately after birth in three litters. One Bckdha.sup.−/− died at P1, probably due to injection failure and one Bckdha.sup.+/− died at P12 with a major growth retardation and was not included in the analysis, leaving 5 Bckdha.sup.−/−, 10 Bckdha.sup.+/− and 6 Bckdha.sup.+/+ mice. This treatment allowed a transient rescue of the MSUD phenotype as 5/5 Bckdha.sup.−/− mice survived more than 14 days without overt clinical symptoms. 3/5 Bckdha.sup.−/− mice exhibited a strictly normal growth until P14, followed by a rapid weight loss, appearance of clinical signs (ataxia, frequent falls) evolving towards a moribund state requiring sacrifice at P19 or P21 (FIG. 5). The 2 last Bckdha.sup.−/− mice displayed growth on a lower curve during the 2 first weeks followed by a growth arrest during the third week with neurological deterioration and weight loss at P20 requiring sacrifice at P21. These results suggest that of non-liver tissues effectively contribute to the whole-body BCKDHA activity responsible for the phenotypic rescue of mice treated with the EF1α hBCKDHA transgene at 10.sup.14 vg/kg.

Example 2

[0071] Bckdhb.sup.−/− Mouse Model Recapitulates the Severe Human MSUD Phenotype

[0072] Bckdhb.sup.−/− mouse model recapitulates the severe human MSUD phenotype, displaying a lethal early phenotype (FIG. 6a) with major accumulation of MSUD markers, leucine (FIG. 6b) and alloisoleucine, in plasma (FIG. 6c).

[0073] High dose gene therapy allows rescue of severe MSUD phenotype of Bckdhb.sup.−/− mice with EF1α hBCKDHB transgene. In order to establish a proof of concept of treatment efficacy, we performed systemic intra-temporal injection of hBCKDHB transgene under the control of the ubiquitous promoter EF1a encapsulated in AAV8 at 10.sup.14 vg/kg at P0, immediately after birth in mice pups. Compared to their wild-type and heterozygous littermates Bckdhb.sup.−/− mice exhibited similar survival and a normal growth (FIG. 7a, b) without overt phenotypic abnormalities at age 3 months, with a dramatic improvement of the biochemical phenotype (FIG. 7c).

REFERENCES

[0074] Throughout this application, various references describe the state of the art to which this invention pertains. The disclosures of these references are hereby incorporated by reference into the present disclosure. [0075] Abi-Wardé, M.-T., Roda, C., Arnoux, J.-B., Servais, A., Habarou, F., Brassier, A., Pontoizeau, C., Barbier, V., Bayart, M., Leboeuf, V., et al. (2017). Long-term metabolic follow-up and clinical outcome of 35 patients with maple syrup urine disease. J. Inherit. Metab. Dis. 40, 783-792. [0076] Baruteau, J Perocheau, D. P., Hanley, J Lorvellec, M., Rocha-Ferreira, E., Karda, R., Ng, J Suff, N., Diaz, J. A., Rahim, A. A., et al. (2018). Argininosuccinic aciduria fosters neuronal nitrosative stress reversed by Asl gene transfer. Nat. Commun. 9, 3505. [0077] Bodner-Leidecker, A., Wendel, U., Saudubray, J. M., and Schadewaldt, P. (2000). Branched-chain L-amino acid metabolism in classical maple syrup urine disease after orthotopic liver transplantation. J. Inherit. Metab. Dis. 23, 805-818. [0078] Bouchereau, J., Leduc-Leballeur, J., Pichard, S., Imbard, A., Benoist, J.-F., Abi Warde, M.-T., Arnoux, J.-B., Barbier, V., Brassier, A., Broué, P., et al. (2017). Neurocognitive profiles in MSUD school-age patients. J. Inherit. Metab. Dis. 40, 377-383. [0079] Chandler, R. J., and Venditti, C. P. (2019). Gene Therapy for Methylmalonic Acidemia: Past, Present, and Future. Hum. Gene Ther. [0080] Chandler, R. J., Tarasenko, T. N., Cusmano-Ozog, K., Sun, Q., Sutton, V. R., Venditti, C. P., and McGuire, P. J. (2013). Liver-directed adeno-associated virus serotype 8 gene transfer rescues a lethal murine model of citrullinemia type 1. Gene Ther. 20, 1188-1191. [0081] Cunningham, S. C., Spinoulas, A., Carpenter, K. H., Wilcken, B., Kuchel, P. W., and Alexander, I. E. (2009). AAV2/8-mediated correction of OTC deficiency is robust in adult but not neonatal Spf(ash) mice. Mol. Ther. J. Am. Soc. Gene Ther. 17, 1340-1346. [0082] Ginocchio, V. M., Ferla, R., Auricchio, A., and Brunetti-Pierri, N. (2019). Current Status on Clinical Development of Adeno-Associated Virus-Mediated Liver-Directed Gene Therapy for Inborn Errors of Metabolism. Hum. Gene Ther. [0083] Grisch-Chan, H. M., Schwank, G., Harding, C. O., and Thöny, B. (2019). State-of-the-Art 2019 on Gene Therapy for Phenylketonuria. Hum. Gene Ther. [0084] Homanics, G. E., Skvorak, K., Ferguson, C., Watkins, S., and Paul, H. S. (2006). Production and characterization of murine models of classic and intermediate maple syrup urine disease. BMC Med. Genet. 7, 33. [0085] Lee, E. K., Hu, C., Bhargava, R., Rozengurt, N., Stout, D., Grody, W. W., Cederbaum, S. D., and Lipshutz, G. S. (2012). Long-term survival of the juvenile lethal arginase-deficient mouse with AAV gene therapy. Mol. Ther. J. Am. Soc. Gene Ther. 20, 1844-1851. [0086] Mazariegos, G. V., Morton, D. H., Sindhi, R., Soltys, K., Nayyar, N., Bond, G., Shellmer, D., Shneider, B., Vockley, J., and Strauss, K. A. (2012). Liver transplantation for classical maple syrup urine disease: long-term follow-up in 37 patients and comparative United Network for Organ Sharing experience. J. Pediatr. 160, 116-121.e1. [0087] Muelly, E. R., Moore, G. J., Bunce, S. C., Mack, J., Bigler, D. C., Morton, D. H., and Strauss, K. A. (2013). Biochemical correlates of neuropsychiatric illness in maple syrup urine disease. J. Clin. Invest. 123, 1809-1820. [0088] Nathwani, A. C., Reiss, U. M., Tuddenham, E. G. D., Rosales, C., Chowdary, P., McIntosh, J., Della Peruta, M., Lheriteau, E., Patel, N., Raj, D., et al. (2014). Long-term safety and efficacy of factor IX gene therapy in hemophilia B. N. Engl. J. Med. 371, 1994-2004. [0089] S Sonnet, D., N O'Leary, M., A Gutierrez, M., M Nguyen, S., Mateen, S., Hsu, Y., P Mitchell, K., J Lopez, A., Vockley, J., K Kennedy, B., et al. (2016). Metformin inhibits Branched Chain Amino Acid (BCAA) derived ketoacidosis and promotes metabolic homeostasis in MSUD. Sci. Rep. 6, 28775. [0090] Strauss, K. A., Puffenberger, E. G., and Morton, D. H. (2013). Maple Syrup Urine Disease. In GeneReviews®, M. P. Adam, H. H. Ardinger, R. A. Pagon, S. E. Wallace, L. J. Bean, K. Stephens, and A. Amemiya, eds. (Seattle (Wash.): University of Washington, Seattle), p. [0091] Suryawan, A., Hawes, J. W., Harris, R. A., Shimomura, Y., Jenkins, A. E., and Hutson, S. M. (1998). A molecular model of human branched-chain amino acid metabolism. Am. J. Clin. Nutr. 68, 72-81. [0092] Wendel, U., Saudubray, J. M., Bodner, A., and Schadewaldt, P. (1999). Liver transplantation in maple syrup urine disease. Eur. J. Pediatr. 158 Suppl 2, S60-64. [0093] Yoneshiro, T., Wang, Q., Tajima, K., Matsushita, M., Maki, H., Igarashi, K., Dai, Z., White, P. J., McGarrah, R. W., Ilkayeva, O. R., et al. (2019). BCAA catabolism in brown fat controls energy homeostasis through SLC25A44. Nature 572, 614-619. [0094] Zinnanti, W. J., Lazovic, J., Griffin, K., Skvorak, K. J., Paul, H. S., Homanics, G. E., Bewley, M. C., Cheng, K. C., Lanoue, K. F., and Flanagan, J. M. (2009). Dual mechanism of brain injury and novel treatment strategy in maple syrup urine disease. Brain J. Neurol. 132, 903-918.