Carnitine palmitoyltransferase II deficiency

The authors investigated 32 patients with the muscle form of CPT II deficiency. Total carnitine palmitoyltransferase enzyme system (CPT) activity was normal but abnormally inhibited by malonyl-CoA, palmitoyl-CoA, and the detergents Triton X and Tween 20. Mutation analysis identified three described mutations (S113L, P50H, and F448L) and two novel mutations (M214T and Y479F). Using modeling techniques, a structure could be identified anchoring the protein in the membrane. Only one of the five mutations (Y479F) is located within this region. NEUROLOGY 2003;60:1351–1353 The carnitine palmitoyltransferase enzyme system (CPT), consisting of CPT I and CPT II, mediates the entry of long-chain fatty acids into the mitochondrial matrix for -oxidation.1 Adult-onset CPT II deficiency, characterized by exercise-induced muscle pain and weakness, rhabdomyolysis, and paroxysmal myoglobinuria, is the most common disorder of lipid metabolism affecting the skeletal muscle.2 The human CPT II gene is located at chromosome 1, spans 20 kb, and contains five exons. A common mutation, S113L, is present in approximately 60% of alleles in addition to a multitude of rare, obviously private mutations.3 There are conflicting reports on the biochemical consequences of CPT II deficiency caused by these mutations.4,5 Patients and methods. Thirty-two patients with CPT II deficiency and two asymptomatic parents were investigated. Control muscle specimens were obtained from patients who had muscle biopsies for diagnosis of muscular symptoms. Patients were deemed to be “normal controls” if they were ultimately found to have no muscle disease according to combined clinical, electromyographic, biochemical, and histologic criteria. Informed consent was obtained from all patients. Total CPT activity was determined using the isotope forward assay as described.4 The physiologic inhibitors malonyl-CoA and palmitoylcarnitine were added to the reaction mixture as indicated. In some experiments, membranes were disrupted by preincubation with Triton X-100 or Tween 20. Genomic DNA was prepared from muscle tissue or blood using standard procedures. Polymorphisms and the S113L mutation were screened as published.5 Primer pairs were synthesized to amplify the entire coding region and completely sequenced. Both strands of genomic DNA were analyzed to confirm a mutation. Restriction enzyme digestion-based assays were established to screen control subjects. Protocols have been published previously.6 Homology searching and alignment of the CPT II sequence was done with the FastA alogarithm.8 Using the PHD neural network method of Rost and Sander, secondary structure and surface accessibility prediction was performed. The surface polarity of predicted helices was checked with an automated helical wheel projection procedure. Three-dimensional modeling of the -helix bundle was done manually with the INSIGHT II software package (Molecular Simulations Inc., San Diego, CA). Results. In all patients, CPT activity using the isotope forward assay was normal. Addition of the inhibitors malonyl-CoA and palmitoylcarnitine or the detergents Triton X and Tween decreased CPT activity in control subjects and patients, but the inhibition was significantly greater in patients than in control subjects (see table E1 on the Neurology Web site). Fourteen index patients were homozygous and 17 were heterozygous for common the S113L mutation. One was homozygous for the P50H mutation. Two other patients were compound heterozygous for the S113L mutation and the P50H mutation. A T-to-C transition in nucleotide position 742 (nucleotide position according to Finocchiaro3) was found once (M214T); an A-to-T transition in position 1547 could be identified leading to the substitution of a tyrosine to a phenylalanine at position 479 (Y479F); and a T-to-C transition in position 1443 leading to a substitution of phenylalanine to leucine at position 448 (F448L) was found in three unrelated patients. This mutation was associated with a 2-base pair (bp) deletion at position 1339 and 1340 (413delAG). Fifty-five healthy control subjects did not have one of the described mutations. All results are summarized in table E-1. A sequence database search found no other proteins homologous to CPT II except for other carnitine C-acyl transferases. However, when the homology search was repeated on short domains, a region was found that shares 50% identical residues and 75% similarity to a membrane-located helix of a mitochondrial ATPase. The domain F448-G497, comprising helix D464-Y479 and a second predicted helix (S488-H496) along with a region N-terminal to D464-Y479, is highly conserved among carnitine acyltransferases. For both helices, a low solvent accessibility is Additional material related to this article can be found on the Neurology Web site. Go to www.neurology.org and scroll down the Table of