Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2016 Jul 29;11(7):e0160447.
doi: 10.1371/journal.pone.0160447. eCollection 2016.

A Spontaneous Missense Mutation in Branched Chain Keto Acid Dehydrogenase Kinase in the Rat Affects Both the Central and Peripheral Nervous Systems

Free PMC article

A Spontaneous Missense Mutation in Branched Chain Keto Acid Dehydrogenase Kinase in the Rat Affects Both the Central and Peripheral Nervous Systems

J Samuel Zigler Jr et al. PLoS One. .
Free PMC article


A novel mutation, causing a phenotype we named frogleg because its most obvious characteristic is a severe splaying of the hind limbs, arose spontaneously in a colony of Sprague-Dawley rats. Frogleg is a complex phenotype that includes abnormalities in hind limb function, reduced brain weight with dilated ventricles and infertility. Using micro-satellite markers spanning the entire rat genome, the mutation was mapped to a region of rat chromosome 1 between D1Rat131 and D1Rat287. Analysis of whole genome sequencing data within the linkage interval, identified a missense mutation in the branched-chain alpha-keto dehydrogenase kinase (Bckdk) gene. The protein encoded by Bckdk is an integral part of an enzyme complex located in the mitochondrial matrix of many tissues which regulates the levels of the branched-chain amino acids (BCAAs), leucine, isoleucine and valine. BCAAs are essential amino acids (not synthesized by the body), and circulating levels must be tightly regulated; levels that are too high or too low are both deleterious. BCKDK phosphorylates Ser293 of the E1α subunit of the BCKDH protein, which catalyzes the rate-limiting step in the catabolism of the BCAAs, inhibiting BCKDH and thereby, limiting breakdown of the BCAAs. In contrast, when Ser293 is not phosphorylated, BCKDH activity is unchecked and the levels of the BCAAs will decrease dramatically. The mutation is located within the kinase domain of Bckdk and is predicted to be damaging. Consistent with this, we show that in rats homozygous for the mutation, phosphorylation of BCKDH in the brain is markedly decreased relative to wild type or heterozygous littermates. Further, circulating levels of the BCAAs are reduced by 70-80% in animals homozygous for the mutation. The frogleg phenotype shares important characteristics with a previously described Bckdk knockout mouse and with human subjects with Bckdk mutations. In addition, we report novel data regarding peripheral neuropathy of the hind limbs.

Conflict of interest statement

Competing Interests: The authors have declared that no competing interests exist.


Fig 1
Fig 1. Linkage map of the frogleg locus on rat chromosome 1.
Ideogram of rat chromosome 1, showing polymorphic markers in the region of the Bckdk gene. Nucleotide sequence positions were determined by locating original marker amplimer sequences to the RGSC 6.0 / rn6 July 2014 assembly of the rat genome. MLINK-derived maximum LOD scores and corresponding theta values indicate linkage distance to the disease locus. The analysis was based on 11 affected and 10 unaffected individuals. Significant linkage scores identifying the initial disease interval (> 3.0) are enclosed in the box.
Fig 2
Fig 2. The effect of the G369E mutation on the structure of BCKDK protein is shown.
The prediction is based on homology modeling of the structures of BCKDK and the G369E mutant as refined using the 7.5 ns molecular dynamics and shown in beige and light blue, respectively. In the mutant, the helix containing residue E369 is moving outward from the ADP binding site due to repulsion from the negatively charged residue E275. This might suggest a decrease in ADP binding activity by the mutant enzyme due to the replacement of glycine in normal BCKDK by the negatively charged glutamic acid in the mutant variant.
Fig 3
Fig 3. Electrophoresis and western analysis of protein lysates from the brains of frogleg and littermate rats.
In panel A are shown the staining patterns for 5 individual rats: lane 1 is lysate from a wild type, lanes 2 and 3 are from rats heterozygous for the Bckdk mutation, and lanes 4 and 5 are from homozygotes. In panel B equal aliquots of the same 5 samples were electrophoresed, transferred to nitrocellulose and probed with an antibody to BCKDH subunit E1α. It is evident that similar amounts of the total enzyme are present in all 5 samples. In panel C the amount of BCKDH phosphorylated at Ser293 of the E1α subunit is determined by probing the same 5 lysates with an antibody specific for the phosphorylated form of the enzyme. While the wildtype and heterozygous animals have a clear band, no immunoreactivity is detected in the 2 samples homozygous for the mutation. To demonstrate that the bands present on panel C represent the phosphorylated enzyme, aliquots of the same lysates were treated with calf intestinal alkaline phosphatase before electrophoresis. The bands present in lanes 1–3 of panel C are no longer visible.
Fig 4
Fig 4. Appearance of frogleg homozygous rat at 20 days of age.
The animal on the left shows the typical appearance of the frogleg pup, relative to a non-frogleg littermate on the right. Note the small size, the thin, rough coat, and the abnormal rotation of the hind limbs of the affected animal.
Fig 5
Fig 5
A) Impairment of the hind limbs of frogleg rats (bottom two panels) compared to wild type Sprague Dawley rats at 4 months (top two panels). B) Left hind limbs of a wildtype (left) and a frogleg (right) rat. Note the inner rotation and extension of the frogleg rat compared with the normal alignment and orientation of the wild type. Also, note the similarity in muscle mass. The cartoon at the bottom of the figure demonstrates the joint alignment and orientation. C) The hind paw of a wild type and a frogleg rat are shown in (left) and (right), respectively. Note the in-turning (pronated) ankles in frogleg with hyperkeratosis of foot pads and toenail overgrowth (arrow), likely related to abnormal or reduced wear.
Fig 6
Fig 6
A. Brains from wild type (left) and frogleg (right) were of similar size. Scale is in centimeters. B. Brain weights, however, differed in wild type (+/+) and homozygote (frogleg) rats of 4 weeks, 10 weeks and 7 months of age, being considerably smaller in the frogleg rats compared to their littermates at all ages tested (mean+SEM).
Fig 7
Fig 7
2D coronal T2 weighted images of representative wild type (A) and frogleg (B) brains at P65. In (C) and (D) 3D reconstructions of the P65 lateral ventricles are shown. Dramatic difference in ventricular size is clearly observed. In (E) and (F) 2D coronal T2 weighted images of wild type and frogleg brains, respectively, are shown at P110. In (G) and (H) the 3D reconstructions for P110 are shown. The difference of ventricular size is not as great as that shown in (C) and (D). Red arrows (B, F) indicate the enlarged ventricles of frogleg brains.
Fig 8
Fig 8
A. Averaged total brain volume of wild type and frogleg rats at P65 and P110; B. Averaged ventricular size of wild type and frogleg rats at P65 and P110. Error bars represent SD. * P< 0.05 (relative to wild type).
Fig 9
Fig 9. Wild type and frogleg testes and epididymis, paraffin, 5μ sections, H and E.
A. Wild type testis, with normal spermatogenesis. B. Frogleg testis, multiple seminiferous tubules with no spermatogenesis; only 1 tubule (arrow) with active spermatogenesis. C. Frogleg epididymis, reduced sperm (oligospermia) with clumped and degenerating cells. Scale bars = 100μm in A and B; 200μm in C.
Fig 10
Fig 10
A. Sciatic compound muscle action potentials (CMAP) are markedly reduced in two representative 7 month old frogleg rats when compared with an age-matched control (left panel). A similar decrease in CMAP was also evident in frogleg rats at 10 weeks of age (right). B. CMAP latencies at both 7 months and 10 weeks of age were marginally higher in frogleg rats. Error bars represent SEM. * P< 0.05 (relative to control).
Fig 11
Fig 11. Transmission electron microscopy images of cross sections of hind limb nerves from a frogleg rat.
A. Sciatic nerve showing evidence of rare active demyelination (arrow). In the ventral root (B) of the sciatic nerve (as well as in the nerve itself) there were axons that were thinly myelinated (arrows in B). In the sural nerve, there was evidence of rare Wallerian-like degeneration, intra-axonal inclusions, and denervated Schwann cells. Scale bar = 2 microns.
Fig 12
Fig 12. Immunolabeling of neuromuscular junctions of soleus and extensor digitorum longus (EDL) muscles.
Representative data for muscles isolated from 7 month old frogleg rat are shown in Panels A for the EDL and Panels B for the soleus. For the EDL, panel A shows staining (green) with antibodies to neurofilament (NF200) and synaptic vesicle protein (SV2), while panel A’ shows staining of the nicotinic acetylcholine receptors at the junction with α-bungarotoxin (red). A” shows the merged image, with yellow indicating normal innervation of each of the 5 junctions shown. In contrast, for the soleus, panel B indicates very little immunoreactivity with SV2 or NF200. Staining of the nicotinic acetylcholine receptors in B’ is normal, but in the merged image (B”) only 2 of the 7 numbered junctions show evidence of partial innervation (arrowheads at numbers 5 and 6), with the other 5 being fully denervated. Scale bar = 50 μm. Panels C and D show the cumulative data for soleus and EDL muscles respectively, for 7 month old animals. Black bars are for wild type and red bars for frogleg animals. In WT EDL and soleus muscles, 97.1% and 98.6% of NMJs are intact. In contrast, while the EDL NMJs were 89.8% intact in the frogleg mutant at 7M, only 4.4% of the NMJs for the soleus remained intact at that age. Signs of denervation were predominant in soleus muscles, with NMJs being either partially (42.6%) or completely (51.1%) denervated. Error bars = SD; * P< 0.05 (Relative to control). In panel E, the time course of denervation in the soleus muscle for frogleg and wild type rats aged 1–7 months is shown.

Similar articles

See all similar articles


    1. Hutson SM. The case for regulating indispensable amino acid metabolism: the branched-chain alpha-keto acid dehydrogenase kinase-knockout mouse. Biochem J. 2006; 400(1):e1–3. - PMC - PubMed
    1. Joshi MA, Jeoung NH, Obayashi M, Hattab EM, Brocken EG, Liechty EA, et al. Impaired growth and neurological abnormalities in branched-chain alpha-keto acid dehydrogenase kinase-deficient mice. Biochem J. 2006; 400(1):153–62. - PMC - PubMed
    1. Novarino G, El-Fishawy P, Kayserili H, Meguid NA, Scott EM, Schroth J, et al. Mutations in BCKD-kinase lead to a potentially treatable form of autism with epilepsy. Science. 2014; 343(6170):506–11. - PubMed
    1. Cook KG, Bradford AP, Yeaman SJ, Aitken A, Fearnley IM, Walker JE. Regulation of bovine kidney branched-chain 2-oxoacid dehydrogenase complex by reversible phosphorylation. Eur J Biochem. 1984; 145(3):587–91. - PubMed
    1. Harris RA, Joshi M, Jeoung NH. Mechanisms responsible for regulation of branched-chain amino acid catabolism. Biochem Biophys Res Commun. 2004; 313(2):391–396. - PubMed

MeSH terms


Grant support

DS received support from National Institutes of Health, National Institutes of Child Health and Human Disease, R21HD059008, AH received support from Dr. Miriam and Sheldon G. Adelson Medical Research Foundation, The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

LinkOut - more resources