Tuesday, October 20, 2015

POLG heterozygote

I raised the question below:
A 4 yo boy developed brief episodes of unresponsiveness.  Past history unremarkable and development normal.  EEG showed generalized and focal paroxysmal abnormalities.  Levetiracetam initiated without good effect at apex dosage of 650 mg bid (~16 kg).  Valproate was then added. On 250-125-250 mg had VPA level  of  66.  AST 31.  VPA increased to 250 mg tid.  One week later, the patient experienced anorexia and vomiting.  Repeat level  122.  AST 399 ALT 261.  VPA stopped.  Two days later AST 208 ALT 218. One week later ALT 64 AST 39.  Therapy with topiramate initiated.  Subsequently, LVT stopped and ethosuximide initiated.  Seizure control on TPM and ETX thus far has been relatively good.
Genetic testing shows a heterozygous disease-causing POLG mutation (c.3151 G>C).

Did this cause the scenario above?

A colleague replied: 

Very interesting, I wish I had a definitive answer!

The POLG c.3151C>G corresponds to the Gly1051Arg change, which is listed as a disease causing mutation (autosomal recessive for this allele).  There have been cases where there has been a deletion of the other copy of the gene (most NextGen and other commercial technologies will pick this up) and mechanisms exist where the other copy may not be expressed -- this takes more investigation and not easy to perform outside of a research lab.  And there are several other genes that conceivably behave in a similar fashion.  (The list can be found at http://www.ncbi.nlm.nih.gov/books/NBK26471/  ).  Although many kids with POLG disease have some developmental delays before their seizures, not all do. 

I would consider repeating the neuroimaging in a month or two (looking for evolving atrophy or other pathology), grab a CSF folate (many of these mtDNA depletion disorders will show a decrease in CSF folate) and consider broadening the search for disease to include the mtDNA depletion disorders. 

Please let us know how this evolves.  I have seen several such cases (one abnormal copy) and have not reached a good understanding in these kids --- but they all were never challenged again with VPA. 

I hope this helps.


  1. Li S, Guo J, Ying Z, Chen S, Yang L, Chen K, Long Q, Qin D, Pei D, Liu X.
    Valproic acid-induced hepatotoxicity in Alpers syndrome is associated with
    mitochondrial permeability transition pore opening-dependent apoptotic
    sensitivity in an induced pluripotent stem cell model. Hepatology. 2015


    Valproic acid (VPA) is widely used to treat epilepsy, migraine, chronic headache, bipolar disorder, and as adjuvant chemotherapy, but potentially causes idiosyncratic liver injury. Alpers-Huttenlocher syndrome (AHS), a neurometabolic disorder caused by mutations in mitochondrial DNA polymerase gamma (POLG), is associated with an increased risk of developing fatal VPA hepatotoxicity. However, the mechanistic link of this clinical mystery remains unknown. Here, fibroblasts from 2 AHS patients were reprogrammed to induced pluripotent stem cells (iPSCs) and then differentiated to hepatocyte-like cells (AHS iPSCs-Hep). Both AHS iPSCs-Hep are more sensitive to VPA-induced mitochondrial-dependent apoptosis than controls, showing more activated caspase-9 and cytochrome c release. Strikingly, levels of both soluble and oligomeric optic atrophy 1, which together keep cristae junctions tight, are reduced in AHS iPSCs-Hep. Furthermore, POLG mutation cells show reduced POLG expression, mitochondrial DNA (mtDNA) amount, mitochondrial adenosine triphosphate production, as well as abnormal mitochondrial ultrastructure after differentiation to hepatocyte-like cells. Superoxide flashes, spontaneous bursts of superoxide generation, caused by opening of the mitochondrial permeability transition pore (mPTP), occur more frequently in AHS iPSCs-Hep. Moreover, the mPTP inhibitor, cyclosporine A, rescues VPA-induced apoptotic sensitivity in AHS iPSCs-Hep. This result suggests that targeting mPTP opening could be an effective method to prevent hepatotoxicity by VPA in AHS patients. In addition, carnitine or N-acetylcysteine, which has been used in the treatment of VPA-induced hepatotoxicity, is able to rescue VPA-induced apoptotic sensitivity in AHS iPSCs-Hep.


    AHS iPSCs-Hep are more sensitive to the VPA-induced mitochondrial-dependent apoptotic pathway, and this effect is mediated by mPTP opening. Toxicity models in genetic diseases using iPSCs enable the evaluation of drugs for therapeutic targets.

  2. An 8 year old boy with a history of autism presented in early June with super refractory seizures, treated with pentobarbital coma, among other interventions. After stabilization of his epilepsy, he was transferred to our facility for rehabilitation care. Molecular genetic testing for POLG deficiency showed 2 pathogenic mutations, c.1399G>A (p.A467T) and c.3286C>T (p.R1096C). c.1399G>A (p.A467T) is a common pathogenic mutation. c.3286C>T (p.R1096C) has been reported in a homozygous state with eight Alpers syndrome patients and in a compund heterozygous state with autosomal recessaive PEO. The patient's course was one of progressive disability leading to hospice status and then his death yesterday.