SEARCH

SEARCH BY CITATION

Keywords:

  • Glucose transporter;
  • Lactate transporter;
  • Glucose metabolism;
  • Generalized epilepsy;
  • GLUT1 deficiency

Summary

  1. Top of page
  2. Summary
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References
  9. Biography
  10. Supporting Information

The availability of glucose, and its glycolytic product lactate, for cerebral energy metabolism is regulated by specific brain transporters. Inadequate energy delivery leads to neurologic impairment. Haploinsufficiency of the glucose transporter GLUT1 causes a characteristic early onset encephalopathy, and has recently emerged as an important cause of a variety of childhood or later-onset generalized epilepsies and paroxysmal exercise-induced dyskinesia. We explored whether mutations in the genes encoding the other major glucose (GLUT3) or lactate (MCT1/2/3/4) transporters involved in cerebral energy metabolism also cause generalized epilepsies. A cohort of 119 cases with myoclonic astatic epilepsy or early onset absence epilepsy was screened for nucleotide variants in these five candidate genes. No epilepsy-causing mutations were identified, indicating that of the major energetic fuel transporters in the brain, only GLUT1 is clearly associated with generalized epilepsy.

The glucose transporter GLUT1 plays an essential role in brain metabolism by facilitating transport of glucose in the neurovascular unit. Mutations in the SLC2A1 gene that encodes GLUT1 lead to deficiency of the transporter and an expanding spectrum of neurologic disorders. Originally described as a cause of a severe encephalopathy,[1] GLUT1 deficiency has recently been shown to account for 10% of patients with early onset absence epilepsy (EOAE),[2, 3] 5% of those with myoclonic astatic epilepsy (MAE),[4] and 1% of all individuals with genetic generalized epilepsies (GGEs).[5] In addition to seizures, subjects with more severe phenotypes may have developmental delay or ataxia. Others may have paroxysmal exercise-induced dyskinesia (PED) with or without epilepsy. An early diagnosis of GLUT1 deficiency is important as it is managed with the ketogenic diet, which typically results in seizure control and may improve cognition.

Because glucose metabolism is critical for neural activity, we hypothesized that mutation of other critical transporters in the cerebral energy metabolism pathways (Fig. 1) may also cause generalized epilepsies. In phenotypes known to have a significant association with GLUT1 deficiency we used Sanger sequencing to screen the other major cerebral glucose (GLUT3) and lactate (MCT1/2/3/4) transporters.

image

Figure 1. Glucose and lactate transport pathways. Graphic showing a simplified version of the major glucose and lactate transport pathways in the brain. The key transfer steps facilitated by the GLUT and MCT transporters are illustrated. Gluc, glucose; Pyr, pyruvate; La, lactate; Glu, glutamate; ECF, extracellular fluid. Adapted from.[9]

Download figure to PowerPoint

Materials and Methods

  1. Top of page
  2. Summary
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References
  9. Biography
  10. Supporting Information

Clinical phenotyping

We selected 119 patients with the two generalized epilepsy syndromes most frequently associated with GLUT1 deficiency—myoclonic astatic epilepsy (n = 80) and early onset absence epilepsy (n = 39)—from the Epilepsy Research Centre database. In all patients, nucleotide and copy number variations in SLC2A1 (GLUT1) had been excluded as part of previous studies.[2-4] The phenotype of MAE was specifically selected, as this disorder is regarded as the epilepsy syndrome most responsive to the ketogenic diet.[4] Myoclonic-astatic epilepsy was defined by onset of multiple afebrile, generalized seizure types between 6 months and 6 years, generalized spike-wave (>2.5 Hz) on electroencephalography (EEG), and at least one of myoclonic, atonic, or myoclonic-atonic seizures. Patients with tonic seizures were excluded. This is in line with our previously published work on GLUT1 deficiency[4] and based on the definition put forward by Oguni and colleagues.[6] Early onset absence epilepsy was defined by onset of typical absence seizures before age 4 years as the major seizure type and generalized spike-wave (>2.5 Hz) on EEG. Those with atonic, myoclonic-atonic, or tonic seizures were excluded from EOAE.[3] Detailed electroclinical information was obtained regarding the epileptology, developmental course, and medical and family history.[2, 4] Genomic DNA was extracted from venous blood by standard methods.

The Austin Health Human Research Ethics Committee, Melbourne, Australia, approved this study (Project No. H2007/02961). Informed consent was obtained from all subjects, or their parents and carers in the case of minors or those with intellectual disability.

Polymerase chain reaction and Sanger sequencing

Five candidate genes of the solute carrier (SLC) transporter family—SLC2A3 (GLUT3), SLC16A1 (MCT1), SLC16A7 (MCT2), SLC16A8 (MCT3), and SLC16A3 (MCT4)—encoding all major brain glucose or lactate transporters except GLUT1 (Fig. 1) were amplified using gene-specific primers designed to the reference human gene transcripts (NCBI Gene; http://www.ncbi.nlm.nih.gov/). Primer sequences are available on request. Amplification reactions were cycled using a standard protocol on a Veriti Thermal Cycler (Applied Biosystems, Carlsbad, CA, U.S.A.). Bidirectional sequencing of all exons and flanking regions including splice sites was completed with a BigDye v3.1 Terminator Cycle Sequencing Kit (Applied Biosystems), according to the manufacturer's instructions. Sequencing products were resolved using a 3730xl DNA Analyzer (Applied Biosystems). All sequencing chromatograms were compared to published complementary DNA (cDNA) sequence; nucleotide changes were detected using Codon Code Aligner (CodonCode Corporation, Dedham, MA, U.S.A.).

Results

  1. Top of page
  2. Summary
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References
  9. Biography
  10. Supporting Information

Clinical syndromes

The EOAE cohort had a median onset age of 18 months and, of those with sufficient data, 22 (63%) of 35 were refractory to trials of two appropriate medicines in adequate doses. The MAE cohort was more refractory, with 56 (80%) of the 70 with sufficient data showing no response to two medications. Around 60% (46/75) of MAE cases had an encephalopathic course, with a poor intellectual outcome following initial regression not reversed by treatment. Of the MAE cases, 19 (24%) of 80 received the ketogenic diet, mainly those diagnosed in the last 10 years when this treatment was more easily available. The majority of those with follow-up data showed at least a 50% seizure response to the diet (11/16, 69%), whereas two (2/16, 13%) had complete remission with the diet.

Candidate gene screening

Screening of SLC2A3 revealed three heterozygous coding substitutions; a novel missense change, p.Ile14Leu, that is likely to be benign because isoleucine and leucine are functionally equivalent amino acids, and two known coding polymorphisms (p.Arg91His and p.Leu258Leu; Table 1). In SLC16A1 we identified only one known heterozygous coding polymorphism, p.Arg490Glu, whereas in SLC16A7 we detected four heterozygous coding polymorphisms, none of which appear to be disease-causing based on their frequency in control populations (Table 1).

Table 1. Gene variants identified in our patient cohort
GeneProteinLocationTypeDNA levelProtein levelSNP IDa
  1. a

    Single nucleotide polymorphism identification (rs) number.

  2. b

    Present at low frequency on Exome Variant Server database but not assigned an rs number.

SLC2A3GLUT3CodingMissensec.40A > Cp.Ile14leu
SLC2A3 GLUT3CodingMissensec.272G > Ap.Arg91Hisrs145936296
SLC2A3 GLUT3CodingSynonymousc.774A > Gp.Leu258Leurs17847967
SLC16A1 MCT1CodingMissensec.1470T > Ap.Asp490Glurs1049434 
SLC16A7 MCT2CodingSynonymousc.334C > Tp.Leu112Leurs200049323
SLC16A7 MCT2CodingMissensec.656G > Tp.Ser219Ilers142586562
SLC16A7 MCT2CodingMissensec.1333A > Tp.Thr445Serrs3763980
SLC16A7 MCT2CodingSynonymousc.1383C > Tp.Asn461Asnrs3763979
SLC16A8 MCT3Noncoding5′UTRc.-5C > T
SLC16A8 MCT3CodingMissensec.124T > Cp.Phe42Leu
SLC16A8 MCT3NoncodingSplicingc.214+1G > Crs77968014
SLC16A8 MCT3CodingMissensec.1384G > Tp.Asp462Tyrrs144999316
SLC16A8 MCT3CodingMissensec.1450G > Ap.Ala484Thr
SLC16A3 MCT4CodingMissensec.883G > Ap.Val295Ile b
SLC16A3 MCT4CodingSynonymousc.831G > Ap.Ala277Ala
SLC16A3 MCT4CodingSynonymousc.1362C > Tp.Asn454Asnrs112638041

We detected most variation in the SLC16A8 gene with five heterozygous variants found including two known polymorphisms and three novel changes (Table 1). Of the novel changes one was located in the 5′UTR (c.-5C > T) and two were missense changes (p.Phe42Leu and p.Ala484Thr). Parents were available and segregation analysis was completed for all five variants; however, all were found to be inherited from unaffected parents, making it unlikely that any are disease-causing.

In SLC16A3, three heterozygous variants were detected only one of which (p.Ala277Ala) was novel but synonymous. Although rare, the p.Val295Ile variant in SLC16A3 is predicted to be benign by standard pathogenicity prediction software (PolyPhen-2; http://genetics.bwh.harvard.edu/pph2/) and the substitution has a low Grantham score of 29, suggesting that the variant is unlikely to be damaging to the MCT4 protein.

The SLC2A3 and SLC16A7 genes are predicted to be tolerant to functional change according to their residual variant intolerance scores (RVIS) of 0.619 and 0.375, respectively.[7] Although the SLC16A1 (RVIS = −0.072) is close to expectation and the SLC16A1 (RVIS = −0.641) gene is predicted to be intolerant to functional change, no RVIS was available for SLC16A8. These intolerance predictions correlate well with the number of variations identified for each gene (Table 1).

Discussion

  1. Top of page
  2. Summary
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References
  9. Biography
  10. Supporting Information

The discovery of mutations in SLC2A1 in a wide spectrum of mild to severe generalized epilepsies was unexpected but important because of treatment implications. However, our hypothesis that similar cases are due to mutations in other glucose and lactate transporters was not supported.

Our data suggest that GLUT1 is the only energy metabolism transporter associated with generalized epilepsy. This could be because it is the only glucose transporter present in the neurovascular unit and thus the rate-limiting step in brain glucose transport. However, GLUT1 is also expressed in astrocytes. Although other members of the GLUT family are expressed in the brain, only GLUT1 and GLUT3 are thought to contribute significantly to brain glucose transport (reviewed in[8]). The exclusive expression of GLUT3 in neurons makes it an excellent candidate for an epilepsy gene; however, we were unable to find GLUT3 mutations in the generalized epilepsy cohort screened here. Similarly, if energy delivery to neurons is dependent on the astrocyte-neuron lactate shuttle,[8] then the lactate transporters would also be attractive epilepsy gene candidates. Although no mutations were identified, it may be that these transporters are so crucial that any deficiency is compensated for by the presence of other known or as yet unidentified transporters. In some regions of the brain, multiple transporters are expressed, as shown in Fig. 1, providing the opportunity for redundancy. Such compensation has been observed in a mouse model of Glut1 haploinsufficiency; developmental expression of Mct1 and Mct2 was increased twofold in the neonatal brain in response to reduced Glut1 in heterozygous null mice.[9]

Studies from other animal models are broadly consistent with our findings. Null heterozygous Glut3 mice exhibited behavioral features suggesting cognitive and social impairments as well as electrographic discharges but no clinical generalized seizures.[10] Of interest, haploinsufficiency of Glut3 was associated with increased concentration of microvascular/glial Glut1 and Mct2, no change in brain glucose, and enhanced lactate uptake. Likewise, in an independent study, Glut3 haploinsufficiency did not impair mouse brain glucose uptake or utilization.[11] In a study of monocarboxylate transporters targeted deletion of Mct3 in mouse altered visual function; however, there were no overt seizures, although these were not specifically tested for.[12]

This study has some limitations. First, certain types of pathogenic variants that we could not detect by Sanger sequencing, such as large copy number variants (CNVs), may exist in these genes. However, this is unlikely, since most mutations identified in GLUT1 deficiency are single nucleotide variants that are readily detectable by our methods and often cause moderate to severe disease. Second, the infrequent inherited gene variants we found may reduce the amount of transporter, but on their own are insufficient to cause disease. Such variants could still behave as susceptibility variants. To test this would require functional studies and, as a common variant association study, a much larger sample size. Third, we intentionally screened the two generalized epilepsy syndromes most frequently associated with GLUT1 deficiency: MAE and EOAE. However, this does not exclude the involvement of the GLUT3 and MCT transporters in the pathogenesis of other types of epilepsy, such as temporal lobe epilepsy, where reduction in brain MCT1 has been reported.[13, 14]

Thus the molecular explanation for syndromes that may respond to the ketogenic diet where GLUT1 variants are not found remains elusive. Analysis with hypothesis-free massively parallel sequencing may elucidate the molecular determinants in these patients. Our results are also consistent with the negative findings of an independent screen of GLUT3 in patients with GLUT1 deficiency syndrome-like disease published while we were writing this report.[15]

Acknowledgments

  1. Top of page
  2. Summary
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References
  9. Biography
  10. Supporting Information

We thank the patients and their families for participating in our research program. Tina Damiano, graphic artist, is acknowledged for designing Figure 1. Elena Aleksoska (University of Melbourne) is acknowledged for performing genomic DNA extractions. Slave Petrovski (Duke University) is thanked for advice on the tolerance of the GLUT and MCT genes to variation. This study was supported by National Health and Medical Research Council Program Grant (628952) to S.F.B and I.E.S, an Australia Fellowship (466671) to S.F.B, a Practitioner Fellowship (1006110) to I.E.S and a Postdoctoral Training Fellowship (546493) to M.S.H.

Disclosure

  1. Top of page
  2. Summary
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References
  9. Biography
  10. Supporting Information

None of the authors has any conflict of interest to disclose. We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.

References

  1. Top of page
  2. Summary
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References
  9. Biography
  10. Supporting Information

Biography

  1. Top of page
  2. Summary
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References
  9. Biography
  10. Supporting Information
  • Image of creator

    Dr.Michael S.Hildebrand is a molecular geneticist with more than ten years experience studying the genetics of neurological disorders

Supporting Information

  1. Top of page
  2. Summary
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References
  9. Biography
  10. Supporting Information
FilenameFormatSizeDescription
epi12519-sup-0001-SupportingInformation.pptapplication/mspowerpoint288K 

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.