Early onset absence epilepsy: 1 in 10 cases is caused by GLUT1 deficiency
Address correspondence to Ingrid E. Scheffer, Melbourne Brain Centre, 245 Burgundy St, Heidelberg, Vic. 3084, Australia. E-mail: firstname.lastname@example.org
Glucose transporter 1 (GLUT1) deficiency caused by mutations of SLC2A1 is an increasingly recognized cause of genetic generalized epilepsy. We previously reported that >10% (4 of 34) of a cohort with early onset absence epilepsy (EOAE) had GLUT1 deficiency. This study uses a new cohort of 55 patients with EOAE to confirm that finding. Patients with typical absence seizures beginning before 4 years of age were screened for solute carrier family 2 (facilitated glucose transporter), member 1 (SLC2A1) mutations or deletions. All had generalized spike-waves on electroencephalography (EEG). Those with tonic and/or atonic seizures were excluded. Mutations were found in 7 (13%) of 55 cases, including five missense mutations, an in-frame deletion leading to loss of a single amino acid, and a deletion spanning two exons. Over both studies, 11 (12%) of 89 probands with EOAE have GLUT1 deficiency. Given the major treatment and genetic counseling implications, this study confirms that SLC2A1 mutational analysis should be strongly considered in EOAE.
Glucose transporter 1 (GLUT1), encoded by solute carrier family 2 (facilitated glucose transporter), member 1 (SLC2A1), is the facilitative glucose transporter across the blood–brain barrier. Mutations in SLC2A1 are found in the GLUT1-deficiency encephalopathy (GLUT1DS, OMIM 606777) (De Vivo et al., 1991; Seidner et al., 1998). GLUT1-deficiency encephalopathy is a severe autosomal-dominant metabolic encephalopathy with refractory infantile-onset seizures, complex motor disorder, cognitive impairment, and microcephaly. Hypoglycorrhachia (low fasting cerebrospinal fluid [CSF] glucose <2.2 mm, CSF/plasma glucose ratio <0.45) has been considered the diagnostic hallmark of GLUT1 deficiency (Klepper & Leiendecker, 2007).
The phenotypic spectrum of SLC2A1 mutations has considerably expanded. It now includes a range of paroxysmal neurologic disorders including epilepsies and movement disorders. The movement disorder paroxysmal exertional dyskinesia (PED) suggests GLUT1 deficiency (Suls et al., 2008; Weber et al., 2008). Other presentations include hemolytic anemia (Weber et al., 2008), alternating hemiplegia of childhood (Rotstein et al., 2009), paroxysmal choreoathetosis with spasticity (Weber et al., 2011), and absence epilepsy beginning from early childhood to young adult life (Suls et al., 2009; Mullen et al., 2010). Autosomal recessive inheritance has also been reported, but is uncommon (Klepper et al., 2009; Rotstein et al., 2010).
Early onset absence epilepsy (EOAE) is an uncommon genetic generalized epilepsy syndrome with onset of typical absence seizures before 4 years of age. We recently showed that 4 (10%) of 34 cases with EOAE had GLUT1 deficiency caused by mutations in SLC2A1 (Suls et al., 2009). Herein, we present a replication study of a new cohort of 55 patients with EOAE.
Materials and Methods
Cases had absence seizures as the predominant seizure type with onset before 4 years of age. All had generalized spike-waves (>2.5 Hz) and absence seizures on electroencephalography (EEG). Exclusion criteria included atonic or tonic seizures. Written informed consent was obtained from all patients and in the case of minors, their parents or legal guardians. The study was approved by the Austin Health Human Research Ethics Committee.
Fifty-five patients with EOAE and 500 blood bank DNA samples were included. The whole coding (10 exons and exon/intron junctions) and promoter regions of SLC2A1 were polymerase chain reaction (PCR) amplified and PCR products sequenced using BigDye chemistry on ABI3730XL DNA analyzer (Applied Biosystems, Carlsbad, CA, U.S.A.). Mutation detection used sequence analysis software (CodonCode, Centerville, MA, U.S.A.). Multiplex ligation-dependent probe amplification (MLPA), using probemix P138-B1 (MRC Holland, Amsterdam, The Netherlands) was performed to detect deletions or duplications.
There were seven mutations, including five missense mutations, a small in-frame deletion, and an exonic deletion (see Table 1 and Figs. S1 and S2).
Table 1. Patients with EOAE and GLUT1 deficiency
|Mutation details|| || || || || || || |
| Mutation||c.476T>G||c.643C>T||c1134-1136del CTT||c.457C>T||c.627G>C||c.277C>T||Exon 3 and 4 deletion|
| Amino acid||Leu159Arg||Leu215Phe||Phe379del||Arg153Cys||Glu209Asp||Arg93Trp||–|
| Exon||4||5||9||4||5||4||3 and 4|
| Location||5th TM||3rd IC||10th TM||2nd IC||3rd IC||1st IC||–|
| Inheritance||De novo||Maternal||Unknown||De novo||Maternal||Unknown||De novo|
|Clinical details|| || || || || || || |
| Current age (year)||14||12||19||4||6||6||27|
| Absence (onset age)||2 years||2 years||3 years||2 years||3 years||18 months||<1 year|
| GTCS (onset age)||4 years||–||5 years||–||–||–||–|
| MJ (onset age)||–||–||–||–||–||–||–|
| Examination||Normal||Normal||Normal||Normal||Normal||Mild action tremor||Ataxia, spasticity|
| PED||No||No||Mild – 2 years||No||No||No||No|
| Intellect||Mild ID||Normal||Normal||Moderate developmental delay||Normal||Normal||Severe ID|
| Seizure control||Weekly absence on VPA||Seizure-free on VPA, LTG, ETX||Frequent absence on LTG||Seizure-free on ketogenic diet||Seizure-free on VPA and LEV||Seizure-free on ketogenic diet||Ongoing absence seizures on multiple medication|
| Fasting CSF glucose (mm) |
Normal ≥2.2 mm
| Fasting serum glucose (mm) |
Normal = 3.5–6 mm
| CSF/serum glucose ratio |
Three of the five missense changes (Patients 1, 2, and 5) were neither previously reported nor found in 1,000 blood bank–control chromosomes. The mutation in patient 4 (c.457C>T) has been reported in classical GLUT1 encephalopathy (Klepper & Leiendecker, 2007). The mutation in patient 6 (c.227C>T) has been reported to cause mental retardation, ataxia, and relatively late-onset (11 years of age) difficult-to-characterize seizures (Joshi et al., 2008). All of the novel SLC2A1 missense mutations and the in-frame deletion affect evolutionarily conserved amino acids (Table 2).
Table 2. Evolutionary conservation of the four novel SLC2A1 mutations
|Human|| ||T|| L ||H|| ||L|| L ||I|| ||G|| F ||V|| ||P|| E ||S|
|Rat|| ||T|| L ||H|| ||L|| L ||I|| ||G|| F ||V|| ||P|| E ||S|
|Orangutan|| ||T|| L ||H|| ||L|| L ||I|| ||G|| F ||V|| ||P|| E ||S|
|Mouse|| ||T|| L ||H|| ||L|| L ||I|| ||G|| F ||V|| ||P|| E ||S|
|Horse|| ||T|| L ||H|| ||L|| L ||I|| ||G|| F ||V|| ||P|| E ||S|
|Cow|| ||T|| L ||H|| ||L|| L ||I|| ||G|| F ||V|| ||P|| E ||S|
|Kangaroo|| ||T|| L ||H|| ||L|| L ||I|| ||G|| F ||V|| ||P|| E ||S|
|Chimpanzee|| ||T|| L ||H|| ||L|| L ||I|| ||G|| F ||V|| ||P|| E ||S|
|Macaque|| ||T|| L ||H|| ||L|| L ||I|| ||G|| F ||V|| ||P|| E ||S|
|Dog|| ||T|| L ||H|| ||L|| L ||I|| ||G|| F ||V|| ||P|| E ||S|
|Opossum|| ||T|| L ||H|| ||L|| L ||I|| ||G|| F ||V|| ||P|| E ||S|
|Zebrafish|| ||T|| L ||H|| ||L|| L ||I|| ||G|| F ||V|| ||P|| E ||S|
|Platypus|| ||T|| F ||H|| ||L|| L ||I|| ||G|| F ||V|| ||P|| E ||S|
|Chicken|| ||T|| F ||H|| ||L|| L ||I|| ||–|| – ||–|| ||–|| – ||–|
|Stickleback|| ||T|| F ||H|| ||L|| L ||I|| ||G|| F ||V|| ||P|| K ||S|
Three mutations arose de novo (Patients 1, 4, and 7), two were inherited, and the inheritance for Patients 3 and 6 could not be determined. Both mothers with mutations were affected: the mother of Patient 2 had mild PED and of Patient 5 had childhood absence epilepsy. The six patients with missense mutations or in-frame deletion had onset of absence seizures at between 18 months and 3 years of age. Two (Patients 1 and 3) had occasional generalized tonic–clonic seizures (GTCS) with onset after 4 years; no patients had myoclonic seizures. Patient 3 had PED. Intellect was normal in four cases. Patient 1 had mild intellectual disability, whereas Patient 4 had moderate developmental delay. Patients 2 and 5 were fully controlled with antiepileptic drugs, whereas Patients 1 and 3 had ongoing absence seizures on medication. Two patients (4 and 6) had complete control of seizures with the classical ketogenic diet.
Patient 7, with the exonic deletion, is 27 years old, and although he had early onset absence seizures, his later course was closer to the classical encephalopathy with severe intellectual disability and ataxia.
Fasting lumbar puncture studies were performed in six of the seven cases. Patient 6 showed abnormal fasting CSF glucose (<2.2 mm). Fasting CSF glucose was borderline at 2.3–2.4 mm in the remaining four patients without exonic deletions. Further attempts to obtain early records in Patient 7 showed a CSF glucose of 2 mm in the first year of life.
We previously showed that 4 of 34 patients with EOAE had GLUT1 deficiency caused by mutations in SLC2A1 (Suls et al., 2009). Here we report our replication of these results in a new, larger cohort of 55 patients with EOAE in which 7 patients had SLC2A1 mutations. Taking both of our studies together, we show that approximately 12% of EOAE cases (11/89) are caused by GLUT1 deficiency.
The novel mutations are likely to be pathogenic for several reasons. They either affect highly conserved amino acids or are deletions of whole exons. The missense mutations change the amino acid polarity and charge in 2/3 (Patients 1 and 5). Of the three novel missense mutations, one arose de novo and two were inherited from affected parents.
Patients with missense mutations or an in-frame deletion could not be distinguished from the wider group with EOAE. Typical absence seizures were seen without distinctive clinical or EEG features. PED was present in a mild form in only one case. A family history of absence epilepsy was present in only one mother. A negative family history is likely in many patients, as de novo mutations are common. Exonic deletions are usually associated with GLUT1 encephalopathy. Patient 7, although included due to initial presentation, had a later course consistent with GLUT1 encephalopathy.
The longstanding diagnostic boundaries for GLUT1 encephalopathy have been a low CSF glucose (<2.2 mm) or CSF/serum glucose ratio of <0.45 (Klepper & Leiendecker, 2007). Only two of six patients tested had low CSF glucose (Patients 6 and 7); the remainder were low-normal. Our results support the view that a CSF glucose of <2.5–2.6 mm is consistent with GLUT1 deficiency (De Vivo & Wang, 2008; Leen et al., 2010). Raising the required CSF/serum ratio for GLUT1 deficiency to <0.50 has also been proposed; however, half our cohort had a ratio ≥0.50 (Leen et al., 2010).
CSF glucose >2.6 mm may be insufficient to exclude GLUT1 deficiency with mild phenotypes. The expanded GLUT1-deficiency spectrum is a recent finding, and the upper limit of CSF glucose compatible with the diagnosis is unclear. Familial cases have shown CSF glucose above 2.6 mm, albeit uncommonly (Suls et al., 2008). A low-normal CSF glucose does not exclude GLUT1 deficiency.
That restriction of glucose delivery to the brain may cause epilepsy when the CSF glucose is normal is not yet adequately explained. It may be that glucose delivery is inadequate for peak demand or that transport is not uniformly reduced; for instance the basal ganglia may be particularly affected in those with PED.
In EOAE and familial GLUT1 deficiency the severity of the epilepsy declines in adulthood (Mullen et al., 2010). In our cases diagnosed early, two have used the ketogenic diet (Patients 4 and 6). Cognitive impairment is evident in less than half (3/7) of cases. However, the normal range for intellect is wide, and other patients may be achieving below their potential without frank intellectual disability. Although a ketogenic diet may be unnecessary, the diagnosis significantly changes the treatment algorithm and it remains to be shown whether ketogenic diet may improve cognition even within the normal range.
We confirm that more than 1 in 10 cases of EOAE has GLUT1 deficiency, which cannot be diagnosed on clinical grounds. We suggest that SLC2A1 mutation screening represents a key etiologic investigation for EOAE when earlier consideration of the ketogenic diet may lead to seizure control and improved cognitive outcome.
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.