Article Text

Splicing of the glutamate transporter EAAT2: a candidate gene of amyotrophic lateral sclerosis
  1. Department of Neurology, University of Ulm, Germany
  1. Dr Thomas Meyer, Department of Neurology, University of Ulm, Albert-Einstein-Allee 11/O25–5, 89081 Ulm, Germany. Telephone 0049 731 177 1200; fax 0049 731 177 1202.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Defective glutamatergic neurotransmission may have a critical role in the pathogenesis of amyotrophic lateral sclerosis (ALS). A reduced synaptic glutamate reuptake has been found in the motor cortex and spinal cord of patients with sporadic ALS.1 More recently, a selective loss of the glial glutamate transporter protein EAAT2 has been described.2 Most recently aberrant splicing of the EAAT2 transcript was reported to be the cause for a reduced expression of the EAAT2 protein.3 Several novel EAAT2 transcripts were cloned from patients with ALS and described as disease specific. In vitro expression studies suggested that proteins translated from these transcripts were rapidly degraded and show a dominant negative effect on normal EAAT2 protein which appears to be the predominat glutamate transporter in the CNS.3 A loss of EAAT2 can lead to neuronal degeneration through abnormal accumulation of the potential neurotoxin glutamate and excitotoxic mechanisms. This pathogenetic concept was supported by the clinical efficacy of antiglutamatergic drugs in patients with ALS and transgenic models. One of the reported transcripts was characterised by the skipping of the protein coding exon 8 of the EAAT2 gene. This transcript was amplified by polymerase chain reaction from ALS-CSF and suggested as a diagnostic tool in ALS.3 Interestingly, this transcript is identical to an alternative splicing product of the EAAT2 transcript which we have recently reported and named EAAT2/C1. This and another splice product, named EAAT2/C2, have been cloned from normal human brain RNA.4

Here we report the cloning of two further EAAT2 transcripts, named EAAT2/C3 and EAAT2/C4. Based on the EAAT2 sequence information we designed specific primers for reverse transcription (RT) of the EAAT2-mRNA using control human brain poly-A+ RNA as template (Clontech, Palo Alto). RT and PCR amplification were performed as described before (RT primer: 5′ CAGTTACCATAG-GATACGCTGG; PCR primers: 5′ GATAGTTGCTGAAGAGGAGGGG; 5′ CATATC-CTTATTT CT- CACGTTTCC).4PCR cloning and DNA sequencing disclosed two novel EAAT2 transcripts which resulted from splicing of protein coding sequences. EAAT2/C3 orginated from a deletion of 234 nucleotides (891–1124; GenBank UO3505) corresponding to exon 6 of the EAAT2 gene which is coding for 78 amino acids in the central part of the putative EAAT2 polypeptide (figure).5 The EAAT2/C4 transcript was characterised by the deletion of 702 nucleotides ranging from position 992 to 1693 (GenBank U03505). The splicing occurred at internal 5′- and 3′-splice sites which showed an uncomplete consensus sequence. EAAT2/C4 resulted from deletion of exons seven to nine and parts of exons six and 10 (figure) with the downstream sequence still in frame.5 At the putative protein level EAAT2/C4 showed a loss of 234 amino acids located in the middle and C-terminal part of the polypeptide.

Schematic presentation of alternative splicing in EAAT2 and its relation to the protein coding exons of the human EAAT2 gene.5 Alternatively spliced exon sequences are shown in grey. Exon and intron lengths are not to scale. CDS=coding sequence.

Our findings contribute to the notion that the EAAT2 transcript is highly variable. Splicing of the EAAT2 transcript is also found under normal conditions and may be part of post-transcriptional EAAT2 gene regulation. Furthermore, alternative EAAT2 transcripts were identified in other species. We conclude that splicing of the EAAT2 transcript is unlikely to be ALS specific. The EAAT2 gene regulation and its pathogenetic relevance are far from completely understood. The use of EAAT2 splicing products as diagnostic tools in ALS would be extremely valuable, but further evidence is necessary before concluding that these splice variants are specifically associated with ALS. However, the evolving knowledge on EAAT2 gene regulation will provide the basis for a comprehensive association study of EAAT2 splicing products in ALS and other neurodegenerative diseases.