Abstracts

Rationale: Variants in SLC6A1, encoding GABA transporter 1 (GAT1), have been associated with a wide range of phenotypes, including epilepsies, autism, and intellectual disability. Recently, three variants have been identified in patients with schizophrenia and no epilepsy. We have shown that the previously studied variants associated with epilepsy have a reduction in ability to uptake GABA. Therefore, to contrast the schizophrenia cohort from epilepsy cohorts, we hypothesized these variants may have different pathomechanisms than those associated with epilepsy or autism, such as an outward leak of GABA through the transporter. We sought to elucidate how these schizophrenia-associated variants differ from those that cause epilepsy, and how they are similar, in order to evaluate the potential usefulness of anti-epileptic drugs for a subset of patients with schizophrenia.

Methods: To predict protein stability, machine learning tools including mCSM, SDM, DUET, DynaMut, INPS-MD, and MAESTRO were used. Radioactive ³H-labeled GABA was used to assess the GABA uptake ability of transfected cells expressing the wildtype or variant transporter. ³H-GABA was also used to measure rate of efflux out of cells pre-loaded with ³H-GABA. Total protein expression was assessed by immunoblotting and flow cytometry. Surface expression was assessed by biotinylation and flow cytometry. Glycosylation patterns were determined by digesting cell lysates with Endo H, which specifically removes endoplasmic reticulum linked glycans, and PNGase F, which removes all glycans, followed by immunoblotting.

Results: In silico analysis predicts all three variants to destabilize the GAT1 protein. GABA uptake function was slightly reduced in all 3 variants, compared to wildtype (A93T 76±14%, R211C 63±9%, W495L 40±7%, compared to WT 100%. n = 3). Total and surface protein expression are not significantly reduced, as measured by Western blots (Total: A93T 97±14%, R211C 80±6%, W495L 95±15%, compared to WT 100%. n = 6. Surface: A93T 55±14%, R211C 80±27%, W495L 87±19%, compared to WT 100%. n = 4), or flow cytometry (Total: A93T 89±6%, R211C 86±3%, W495L 92±6%, compared to WT 100%. n = 3. Surface: A93T 95±5%, R211C 85±15%, W495L 75±22%, compared to WT 100%. n = 3). Glycosylation is also not noticeably affected (n = 1; further replicants in progress). Rate of efflux is not altered (A93T 38±4, R211C 45±6, W495L 43±5, compared to WT 49±5. n = 4).

Conclusions: The range of phenotypes associated with the GAT1-encoding SLC6A1 has expanded to include schizophrenia. In contrast to those variants associated with epilepsy and autism, the variants of schizophrenia caused a relatively mild reduction of GABA uptake. This suggests some similarities in pathomechanism between epilepsy and schizophrenia variants in SLC6A1. This may open up new treatment options for schizophrenia, including both pharmacological or genetic means that restore GAT1 function. Next steps include investigation of the effect of both these schizophrenia-associated and epilepsy-associated GAT1 variants on the release of dopamine.

Funding: Please list any funding that was received in support of this abstract.: SLC6A1 Connect.