Based on previous findings of significant differences in proteomes and transcriptomes of patients with vs. without epilepsy, a multinational group of investigators utilized proteomic and transcriptomic techniques to attempt to elucidate the molecular signaling networks associated with sudden unexpected death in epilepsy (SUDEP).
The first experiment utilized proteomics, the large-scale study of all proteins within cells, to analyze gene products in autopsy materials of patients with SUDEP and patients with non-SUDEP epilepsy through the North American SUDEP Registry. They found no proteomic differences between patients with SUDEP (n = 12) and non-SUDEP epilepsy patients (n = 14) in hippocampal CA1-3, dentate gyrus, or frontal cortex, areas thought to be involved in ictogeneis.
Secondly, they analyzed the complete set of RNA molecules expressed from hippocampal and temporal neocortical surgical specimens of patients with mesial temporal lobe epilepsy. Patients were divided into low- (n=6) or high SUDEP risk (n=8) groups based on the duration of postictal generalized EEG suppression. Transcriptomics identified 55 genes (including 37 protein-coding and 15 long noncoding RNAs) differentially expressed between the low- and high SUDEP risk groups’ hippocampi. Among the protein coding genes, the greatest decrease in expression in the high risk group was seen in GFRA1, a receptor for glial cell-derived neurotrophic factor GDNF. As GDNF has been associated with seizure suppression in animal models, decreased expression of GFRA1 may reflect a decrease in GDNF-associated seizure suppression in SUDEP patients. The greatest increase in gene expression in the high risk group was of sarcoglycan gamma (SGCG), a sarcolemmal transmembrane glycoprotein thought to link the cytoskeleton and the extracellular matrix. The increased expression of SGCG could result in aberrant cerebrovascular organization. The significance of the differential expression of long noncoding RNAs is not yet clear.
In sum, compared to other patients with epilepsy, no proteomic difference was detected in the SUDEP autopsy tissue and limited differences were seen in the hippocampal transcriptomes of patients hypothesized to be at high risk of SUDEP based on duration of postictal generalized EEG suppression. The authors emphasized the need to study epilepsy subtypes separately and to explore the role of other brain regions in SUDEP
COMMENTARY: Significant differences in proteomes and transcriptomes of the hippocampus and cortex have been demonstrated between patients with epilepsy and controls.  While the brain regions mediating SUDEP are not clearly elaborated, some data suggests that areas involved in epileptogenesis may be implicated in SUDEP as well. The absence of significant differences in proteomes and transcriptomes between SUDEP/high-risk and non-SUDEP/low-risk epilepsy patients in this study suggests that ictogenic brain regions may not be driving SUDEP pathophysiology. Future studies of other brain areas, such as the brainstep, are needed . Tissue banks would also be of help.
Furthermore, transcriptomics analyses should be reevaluated when we know more about the roles of long noncoding RNAs. There are still unknowns such as epilepsy type, genetics, and heart involvement in SUDEP. More autonomic parameters and underlying causes must be sought. Different mechanisms in SUDEP compared to epilepsy pathophysiology could be elucidated in future studies. In this way, as pediatric neurologists we will be more able to counsel for future risks and protect our patients from SUDEP.
The author(s) have declared that no competing interests exist