Identification of CaMKII splice variants expressed in the human hippocampus

Purpose: The goal of this study was to use deep sequencing to identify all splice variants of Calcium/calmodulin-dependent kinase II (CaMKII) expressed in the human hippocampus. Methods: Transcripts of CaMKII-encoding genes (CAMK2A, CAMK2B, CAMK2G, and CAMK2D) were sub-amplified by PCR from total RNA extracted from human hippocampal tissue samples from 3 donors. Illumina sequencing libraries were constructed by PCR from these initial pools of amplicons and sequenced on an Illumina MiSeq instrument. Sequencing reads passing quality controls were clustered on the basis of sequence identity or near-identity. Consensus sequences of clusters were mapped with known exons of CaMKII genes to identify the splice variant represented by each cluster. Donor 1 replicate 2, Donor 2, and Donor 3 libraries from genes CAMK2B, CAMK2G, and CAMK2D were first sequenced on a MiSeq Nano flow cell, then re-pooled for read balancing and sequenced again of a full-size MiSeq flow cell. For each library, reads from Nano and full-size flow cells were combined for subsequent analysis. Results: We perfomed the first comprehensive survey of CaMKII transcripts expressed in individual tissue samples (human hippocampus). We detected a total of 79 splice variants of the four human CaMKIIs: CaMKIIa (3), CaMKIIß (30), CaMKII? (24), and CaMKIId (22), across tissue samples from 3 donors. This represents the vast majority of possible in-frame CaMKII splice variants (Sloutsky and Stratton, European Journal of Neuroscience, 2020; https://doi.org/10.1111/ejn.14761). Overall design: Deep sequencing of sub-amplicons of transcripts from four CaMKII-encoding genes (CAMK2A, CAMK2B, CAMK2G, and CAMK2D)