Evaluation of human-specific RNA enrichment kits for transcriptome sequencing in a non-model organism: the African savanna elephant (Loxodonta africana)

Background The use of whole blood to study immune responses in non-model organisms, like African savanna elephants ( Loxondota africana ), has been limited by the lack of species-specific resources. In whole blood, overrepresented globin mRNA and ribosomal RNA can dominate transcriptomic analyses, masking important biologically informative transcripts and therefore need to be removed for downstream applications such as next generation sequencing. Although kits to deplete globin mRNA and rRNA are commercially available for humans, rats and mice, similar kits are not available for non-model species. Therefore, this study aimed to evaluate the efficiency of cross-species globin mRNA (GLOBINclear™-Human Kit) and ribosomal RNA (NEBNext® rRNA Depletion Kit v2) depletion strategies for African savanna elephant whole blood-derived RNA, with the goal of establishing an optimized preprocessing workflow for downstream RNA sequencing. Results Following bioinformatic processing, an average of 127.9–278 million paired-end reads per sample were retained with Phred quality scores ranging from Q34 to Q38 across all enriched and unprocessed total samples. Depletion reduced the proportion of rRNA reads from 57.5–73.1% in unprocessed samples to < 3.5% in the enriched samples. However, no significant reduction was observed in the proportion of globin mRNA reads following enrichment. Despite this, genome alignment metrics improved substantially following RNA enrichment, with alignment efficiency increasing from 25.3–40.5% in the unprocessed samples, to 86.2–90.5% in enriched samples. BLAST results revealed moderate sequence similarity between human and African savanna elephant haemoglobin genes (identity: 73.8–79.6%) and high sequence similarity for the rRNA gene targets (identity: 76 to 100% %). Conclusions Human-specific RNA enrichment kits effectively depleted African savanna elephant ribosomal RNA prior to sequencing, resulting in a substantial increase in the proportion of uniquely mapped reads to the elephant reference genome. This pre-processing approach enhances the quality of RNA-seq data and supports downstream transcriptomic analyses in African savanna elephant, which will enable investigation of immune-related pathways in this endangered species. The application of this optimized workflow has potential to improve genomic reference resources and enhance the utility of transcriptomics for advancing the understanding of immune responses in African savanna elephants.