Transcriptome mining of squalene cyclases in Lygodium japonicum

Abstract Transcriptome mining is a powerful tool for gene discovery in plant sciences. Biosynthetic enzyme transcripts of specialized metabolites have been isolated using transcriptome datasets of non-model plants; however, this methodology has not been applied to fern metabolites. Herein, we attempted to isolate transcripts of squalene cyclases from the Lygodium japonicum transcriptome database. A BLAST search revealed the presence of four isotigs (12872, 12873, 33018, and 33763) annotated as squalene cyclase, two of which contained full-length open-reading frame sequences. However, both isotigs failed to be amplified; our investigation revealed them to be chimeric sequences due to incorrect assembly. We also revealed that part of isotigs 12873 and 33018 constituted a single gene, designated LJH. Conversely, homology-based PCR resulted in the isolation of a squalene cyclase homolog, LjSCx, distinct from the four isotigs. Biochemical characterization of LJH and LjSCx revealed that LJH is a hopene synthase gene and LjSCx is a pseudogene. These findings highlight the usefulness and limitations of transcriptome datasets, especially with respect to genes comprising a gene family.