Fig. 1

Insertion of the 18-bp sequence into the CDK5RAP2-PDGFRα fusion gene. a. Schematic diagram showing the structures of full-length CDK5RAP2, full-length-PDGFRα, and CDK5RAP2-PDGFRα fusion protein with its junction structure between CDK5RAP2 and PDGFRα. The N-terminal (aa 1 to 494) of CDK5RAP2 (exons 1–13), containing the coiled-coil domains (dark gray boxes), fuses with the C-terminal (aa 581 to 1089) of PDGFRα (exons 12–22), containing the tyrosine kinase domain, resulting in the 1016 aa CDK5RAP2-PDGFRα fusion protein. The 40-bp junction is composed of a 22-bp PDGFRα inverted intron 9 (aa 494–501, light gray box) and an 18-bp from an unknown source (aa 501–508). b. NCBI BLAST Search revealed that the18-bp is 100% identical to a sequence in Marinobacter sp. Hb8 complete genome (CP017715.1). Of the 18-bp, 17-bp (2–18) is 100% identical to a sequence in Rhodococcus fascians D188 complete genome (CP015235.1), Rhodococcus sp. PBTS2 complete genome (CP015220.1), Micrococcus luteus strain trpE16 genome (CP007437.1) and Micrococcus luteus NCTC 2665 complete genome (CP001628.1). The “a” (gray) is likely added during the fusion event, which together with “aa” in the inverted intron 9 sequence encodes K that does not exist in PDGFRα. c. Sequence alignment of intron 10 from PDGFRα [GenBank Accesssion # NP_006206.4] with AluSz6 retrotransposon [Accession # DF0000052]. AluSz6 was found using Dfam 2.0 database. Sequences were aligned by CLUSTAL OMEGA 1.2.2 Multiple Sequence Alignment software. Nucleotide numbers are indicated on either side. Asterisks denote identical sequences between hPDGFRα and AluSz6. Broken lines represent missing nucleotide sequences in AluSz6