Nishiyama et al. Vol. 114, No. 5, November 1991. Pages 359–371.
The cDNA sequence coding for the core protein of the rat NG2 proteoglycan published in The Journal of Cell Biology article by Nishiyama et al. (1991) contains three errors between nucleotides 6208 and 6357. The correct nucleotide sequence and the deduced amino acid sequence of this segment are shown in Fig. 1. The correct nucleotide sequence contains the following three insertions: T at position 6209; C at position 6325; and G at position 6356. The first insertion alters the reading frame, which is shifted again with the second insertion and then returns to the original reading frame after the third insertion, changing the published amino acid sequence from residues 2047–2095.
The altered amino acid sequence corresponds to a region in the extracellular domain of NG2 NH2-terminal to the transmembrane region. This segment encompasses a region in the published sequence that was reported to contain six cysteine residues. These cysteine residues do not exist in the correct sequence. Thus, the correct primary structure does not have sequence similarity to the insulin receptor-related receptor (Shier and Watt, 1989) as originally reported. The correct sequence also does not contain the arginine-arginine sequence originally reported to be present at positions 2096/2097, which was speculated to be a possible site of cleavage by trypsin (Nishiyama et al., 1995). The correction of the rat NG2 core protein sequence does not alter any of the results that we have published since the first report of the sequence. The correction has been submitted to GenBank (accession number X56541).