Exon structure of the human CGM6 gene, DNA sequence of the putative promoter and determination of the transcriptional start sites. (A) Restriction endonuclease map of the genomic insert from cosmid clone F19632, encompassing the CGM6gene, flanked by a pair of EcoRI/SfiI restriction endonuclease sites in the vector (S =SstI, E = EcoRI, B = BamHI, N =NarI). Boxes indicate the exons (L = leader, N = N-terminal domain, A and B = immunoglobulin constant-like domains, M = membrane domain). Open boxes correspond to the 5′- and 3′-untranslated regions. The bar above the first exon and flanking sequence designates the Sst I restriction endonuclease DNA fragment, whose sequence is shown in (B). The nucleotide sequence is numbered from the upstream SstI restriction endonuclease site (numbers at the left). The derived amino acid sequence from the leader exon is shown below the DNA sequence and the start of the intron 1 donor site (GT) is inversely shaded. Multiple transcriptional start sites are indicated as open triangles above the corresponding nucleotides. Further upstream, SP-1 and C-EBP concensus sequences are overlined and labeled. A simple repetitive DNA sequence is underlined. (C) Primer extension analysis to determine the transcriptional start site of the CGM6 gene (lane 1). Extension products only seen using CML leukocyte RNA are indicated by bars at the left and their nucleotide positions relative to the translational start site (+1) are included. Open triangles point out unspecific extension products that were also seen using RNA from a colon adenocarcinoma (LoVo) and a cervix carcinoma (HeLa) cell line as templates (data not shown). Lanes 2 to 5 show the DNA sequencing reaction with the 1.7-kbSstI restriction endonuclease DNA fragment as a template and using the same oligonucleotide primer as for the primer extension analyses. This allows direct identification of the transcriptional start sites.