ATG Start Site Research Articles

Structural organization of the human S-antigen gene. cDNA, amino acid, intron, exon, promoter, in vitro transcription, retina, and pineal gland.

S-Antigen (S-Ag) is a major soluble photoreceptor protein involved in the visual transduction cascade. Several S-Ag cDNAs and a gene coding for human S-Ag were isolated from cDNA and gene libraries. The gene sequences of the coding, noncoding, and 5'-flanking regions of the gene were determined. The S-Ag gene was approximately 50 kbp (kilobase pairs) in length and contained 16 exons and 15 introns. The length of most exons was less than 100 base pairs (bp) and the smallest one was only 10 bp. In contrast, the length of most introns was larger than 2 kbp, and the gene comprised 97% intron and 3% exon. The splice sites for donor and acceptor were in good agreement with the GT/AG rule. The S-Ag protein of 403 amino acid residues was translated from a mRNA of 1.9 kbp, and the mRNA was transcribed from a gene of 50 kbp. The 5'-flanking region of the gene, approximately 1.1 kbp long, had no known regulatory elements for transcription such as TATA, GC, and CCAAT boxes. Interestingly, the 5'-flanking region had promoter activity in an in vitro transcription assay using a nuclear extract of rat brain. A major transcription start site was found at 387 bp upstream from the translation start site ATG. Our results indicate that the sequence of S-Ag promoter differs from other known promoters and may, perhaps, be specific for photoreceptor rod cells and pinealocytes.

Genomic organization of human complement component C3

Six overlapping clones, spanning the entire C3 gene and the 5′ flanking region, were isolated from two genomic λ libraries. Thirty exons, covering the complete β chain and part of the α chain as far as the C3d region, were analyzed. The full exon-intron organization of the gene was deduced by combining our data with the reported organization of the α′ chain (Barnum et al., 1989, J. Biol. Chem. 264: 8471–8474). The complete gene is 41 kb and consists of 41 exons. The C3 β chain spans 13 kb from exon 1 to exon 16. Exon 16 encodes both α and β chains. The α chain is 28 kb and contains 26 exons, including exon 16. The 5′ flanking region was sequenced up to 514 bases upstream from the ATG start site. The major transcription initiation site was mapped to an adenine residue 61 bases upstream from the signal peptide by primer extension analysis of poly(A) + mRNA from hepatocytes and U937 cells. The TATA box motif was assigned at position-85. Several putative binding sites for transcription factors were found in the 5′ flanking region, suggesting possible pathways for the regulation of the C3 gene.

Species-Specific High Molecular Weight Forms of Basic Fibroblast Growth Factor

bFGF was extracted from either mouse, rat and human cell lines or mouse, rat bovine and human brain tissue and partially purified by cation exchange chromatography and heparin-affinity chromatography. When the heparin-affinity purified proteins were probed on Western blots with antisera against either a highly conserved internal bFGF sequence or recombinant 18 kDa bFGF, species-specific forms of bFGF were detected. bFGF proteins from rat and mouse sources were of apparent molecular weight 18,000, 21,500 and 22,000 whereas those from human sources were of 18,000, 22,500 and 24,000. Bovine bFGF proteins were similar to the multiple human bFGFs. The 22.5 kDa and 24 kDa proteins from human cells were also recognized by an antibody specific for the N-terminally extended forms of human bFGF, whereas this antibody failed to detect 18 kDa bFGF. We showed that the differences in molecular weight between human and rat bFGFs are consistent with the predicted ATG (methionine) or alternative CTG (leucine) translational start sites in the 5' upstream sequences of bFGF cDNAs. In addition we show that, irrespective of the species of origin, the larger bFGF proteins may be separated from 18 kDa bFGF by Mono S chromatography.

Use of antisense RNA to help identify a genomic clone for the 5′ region of mouse β-glucuronidase

It was possible to gauge the inhibition of mouse β-glucuronidase expression by injecting RNA, made from both strands of subclones of a cosmid containing the complete gene, into mouse blastomeres at the four-cell stage. Although our initial screen did not identify the 5′ region, we were able to isolate a subclone containing homology to 20 bp coding for N-terminal amino acids of rat and human β-glucuronidase structural genes. Antisense RNA prepared from one strand of the 350 bp Pst I subclone inhibited β-glucuronidase expression by 89% while RNA prepared from the other strand had little effect. The subclone appears to correspond to the 350 bp fragment identified by others as one including the ATG start site of mouse β-glucuronidase.

Comparison of the 5′ regions of human and mouse carbonic anhydrase II genes and identification of possible regulatory elements

The nucleotide sequence of the 5′ region of the human carbonic anhydrase II gene has been determined. This sequence begins 643 base pairs upstream from the ATG start site and continues through exon 1, intron 1, exon 2 and the adjoining 125 nucleotides of intron 2. The human sequence is compared with homologous regions of the mouse (YBR strain) carbonic anhydrase II gene by aligning the two sequences for optimal homology. In addition to a TATA box and a putative CCAAT box (CCACC in human and CCACT in mouse), three conserved tandem-repeat elements in mouse and two in human (consensus: cCNGTCACCTCCgC) are located 15 and 22 base pairs upstream, respectively, from the CCAAT boxes in the human and mouse sequences. This repeat element is similar to a tandem repeat sequence located at about the same position in mammalian β-globin genes, and may represent regulatory elements common to both the carbonic anhydrase and β-globin genes. The regions surrounding exon 1 are extremely G + C-rich in both human and mouse genes. In addition, severel CCGCCC or GGGCGG sequences which may be important for transcriptional efficiency are found in the 5′ flanking regions of the human and mouse genes.