Nucleotide sequence of cloned cDNA for human sphingolipid activator protein 1 precursor

Two cDNA clones encoding prepro-sphingolipid activator protein 1 (SAP-1) were isolated from a lambda gt11 human hepatoma expression library using polyclonal antibodies. These had inserts of approximately 2 kilobases (lambda-S-1.2 and lambda-S-1.3) and both were both homologous with a previously isolated clone (lambda-S-1.1) for mature SAP-1. We report here the nucleotide sequence of the longer two EcoRI fragments of S-1.2 and S-1.3 that were not the same and the derived amino acid sequences of mature SAP-1 and its prepro form. The open reading frame encodes 19 amino acids, which are colinear with the amino-terminal sequence of mature SAP-1, and extends far beyond the predicted carboxyl terminus of mature SAP-1, indicating extensive carboxyl-terminal processing. The nucleotide sequence of cDNA encoding prepro-SAP-1 includes 1449 bases from the assigned initiation codon ATG at base-pair 472 to the stop codon TGA at base-pair 1921. The first 23 amino acids coded after the initiation ATG are characteristic of a signal peptide. The calculated molecular mass for a polypeptide encoded by 1449 bases is approximately 53 kDa, in keeping with the reported value for pro-SAP-1. The data indicate that after removal of the signal peptide (23 amino acids) mature SAP-1 (78 amino acids) is generated by removing an additional 7 amino acids from the amino terminus and approximately 373 amino acids from the carboxyl terminus. One potential glycosylation site was previously found in mature SAP-1. Three additional potential glycosylation sites are present in the processed carboxyl-terminal polypeptide, which we designate as P-2. The molecular mass of glycosylated pro-SAP-1 is estimated at approximately 69 kDa, assuming glycosylation of all four sites. The value is close to the reported 70-kDa value for glycosylated pro-SAP-1. A computer search failed to reveal homology between P-2 and the sequence of any other protein; its function is uncertain. The 3' untranslated region is composed of 90 base pairs and is incomplete, since it does not contain a polyadenylylation site or a poly(A) tail.