Characterization of an abundant Schistosoma mansoni transcript with no homologs in the databases.

Abstract

The expressed sequence tag (EST) approachthat we have used in the Schistosoma mansoni Ge-nome Project is a powerful technique for the dis-covery of new genes of the parasite (GR Franco etal. 1995 Gene 152: 141-147, E Dias Neto et al.1997 Gene 186: 135-142). In a recent compara-tive study of gene expression in distinct develop-mental stages of the parasite life cycle using theEST strategy, we identified 466 different genes.From this total, 427 were novel and 333 of themcould not be identified based on homologies withdatabase sequences (GR Franco et al. 1997 DNAResearch 4 : 231-240). The high frequency of someof these “unknown” genes in different cDNA li-braries suggests that they might have importantroles in the biology of S. mansoni and thus mayconstitute possible targets for drug design or vac-cine production. One of these genes, highly abun-dant in one of four adult worm libraries that weare studying in our laboratory, was selected forfurther characterization.After clustering analysis of ESTs from differ-ent S. mansoni cDNA libraries using the programICATOOLS (Franco et al. 1997 loc. cit. ), we iden-tified a cluster composed of 16 ESTs from bothcDNA ends that corresponded to an unknown genehighly frequent in an adult worm cDNA library.We have called this gene AUT1 for abundant un-known transcript 1. A single strand consensus ofapproximately 1.6 kb long was derived from thealignment of the 16 EST sequences using the pro-gram DNAsis. In order to obtain the cDNA full-length sequence from both strands, the cDNA clonecontaining the largest insert was digested with therestriction enzymes HindIII and SphI (Fig. 1), thefragments produced were further cloned into thepUC18 vector (Pharmacia) and completely se-quenced on both strands using the Thermo-Sequenase Fluorescent Labeled Primer Cycle Se-quencing kit (Amersham Life Science) and theA.L.F. Automated DNA Sequencer (Pharmacia).Another strategy used to obtain the cDNA sequencefrom both strands was the amplification by poly-merase chain reaction (PCR) of different regionsof the cDNA using specific primers designed forthe gene. The fragments were cloned into thepUC18 vector (SmaI cloning site) using theSureClone Ligation kit (Pharmacia) and sequencedas before. The sequences generated from both pro-cesses were aligned using the DNAsis program andthe cDNA full-length sequence from both direc-tions was obtained, totaling 1520 bp (Fig. 2). ThecDNA was translated into the six possible framesand the length of the longest open reading frame(ORF) was 1005 bp long, potentially encoding aprotein of 335 amino acids (Fig. 2).Analysis of the primary structure of the puta-tive protein coded by AUT1 gene shows one po-tential site for N-linked glycosylation, as well aseight sites for phosphorylation by protein kinaseC, five for phosphorylation by casein kinase II andeight sites for cAMP-dependent protein kinase,suggesting this protein might be phosphorylatedin the organism. The protein does not contain anysignal sequence responsible for translocation acrossthe endoplasmic reticulum (ER) membrane, as seenin secretory and plasma membrane spanning pro-teins, or stretches of hydrophobic residues forplasma membrane insertion. Several searches wereperformed on distinct databases of protein se-quences, typical protein domains and families ofproteins as an attempt to identify the predicted pro-tein or a specific domain on it. All the searchesprovided neither homology with any sequences indatabases nor identifiable structural domains.Primers designed for PCR amplification of afragment containing the complete coding region