Evidence for a class of very small introns in the gene for hypoxanthine-guanine phosphoribosyltransferase in Schistosoma mansoni

Heterologous expression of hAID in E. coli leads to the production of a splice isoform of AID: hAIDδC, a mystery to be explored

Activation-induced cytidine deaminase (AID) is a key player that initiates antibody diversification in activated B-cell. AID mediates somatic hypermutation (SHM) and class switch recombination (CSR) via the deamination of cytosine to uracil at the Ig locus, resulting in the production of high-affinity antibodies. AID is predominantly restricted to Ig genes, whereas off-targeting of AID leads to lymphocyte-related malignancies. Interestingly, apart from FL-AID other splice isoforms of AID are highly expressed in the lymphocyte malignancies. In our study, we found that the heterologous expression of hAID-FL in E. coli cells produced two induced bands of hAID as demonstrated by SDS-PAGE and western blotting. Remarkably, peptide mapping data predicted that one band is hAID-FL and the other is its splice isoform, hAIDδE4a. To get an insight into why E. coli cells expressed hAID-FL and hAID variant, we mutated the 5' and 3' splice site of a putative intron of hAID, but it failed to produce only hAID-FL. Incidentally, hAID expressed with fusion partners also displayed two bands, and peptide mapping data strongly suggest that besides hAID-FL, the lower band showed a significant number of amino acids missing towards the C-terminal domain (named as hAIDδC). Our results are the first report to show that expression of recombinant hAID alone or irrespective of solubilization tags in E. coli cells produced hAID-FL and hAIDδC. It will be fascinating to explore the potential mechanism underlying the expression of hAIDδC from recombinant hAID plasmid in E. coli cells.

Gene structure and splicing in schistosomes

Schistosomes are blood dwelling platyhelminths with a complex life cycle and persist in the definitive host during decades, indicating that they are very successful parasites. The challenge of infecting two hosts from different evolutionary branches, namely an invertebrate snail and a vertebrate, suggests that Schistosomes must display a very sophisticated genetic program to circumvent all the barriers imposed by the hosts' immune systems. Recent large-scale genome and transcriptome data from Schistosomes are facilitating the analysis of gene structure and splicing. Studying the structure of genes coding for secreted proteins is of particular interest since these proteins mediate processes in the host-parasite interface. Recent description of Micro-Exon Genes (MEGs), polymorphic mucin genes (SmPoMucs) and venom allergen-like (SmVALs) proteins with unusual gene structure apparently oriented towards generation of protein variability through alternate splicing, and the presence of multiple copies of these genes, indicate that the parasite developed a sophisticated system to interact with its hosts. This opens up opportunities for further studies with the use of proteomic techniques to better characterize the protein variability created by these systems and their role in parasite survival. Complete description of the functions of these variable proteins will greatly contribute to our understanding of host-parasite interactions.

Stage-specific alternative splicing of the heat-shock transcription factor during the life-cycle of Schistosoma mansoni

Stage-specific alternative splicing of the heat-shock transcription factor of Schistosoma mansoni (SmHSF) generates isoforms with structural diversity that may modulate the activity of SmHSF at different life-stages, and thus may regulate the expression of different genes at different developmental stages. RT-PCR, cloning and DNA-sequence analyses showed stage-specific alternative splicing inside the DNA-binding domain (DBD) involving introns I1 and I2, and beyond the DBD involving introns I4a and I7. Retention of introns I2 and I4 would inactivate SmHSF since they contain termination codons. Retention of intron I1 would add 11 amino acids inside the DBD and may change the DNA-binding specificity of SmHSF; intron I7 would add 13 amino acids to the effector region of HSF. Retention of introns was more pronounced in cercariae (larval stage living in water) than in adult worms (parasitic form in mammals). The isoforms were expressed in bacteria, but functional evaluation was not feasible, because only the isoform lacking introns was soluble while isoforms with introns were insoluble. However, stage-specific alternative splicing that changed HSF function in vivo was evidenced in intact cercariae. The cercarial SmHSF mRNA was enriched with introns I2 and I4a that contain termination codons. Therefore, translation of the SmHSF mRNA was impaired, and the SmHSF protein was undetectable. Consequently, the HSP70 gene could not be transcribed, and the HSP70 mRNA was missing. Alternative splicing was observed for short DNA segments (33-45 bp) bound by splice signals, located in the coding region. These are not bona fida exons since they are not flanked by introns. Yet, they are not regular introns since they are often found in mature mRNA. Alternative splicing of these DNA segments caused structural diversity that could modulate the function of the gene product.

Genomic organization of the Schistosoma mansoni aspartic protease gene, a platyhelminth orthologue of mammalian lysosomal cathepsin D

Schistosomes are considered the most important of the helminth parasites of humans in terms of morbidity and mortality. Schistosomes employ proteolytic enzymes to digest host hemoglobin from ingested human blood, including a cathepsin D-like, aspartic protease that is overexpressed in the gut of the adult female schistosome. Because of its key role in parasite nutrition, this enzyme represents a potential intervention target. To continue exploration of this potential, here we have determined the sequence, structure and genomic organization of the cathepsin D gene locus of Schistosoma mansoni. Using the cDNA encoding S. mansoni cathepsin D as a probe, we isolated several positive bacterial artificial chromosomes (BAC) from a BAC library that represents an approximately 8-fold coverage of the schistosome genome. Sequencing of BAC clone 25-J-24 revealed that the cathepsin D gene locus was approximately 13 kb in length, and included seven exons interrupted by six introns. The exons ranged in length from 49 to 294 bp, and the introns from 30 to 5025 bp. The genomic organization of schistosome cathepsin D was similar in sequence, structure and complexity to human cathepsin D, including to a greater or lesser extent the conservation of all six exon/intron boundaries of the schistosome gene. It was less similar to aspartic protease genes of the nematodes Caenorhabditis elegans and Haemonchus contortus, and dissimilar to those of plasmepsins from malarial parasites. Examination of the introns revealed the presence of endogenous mobile genetic elements including SR2, the ASL-associated retrotransposon, and the SINE-like element, SMalpha. Phylogenetically, schistosome cathepsin D appeared to be more closely related to mammalian cathepsin D than to other sub-families of eukaryotic aspartic proteases known from mammals. Taken together, these features indicated that schistosome cathepsin D is a platyhelminth orthologue of mammalian lysosomal cathepsin D.

Echinococcus granulosus: Cloning and Functional in Vitro Characterization of an Actin Filament Fragmenting Protein

We report the isolation and characterization of an Echinococcus granulosus gene that codes for a protein with actin filament fragmenting and nucleating activities (EgAFFP). The genomic region corresponding to the EgAFFP gene presents a coding sequence of 1110 bp that is interrupted by eight introns. The EgAFFP deduced amino acid sequence is about 40% homologous to those of several members of the gelsolin family, such as Physarum polycephalum fragmin, Dictyostelium discoideum severin, and Lumbricus terrestris actin modulator. As do other proteins of the same family, EgAFFP presents three repeated domains, each one characterized by internal conserved amino acid motifs. Assays with fluorescence-labeled actin showed that the full-length recombinant EgAFFP effectively binds actin monomers in both a calcium-dependent and calcium-independent manner and also presents actin nucleating and severing activities.

Cloning the gene encoding Schistosoma mansoni p50, an immunophilin

A 2.2-kb fragment of genomic DNA encoding Schistosoma mansoni immunophilin p50 (Smp50) was identified on a 14-kb genomic clone. The sequence of Smp50 reveals seven exons interrupted by six small introns ranging from 28-35 bp in size. The transcription start point, defined by primer extension analysis of schistosome RNA, begins at 30 bp upstream from the start AUG codon. Smp50 lacks a TATA box and appears to be a single-copy gene.

RNA trans-splicing in flatworms. Analysis of trans-spliced mRNAs and genes in the human parasite, Schistosoma mansoni

Characteristics of trans-splicing in Schistosoma mansoni were examined to explore the significance and determinants of spliced leader (SL) addition in flatworms. Only a small subset of mRNAs acquire the SL. Analysis of 30 trans-spliced mRNAs and four genes revealed no discernable patterns or common characteristics in the genes, mRNAs, or their encoded proteins that might explain the functional significance of SL addition. While the mRNA encoding the glycolytic enzyme enolase is trans-spliced, mRNAs encoding four other glycolytic enzymes are not, indicating trans-splicing is not prevalent throughout this metabolic pathway. Although the 3' end of flatworm SLs contribute an AUG to mRNAs, the SL AUG does not typically serve to provide a methionine for translation initiation of reading frames in recipient mRNAs. SL RNA expression exhibits no apparent sex, tissue, or cell specificity. Trans-spliced genes undergo both cis- and trans-splicing, and the sequence contexts for these respective acceptor sites are very similar. These results suggest trans-splicing in flatworms is most likely associated either with some property conferred on recipient mRNAs by SL addition or related to some characteristic of the primary transcripts or transcription of trans-spliced genes.

Intracellular promiscuity in Schistosoma mansoni: nuclear transcribed DNA sequences are part of a mitochondrial minisatellite region

It has been shown that the mtDNA of the parasitic trematode Schistosoma mansoni is hypervariable in size. We report here that this length variation is due to a large polymorphic minisatellite composed of two types of repeated sequences of 558 bp and 62 bp. Each minisatellite repeat is made up of a large 558-bp component and a variable tandem array of the small 62-bp unit. Of more fundamental interest was the finding that both the 558-bp and 62-bp components have significant homology with a gene, SM750, previously identified in the nuclear genome of S. mansoni. The small 62-bp unit is identical to the nuclear polymorphic repeat element, which is apparently spread throughout the nuclear genome and is abundant among transcripts, in addition to being present in five tandem copies in SM750. The presence, in the S. mansoni mtDNA, of fragments of genes that are present in and transcribed from the nuclear genome raises the question of the origin of these sequences. The arrangement and the variability that the mtDNA minisatellite embodies were explored as an identity test for S. mansoni based on the use of PCR for tallying the relative abundance of the several repeat numbers of the tandem arrays of the 62-bp unit within the minisatellite structure.

Cloning and characterization of the SmIMP25 integral membrane protein of the parasitic helminth Schistosoma mansoni

The cDNA and genomic clones encoding a 25 kDa integral membrane protein, termed SmIMP25, were isolated from Schistosoma mansoni. The 2.2 kb SmIMP25 mRNA was found in all developmental stages of the parasite tested: miracidium, sporocyst, cercaria and adult worm. The SmIMP25 gene is at least 16 kb long and it is split by four introns ranging in size from 36 bp to > or = 9 kb. Excluding the introns, the gene and the cDNA show 100% sequence identity. The cDNA has an open reading frame encoding a protein 223 amino acids long. The predicted sequence reveals a distinct hydrophobic domain of 20 amino acids located 12 residues from the carboxyl-terminal end. The properties of this domain (marked hydrophobicity, size, flanking by charged residues and C-terminal location) are typical of the transmembrane segments of integral membrane proteins. The presence of three potential N-glycosylation sites is also consistent with membrane proteins that are often glycosylated at the extracellular domain. Accordingly we propose that SmIMP25 is an integral membrane protein in which residues 1-191 are extracellular, residues 192-211 comprise the hydrophobic domain that spans the membrane, and residues 212-223 are intracellular. The SmIMP25 was synthesized as a fusion protein in bacteria and antibodies were elicited in rabbits. Antibodies against SmIMP25 specifically precipitated a 25 kDa protein from cell-free products programmed by schistosome mRNA, in agreement with the size of the protein predicted from the cDNA sequence. Immunofluorescence studies showed SmIMP25 on the surface of the parasite. Surface molecules expressed at the host-parasite interface are likely to provide information on host parasite relationship and may serve as targets for protective immunity.

Dinucleotide biases in the platyhelminth Schistosoma mansoni

The analysis of dinucleotide biases in coding and flanking regions, introns, rDNA and repetitive sequences, in the flatworm Schistosoma mansoni is reported. Except for rDNA, all regions display CpG avoidance and TpG plus CpA excess, which might be evidence of the presence of 5mC. The distribution and hierarchies of dinucleotides differ from the data published for invertebrate and vertebrate coding sequences.

Cloning and characterization of the gene encoding Schistosoma mansoni glutathione peroxidase

Antioxidant enzymes are thought to play a crucial role in the survival of the parasite, Schistosoma mansoni, during its migration through the tissues of the definitive host. We recently cloned the cDNA encoding one such enzyme, glutathione peroxidase (Gpx). In order to elucidate the regulation of expression of this gene, we describe the cloning and characterization of a Gpx gene of S. mansoni. An initial screen of a lambda EMBL4 genomic library using the corresponding cDNA sequence as a probe yielded 14 positive clones, two of which have so far been analyzed in detail. The complete Gpx gene contains five introns, four of which, located at the 5' end, are extremely short (30-51 bp) and the last of which is approximately 6 kb long. We present the sequence of the gene including 73 bp at the 5' end, the complete sequence to 137 bp downstream from the penultimate exon, 164 bp upstream and 131 bp downstream from the last 3' exon. The potential mRNA cap site is situated 219 bp upstream from the ATG start codon. All intron/exon junctions correspond to the conventional eukaryotic splice signal. Analysis of the 5' flanking region revealed the presence of a potential TATA box at--26 bp from the cap site, but no CAAT-like element is present. Southern blot analysis showed a unique Gpx gene organisation in the S. mansoni genome.

Structure of the gene encoding a putative Schistosoma mansoni tegumental antigen precursor

In order to obtain the complete gene encoding the putative precursor of a 15-kDa Schistosoma mansoni tegumental antigen (Sm15), two cDNAs (A70 and A184) and two fragments of independent genomic clones were subcloned and sequenced. The collated sequence contains 4700 nucleotides and represents the full length open reading frame of the gene, encoding a protein of 1032 amino acids with a calculated molecular mass of 116,900. Thus, the gene encodes a much longer protein than that identified in the tegumental membranes, suggesting that it encodes a precursor that is subsequently highly processed. A 964-bp region composed of 5 closely related repeats was found to be present within the translated frame. The predicted protein is highly acidic and there is no indication of hydrophobic domains that may represent transmembrane regions or indicate attachment of a GPI anchor. The coding region has no homologies in the currently available data bases. In the 5' non-transcribed area a copy of the SM alpha repeat family is present. The coding region is preceded by putative CCAAT and TATA boxes that may be involved in the control of expression.

Characterization of the Schistosoma mansoni gene encoding the glycolytic enzyme, triosephosphate isomerase

The complete gene encoding Schistosoma mansoni triosephosphate isomerase (TPI) was isolated from a lambda phage genomic library on 2 overlapping clones. These genomic clones have been characterized by restriction mapping and DNA sequencing of the 5' flanking region, the exons, the intron boundaries and the polyadenylation addition site. S. mansoni TPI is encoded by 6 exons spanning a region of about 12 kb. The 5 introns are located at positions precisely analogous to those of mammalian TPI genes but one of the 6 mammalian TPI introns is missing in S. mansoni. We find no evidence of spliced leader involvement in TPI gene expression. The gene is preceded by at least 4 tandem copies of a 2.5-kb repetitive sequence. While the 12-kb size for the S. mansoni TPI gene is much larger than the 3-4 kb typical of mammalian TPI genes, the 42-bp first intron is unusually short. The transcription initiation site for the S. mansoni TPI gene is heterogeneous. Genomic Southern blot analysis suggests that TPI is expressed from a single copy gene.

Isolation of purine auxotrophic mutants of Lactococcus lactis and characterization of the gene hpt encoding hypoxanthine guanine phosphoribosyltransferase

Five purine auxotrophic mutants of Lactococcus lactis were isolated. L. lactis was capable of converting adenine, guanine and hypoxanthine to AMP, GMP and IMP, respectively, indicating the existence of adenine phosphoribosyltransferase (APRT) and hypoxanthine guanine phosphoribosyltransferase (HGPRT) activities. A 1.3 kb DNA fragment from L. lactis was cloned by complementation of the hpt mutation in Escherichia coli. Introduction of this fragment into L. lactis resulted in an increase in HGPRT activity. In vitro transcription and translation analysis showed that the fragment coded for a polypeptide with M(r) of 22,000. The nucleotide sequence of this hpt gene was determined.

Alternative splicing of the Schistosoma mansoni gene encoding a homologue of epidermal growth factor receptor

The complete coding DNA for a Schistosoma mansoni homologue of the epidermal growth factor receptor (SER) was characterized from cDNA clones obtained by homology to the tyrosine kinase domain of erbB. The DNA sequence predicts a 200-kDa translation product that contains a secretory leader, a cysteine-rich extracellular domain, a hydrophobic transmembrane sequence, and an intracellular tyrosine kinase domain. The SER transcript is present in cercariae and adult schistosomes. In addition to SER transcripts, schistosomes produce at least 3 variant transcripts encoding truncated SER products that include the secretory leader and a small portion of the extracellular domain followed by short sequences of unrelated, C-terminal amino acids. Based on these sequences, 2 of the variant mRNAs (class 2 and 5) appear to encode soluble, secreted proteins while one (class 4) encodes an SER variant protein with a hydrophobic C-terminus that may serve as a membrane anchor. Class 2 SER variant transcripts are present at levels comparable to SER transcripts in adult worms but are not detected in cercariae. Class 4 and 5 SER variant transcripts are also found within adult worms but at lower levels. Genomic cloning and characterization demonstrate that the variant SER transcripts arise through alternative splicing of the SER gene.

Gut-specific and developmental expression of a Caenorhabditis elegans cysteine protease gene

A Caenorhabditis elegans cysteine protease gene fragment, amplified by PCR using conserved eukaryotic protease gene sequences as primers, was used as a probe to isolate cDNA and genomic clones. The genomic clone, which had a coding sequence of 987 bp interrupted by 2 small introns, was physically mapped to the middle of linkage group V. The predicted amino acid sequence of the mature C. elegans cysteine protease was homologous to those of other eukaryotic cysteine proteases, particularly to that of the nematode parasite Haemonchus contortus (50%) and to the cathepsin B-like hemoglobinase of the trematode parasite Schistosoma mansoni (54%). The pro region of the C. elegans protease was homologous only to that of the H. contortus enzyme, implying a similar mechanism of protease activation. The C. elegans cysteine protease gene was temporally regulated: abundant 1.1-kb transcripts were detected in larvae and adults, but not in embryos. Transcription also was spatially regulated, occurring only in the intestine. Like the vitellogenin genes, which also are transcribed exclusively in the intestine, the 5' end of the C. elegans cysteine protease gene had at least one copy of each of 2 heptameric sequences which may be transcriptional regulatory elements governing gut-specific expression.

Characterisation of the gene encoding a candidate vaccine antigen of Theileria parva sporozoites

We have cloned and characterised the gene encoding the 67-kilodalton stage-specific surface antigen, p67, of Theileria parva (Muguga) sporozoites. The gene which is present in a single copy, is divided into 2 exons by an intron 29 bp long and is transcribed into mRNA of about 2500 nucleotides. The gene is present in all stocks of T. parva and there is a related gene in Theileria annulata. The deduced amino acid sequence of 709 residues predicts that p67 is a membrane protein and that it lacks tandemly repeated sequences. Recombinant p67 has been expressed in Escherichia coli as a fusion protein with Sj-26, a glutathione-S-transferase of Schistosoma japonicum. Antibodies to purified recombinant proteins containing residues 9-316 or 397-709 of p67 bind to p67 in immunoblots and neutralise sporozoite infectivity in vitro. Recombinant p67 is, therefore, a candidate antigen for development of an anti-sporozoite vaccine for East Coast fever in cattle.

Cloning and sequence analysis of the Schistosoma mansoni membrane glycoprotein antigen gene GP22

A family of Schistosoma mansoni proteins (18-22 kDa, pI 5.3-5.8) are biosynthesized in juvenile worms and immunoprecipitated by antibodies uniquely present in protective Fischer rat antiserum. A cDNA clone, lambda gt11-40, expressing epitopes common to this protein family was used to obtain a genomic DNA clone, by hybridization with a lambda gt11-40 oligonucleotide probe. In the 1.37 kb of genomic DNA sequenced, an open reading frame of 182 amino acids was identified on the strand corresponding to lambda gt11-40 coding sequences, and those of identical independently isolated cDNA clones defining a 25-kDa surface membrane glycoprotein. The new S. mansoni gene is termed GP22. There are two candidate promoters, confirmed by primer extension studies with worm RNA. Promoter 1 (P1) is preceded by a G + C-rich region and potential CAAT sequences, and is to the 5'-side of P2. Transcription from P1 is initiated at 2 different sites, apparently producing mRNAs with different translation start sites (ATG). Decoding these mRNAs yields protein products of 182 (P1), 175 (P1), 140 (P2) and 136 (P2) amino acids. The polypeptides share the following features: a hydrophobic segment near the carboxy terminus sufficient to span a lipid bilayer, with a consensus sequence for thio-esterification by a fatty acid; an external domain containing 2 potential N-linked glycosylation sites; and a candidate leucine-zipper motif, suggesting the protein may exist as a dimer on the worm surface. While sharing these common features in their carboxy terminal regions, the three proteins differ in the length and properties of their amino termini. The 140-amino acid protein has a short hydrophobic amino terminus, while the 175- and 182-amino acid proteins have more extensive hydrophobic sequences, each preceded by a hydrophilic amino terminal sequence. The heterogeneity observed in 2-dimensional gels of the antigen may be explained in part by the size and charge differences among the proteins deduced from the sequence and transcription pattern of this gene. The possibility of stage-specific regulated expression of this candidate vaccine antigen family is an attractive concept, potentially accounting for the phenomenon of concomitant immunity observed in the rat and perhaps other schistosome hosts.

Mutational analysis of the structure and function of the adenine phosphoribosyltransferase enzyme of Chinese hamster

We have analyzed the adenine phosphoribosyltransferase (APRT) enzyme from Chinese hamster ovary cells through the study of mutants that are able to grow in the presence of the toxic adenine analogue 8-azaadenine. The distribution of the amino acid alterations was analyzed in terms of the binding regions for the purine and phosphoribosylpyrophosphate substrates and a comparison was made with mutants known in human APRT and human, mouse and hamster hypoxanthine-guanine phosphoribosyltransferase. A number of mutants were found to cluster in several regions of the amino acid sequence. Residual enzyme activity with adenine was determined and this was correlated with substrate binding regions. A model of the secondary structure features is proposed.