Identification and characterization of the immunogenic cytotoxic TvCP39 proteinase gene of Trichomonas vaginalis

TvCP39 is a 39 kDa cysteine proteinase (CP) involved in Trichomonas vaginalis cytotoxicity that has been found in vaginal secretions and is immunogenic in patients with trichomonosis. The goal of this work was to identify, clone, express, and characterize the tvcp39 gene. The tvcp39 gene was identified using a proteomic approach, and the complete gene was amplified using PCR, cloned, and sequenced. TvCP39 is encoded by a 915-bp cathepsin L-like CP gene. A fragment corresponding to the mature region (TvCP39r) was expressed, purified, and used to produce rabbit polyclonal antibodies and in functional assays. In one- and two-dimensional western blot assays, the anti-TvCP39r antibody reacted with two protein bands of ~28 and 27 kDa and three spots of ~28, 27, and 24 kDa in trichomonad proteinase-rich extracts that could correspond to the mature and processed fragments of the TvCP39 peptidase. The anti-TvCP39r antibody reacted with the parasitic surface and the native TvCP39 present in vaginal washes from patients with trichomonosis. Moreover, the recombinant TvCP39 protein bound to the surface of HeLa cells and protected HeLa cell monolayers from trichomonal destruction in a concentration-dependent manner. In conclusion, our data support TvCP39 as one of the surface proteinases that is glycosylated and is involved in trichomonal cytotoxicity. Thus, TvCP39 is the first glycosylated cysteine proteinase detected in T. vaginalis.     

Site-Directed Spin-Labeling Study of the Light-Harvesting Complex CP29

The topology of the long N-terminal domain (∼100 amino-acid residues) of the photosynthetic Lhc CP29 was studied using electron spin resonance. Wild-type protein containing a single cysteine at position 108 and nine single-cysteine mutants were produced, allowing to label different parts of the domain with a nitroxide spin label. In all cases, the apoproteins were either solubilized in detergent or they were reconstituted with their native pigments (holoproteins) in vitro. The spin-label electron spin resonance spectra were analyzed in terms of a multicomponent spectral simulation approach, based on hybrid evolutionary optimization and solution condensation. These results permit to trace the structural organization of the long N-terminal domain of CP29. Amino-acid residues 97 and 108 are located in the transmembrane pigment-containing protein body of the protein. Positions 65, 81, and 90 are located in a flexible loop that is proposed to extend out of the protein from the stromal surface. This loop also contains a phosphorylation site at Thr81, suggesting that the flexibility of this loop might play a role in the regulatory mechanisms of the light-harvesting process. Positions 4, 33, 40, and 56 are found to be located in a relatively rigid environment, close to the transmembrane protein body. On the other hand, position 15 is located in a flexible region, relatively far away from the transmembrane domain.     

Deletion analysis of yeast Sec65p reveals a central domain that is sufficient for function in vivo

The Saccharomyces cerevisiae SEC65 gene encodes a 32 kDa subunit of yeast signal recognition particle that is homologous to human SRP19. Sequence comparisons suggest that the yeast protein comprises three distinct domains. The central domain (residues 98–171) exhibits substantial sequence similarity to the 144 residue SRP19. In contrast, the N-terminal and C-terminal domains (residues 1–97 and 172–273 respectively) share no similarity to SRP19, with the exception of a cluster of positively charged residues at the extreme C-terminus of both proteins. Here, we report the cloning of a Sec65p homologue from the yeast Candida albicans that shares the same extended domain structure as its S. cerevisiae counterpart. This conservation of sequence is reflected at the functional level, as the C. albicans gene can complement the conditional lethal sec65-1 mutation in S. cerevisiae . In order to examine the role of the N- and C- terminal domains in Sec65p function, we have engineered truncation mutants of S. cerevisiae SEC65 and tested these for complementing activity in vivo and for SRP integrity in vitro . These studies indicate that a minimal Sec65p comprising residues 76–209, which includes the entire central SRP19-like domain, is sufficient for SRP function in yeast.     

Cerato-platanin, a phytotoxic protein from Ceratocystis fimbriata: expression in Pichia pastoris, purification and characterization

Cerato-platanin (CP) is a phytotoxic protein secreted by the Ascomycete Ceratocystis fimbriata f.sp. platani. This Ascomycete causes canker stain which is a severe disease with a high incidence in the European Platanus acerifolia. CP probably plays a role in the disease, eliciting defence-related responses in the host plants. CP is a 120 amino acid protein, containing 40% hydrophobic residues and two S-S bridges. In the EMBL data bank CP is the first member of a new fungal protein family known as the Cerato-Platanin Family. The N-terminal region of CP shows a high similarity with that of cerato-ulmin, a phytotoxic protein produced by the Ophiostoma species and that belongs to the hydrophobin family. Hydrophobins are hydrophobic proteins secreted by many saprophytic or pathogenic fungi and have a remarkable ability to self-assemble into a rodlet structure takes part in physiological and/or pathological processes. The methyltrophic yeast Pichia pastoris was used to obtain a high-level expression of recombinant CP (rCP) and the pPIC9 vector was chosen to bring about extra-cellular secretion of the protein. The preliminary structural and functional characterization presented here reveals no significant differences between the native and the recombinant protein. We also show that CP self-assembles in solution. The availability of rCP will allow its three-dimensional structure to be determined, facilitating an understanding of the role of CP in the pathogenesis of canker stain. It is also an excellent model for investigating the mechanism of action of the other proteins related to CP.     

Characterization of an apoptosis inhibitory domain at the C-termini of FE65-like protein

TR2(L) is a 56-amino-acid polypeptide that has been shown to block TNF cytotoxicity. FE65-like (FE65L) proteins possess this conserved TR2(L) sequence at their C-termini, whereas variations in the sequences are found in the FE65 proteins. To further analyze the antiapoptotic function of TR2(L), here we utilized an isolated murine partial FE65L cDNA that encodes an N-terminal phosphotyrosine-binding domain (PTB) and the conserved C-terminal TR2(L) sequence. When L929 cells were stably transfected with the FE65L cDNA or its 3' end TR2(L) DNA sequence, these cells became resistant to TNF killing. Replacement of the N-terminal PTB domain with GFP failed to abolish the FE65L-mediated TNF resistance. Ablation of the C-terminal TR2(L) sequence through frame-shift mutation resulted in a complete loss of the FE65L function against TNF. Various protein kinase inhibitors, including lavendustin A, tyrphostin, H7, and staurosporine, which may affect the PTB domain function, could not abolish the FE65L-mediated TNF resistance. A prolonged exposure of L929 cells to these inhibitors for 24 h resulted in cell death, whereas FE65L significantly blocked the cell death. Polyclonal antibodies were generated against a synthetic peptide and shown to interact with a 38-kDa FE65L in L929 cells. Hyaluronidase downregulates the expression of FE65L gene and protein in L929 cells, and this correlates with its enhancement of TNF killing of these cells. Together, our data indicate that the TR2(L) amino acid sequence is an apoptosis-inhibitory domain commonly present in the FE65 and FE65-like family proteins.     

FE65 interaction with the ApoE receptor ApoEr2

The adaptor protein FE65 interacts with the beta-amyloid precursor protein (APP) via its C-terminal phosphotyrosine binding (PTB) domain and affects APP processing and Abeta production. Our previous data demonstrate that the apoE receptor ApoEr2 co-precipitated with APP and suggest that there are extracellular and intracellular interactions between these two transmembrane proteins. We hypothesized that FE65 acts as an intracellular link between ApoEr2 and APP. Co-immunoprecipitation experiments in COS7 cells demonstrated an interaction between ApoEr2 and FE65 that depended on the N-terminal PTB domain of FE65. Full-length FE65 increased co-immunoprecipitation of ApoEr2 and APP. Full-length FE65 also increased surface expression of ApoEr2, as determined by surface protein biotinylation and live cell surface staining. Constructs containing both the C- and N-terminal PTB domains of FE65 increased secreted APP, secreted ApoEr2, APP C-terminal fragment, and ApoEr2 C-terminal fragment, but constructs containing only single PTB domains did not affect APP or ApoEr2 processing. In addition, full-length FE65 decreased Abeta to a significantly greater extent than individual FE65 domains. These data suggest that FE65 can bind APP and ApoEr2 at the same time and affect the processing of each.     

Envelope proteins of human T cell leukemia virus type I: characterization by antisera to synthetic peptides and identification of a natural epitope

Three peptides corresponding to selected regions of the env gene products of human T cell leukemia virus type I were synthesized by solid-phase Merrifield techniques. The sequence of peptide designated SP-65 was identical to the predicted C-terminal 12 residues of the transmembrane protein p21env, and peptide SP-74 was inferred from a region shown to be highly conserved among mammalian retroviruses. The third peptide, SP-70, was derived from a C-terminal region of the surface glycoprotein gp46. Antibodies to each peptide were raised in rabbits and were used to identify and further characterize the proteins coded by the env gene. Despite being present at very low levels in purified viral preparations, these proteins were chromatographed by reverse-phase high pressure liquid chromatography and were located by Western blot analysis of the column fractions. Anti-SP-70 recognized the surface glycoprotein (gp46) and also its C-terminal cleavage fragment (gp16). Anti-SP-65 and anti-SP-74 both reacted with the hydrophobic transmembrane protein (p21) and provided evidence that this protein does not undergo apparent C-terminal processing during viral maturation, unlike the trans-membrane protein of murine leukemia virus. As expected, anti-SP-74 also reacted with homologous proteins from other Type C and Type D viruses, confirming that peptide SP-74 corresponds to a broadly conserved region of retroviral transmembrane proteins. SP-70, which is predicted to be quite near the C terminus of the major surface glycoprotein, was also reactive with sera of HTLV-I-positive patients, indicating that this peptide corresponds to, or is part of, a native epitope recognized by the natural host.     

Targeting of specific domains of diphtheria toxin by site-directed antibodies.

Antibodies highly selective for two functionally distinct regions of diphtheria toxin (DTx) were prepared using synthetic peptide conjugates as immunogens. Three peptides were selected for synthesis: sequence DTx141-157 on fragment A, which contains the putative protein elongation factor (EF-2) ADP-ribosyltransferase site; DTx224-237 on fragment B, selected on the basis of forming a predicted surface loop; and DTx513-526 on fragment B, forming a part of the region containing the putative receptor binding domain. All of the anti-peptide antibodies recognized the corresponding peptide, and also reacted with the toxin, specifically with the fragment containing the sequence against which they were raised, confirming the utility of this approach in generating fragment-specific antibodies. The anti-peptide antibody with the highest binding titre both to the peptide and to the native toxin was the one prepared against the sequence with the highest surface and loop likelihood indices of the three peptides selected. The similarity of the reactivity profiles with peptide and native and denatured toxin is consistent with the prediction that the region selected occurs in a surface loop and that the structure of the peptide is similar to the conformation of this region in the native protein. The epitopes for two of the anti-peptide antibodies were mapped. The results indicated that even though the antisera were raised to peptides containing 14 amino acids (aa) they were directed predominantly against a narrow region within the peptide, consisting of only 5-6 aa residues.(ABSTRACT TRUNCATED AT 250 WORDS)     

Immunological cross-reactivity between human tripeptidyl peptidase II and fibronectin.

Tripeptidyl peptidase II (TPP II) is a large intracellular exopeptidase with an active site of the subtilisin type. Affinity-purified hen antibodies against human erythrocyte TPP II cross-reacted with fibronectin in an immunoblot analysis. Furthermore, antibodies against human fibronectin cross-reacted with TPP II. Antibodies against a 65 kDa cell-binding fragment of fibronectin specifically reacted with TPP II, whereas antibodies against the collagen-binding domain, the main heparin-binding domain or the N-terminal fibrin-binding domain did not react. Moreover, the affinity-purified antibodies against TPP II reacted with a 105 kDa cell-binding fragment of fibronectin but not with the fibrin-binding domain or the collagen-binding domain. When native TPP II was dissociated into smaller units through dialysis against a dilute Tris buffer, it could be digested by chymotrypsin into three stable fragments of 70 kDa, 42 kDa and 20 kDa. It could be demonstrated that the 42 kDa fragment was specifically recognized by antibodies against the 65 kDa cell-binding fragment of fibronectin. Furthermore, labelling with di-[3H]isopropyl phosphorofluoridate and N-terminal sequence determination showed that the 70 kDa fragment contained the active-site serine residue. In conclusion, our findings suggest that one domain of the TPP II molecule bears structural resemblance to a cell-binding fragment of fibronectin.     

Comparative structural analysis of human and rat 65 kDa tumor-associated phosphoproteins

• 1.1. A 65 kDa-tumor-associated protein (p65) was isolated from human and rat carcinoma cell culture media. Antibodies raised to the rat protein recognized an antigenically related protein in human cancer cell line.
• 2.2. Amino acid composition, N-terminal and internal sequence as well as peptide map and western blot analysis of the p65 strongly suggest a high degree of homology between the human and rat p65 proteins.
• 3.3. Homology searches indicated that p65 was not homologous to previously sequenced proteins, but that it may be related to proteins of the steroid receptor superfamily of genes, especially c-erb A gene.

     

Primary structure of a mouse mastocytoma proteoglycan core protein.

The complete nucleotide sequence of a mouse mastocytoma proteoglycan core protein mRNA was determined. The mRNA, estimated to contain 1.1 kb, encodes a protein with an Mr of 16715. A 21-amino acid-residue region of the protein is composed of alternating serine and glycine residues. Southern-blot analysis of mouse genomic DNA with cDNA containing sequences corresponding to the Ser-Gly repeat region revealed more than 15 gene fragments. Hybridization with a probe corresponding to the N-terminal portion of the core protein identified two fragments, and cDNA covering the C-terminal part of the core protein and the 3' untranslated part of the mRNA hybridized to a single fragment. Antibodies against the core protein, obtained after immunization of rabbits with a fusion protein, reacted with both chondroitin sulphate proteoglycans and heparin proteoglycans produced by the tumour. In immunoblotting of a microsomal fraction from the mastocytoma, the antiserum recognized a single protein (Mr 17,000), which probably represents the core protein before glycosylation.     

Opioid-binding cell adhesion molecule (OBCAM)-related clones from a rat brain cDNA library.

The AKin10 gene from Arabidopsis thaliana encoding a putative Ser/Thr protein kinase (PK) has been isolated and characterized. The AKin10-encoding gene is located on a genomic 5.4-kb BamHI fragment and contains ten introns, one being located in the 5' untranslated region. The deduced amino acid sequence of AKin10 is 65% identical over the catalytic domain to the yeast PK (SNF1). SNF1 is essential for the derepression of many glucose-repressible genes, including Suc2 which encodes invertase. Southern blot hybridization experiments suggested the presence of one copy of the gene per haploid genome of A. thaliana. Northern hybridization experiments indicated that this gene is expressed in roots, shoots and leaves. AKin10 may play an important role in a signal transduction cascade regulating gene expression and carbohydrate metabolism in higher plants.     

Co-operative binding of human fibronectin to Sfbl protein triggers streptococcal invasion into respiratory epithelial cells

Streptococcal fibronectin binding protein I (SfbI) mediates adherence to and invasion of Streptococcus pyogenes into human epithelial cells. In this study, we analysed the binding activity of distinct domains of SfbI protein towards its ligand, the extracellular matrix component fibronectin, as well as the biological implication of the binding events during the infection process. By using purified recombinant SfbI derivatives as well as in vivo expressed SfbI domains on the surface of heterologous organism Streptococcus gordonii , we were able to dissociate the two major streptococcal target domains on the human fibronectin molecule. The SfbI repeat region exclusively bound to the 30 kDa N-terminal fragment of fibronectin, whereas the SfbI spacer region exclusively bound to the 45 kDa collagen-binding fragment of fibronectin. In the case of native surface-expressed SfbI protein, an induced fit mode of bacteria–fibronectin interaction was identified. We demonstrate that binding of the 30 kDa fibronectin fragment to the repeat region of SfbI protein co-operatively activates the adjacent SfbI spacer domain to bind the 45 kDa fibronectin fragment. The biological consequence arising from this novel mode of fibronectin targeting was analysed in eukaryotic cell invasion assays. The repeat region of SfbI protein is mediating adherence and constitutes a prerequisite for subsequent invasion, whereas the SfbI spacer domain efficiently triggers the invasion process of streptococci into the eukaryotic cell. Thus, we were able to dissect bacterial adhesion from invasion by manipulating one protein. SfbI protein therefore represents a highly evolved prokaryotic molecule that exploits the host factor fibronectin not only for extracellular targeting but also for its subsequent activation that leads to efficient cellular invasion.     

Characterization of a gene encoding a Pichia pastoris protein disulfide isomerase

Protein disulphide isomerases belong to the thioredoxin superfamily of protein-thiol oxidoreductases that have two double-cysteine redox-active sites and take part in protein folding in the endoplasmic reticulum (ER). We report here the cloning of a Pichia pastoris genomic DNA fragment (2919 bp) that encodes the full length of a protein disulphide isomerase (PpPDI). The deduced amino acid sequence of PDI consists of 517 residues and carries the two characteristic PDI-type redox-active domains -CGHC-, separated by 338 residues, and two potential N-glycosylation sites. The N-terminal end forms a putative signal sequence, and an acidic C-terminal region represents a possible calcium-binding domain. Together with the -HDEL ER retrieval sequence at the C-terminus, these features indicate that the gene encodes a redox-active ER-resident protein disulphide isomerase. The nucleotide sequence, which also contains two other open reading frames, has been submitted to the EMBL Nucleotide Sequence Database, Accession No. AJ302014.     

Exploring the structure of the N-terminal domain of CP29 with ultrafast fluorescence spectroscopy

A high-throughput Förster resonance energy transfer (FRET) study was performed on the approximately 100 amino acids long N-terminal domain of the photosynthetic complex CP29 of higher plants. For this purpose, CP29 was singly mutated along its N-terminal domain, replacing one-by-one native amino acids by a cysteine, which was labeled with a BODIPY fluorescent probe, and reconstituted with the natural pigments of CP9, chlorophylls and xanthophylls. Picosecond fluorescence experiments revealed rapid energy transfer (~20–70 ps) from BODIPY at amino-acid positions 4, 22, 33, 40, 56, 65, 74, 90, and 97 to Chl a molecules in the hydrophobic part of the protein. From the energy transfer times, distances were estimated between label and chlorophyll molecules, using the Förster equation. When the label was attached to amino acids 4, 56, and 97, it was found to be located very close to the protein core (~15 Å), whereas labels at positions 15, 22, 33, 40, 65, 74, and 90 were found at somewhat larger distances. It is concluded that the entire N-terminal domain is in close contact with the hydrophobic core and that there is no loop sticking out into the stroma. Most of the results support a recently proposed topological model for the N-terminus of CP29, which was based on electron-spin-resonance measurements on spin-labeled CP29 with and without its natural pigment content. The present results lead to a slight refinement of that model.     

Cloning and characterization of a gene, mpsA, encoding a protein associated with intracellular magnetic particles from Magnetospirillum sp. strain AMB-1

Proteins located within the lipid bilayer, surrounding the intracellular bacterial magnetic particles (BMP) from Magnetospirillum sp. AMB-1, were separated using SDS-PAGE. Several major proteins of approximate molecular weight 66.2, 35.6, and 24.8 kDa were identified. The N-terminal amino acid sequence of one of these proteins, designated MpsA, was determined and used to design a pair of PCR primers which amplified a 105 bp DNA fragment from AMB-1 genomic DNA. Gene-walking, using anchored PCR, was used to determine the complete nucleotide sequence (954 bp) of the mpsA gene. The mpsA encodes a 317 amino acid protein which does not have an N-terminal cytoplasmic transport signal sequence. Intracellular localization studies were carried out using an mpsA-luc gene fusion expressed in AMB-1 following gene transfer by conjugation. The gene fusion was constructed by cloning a 1.6 kb mpsA fragment upstream of luc in the conjugal plasmid pKLC. The MpsA-Luc fusion protein was preferentially located on the magnetic particle membrane. Although the function of MpsA remains unknown, homology searches suggest similarity with the alpha subunit of acetyl-CoA carboxylase and the CoA-binding motif.