Using Bayesian multinomial classifier to predict whether a given protein sequence is intrinsically disordered

Intrinsically disordered proteins (IDPs) lack a well-defined three-dimensional structure under physiological conditions. Intrinsic disorder is a common phenomenon, particularly in multicellular eukaryotes, and is responsible for important protein functions including regulation and signaling. Many disease-related proteins are likely to be intrinsically disordered or to have disordered regions. In this paper, a new predictor model based on the Bayesian classification methodology is introduced to predict for a given protein or protein region if it is intrinsically disordered or ordered using only its primary sequence. The method allows to incorporate length-dependent amino acid compositional differences of disordered regions by including separate statistical representations for short, middle and long disordered regions. The predictor was trained on the constructed data set of protein regions with known structural properties. In a Jack-knife test, the predictor achieved the sensitivity of 89.2% for disordered and 81.4% for ordered regions. Our method outperformed several reported predictors when evaluated on the previously published data set of Prilusky et al. [2005. FoldIndex: a simple tool to predict whether a given protein sequence is intrinsically unfolded. Bioinformatics 21 (16), 3435-3438]. Further strength of our approach is the ease of implementation.     

Gir2 is an intrinsically unstructured protein that is present in Saccharomyces cerevisiae as a group of heterogeneously electrophoretic migrating forms

Gir2 is a highly acidic cytoplasmic protein of Saccharomyces cerevisiae of unknown function that shows an anomalous migration on SDS-PAGE. Based on its large Stokes radius and thermostability, we have previously suggested that Gir2 lacks extensive secondary structure. Here we report that Gir2 is extremely sensitive to proteolysis when compared to glutathione-S-transferase, a highly structured protein, further indicating its unfolded nature. Prediction based on the FoldIndex program also indicates that Gir2 is a disordered protein. Using truncated forms of Gir2 we show that the N-terminal half of this protein, with its high content of acidic amino acid residues, is responsible for the anomalous electrophoretic behavior of Gir2. Because all these features are hallmarks of intrinsically unstructured proteins (IUP), we propose that Gir2 is another representative of the IUP group of proteins. Additionally, we describe that the endogenous yeast Gir2 shows heterogeneous electrophoretic mobility, which is not due to proteolytic cleavage.     

The human cardiac hormone fragment N-terminal pro B-type natriuretic peptide is an intrinsically unstructured protein

The cardiac hormone B-type natriuretic peptide (BNP) is synthesized as a prepro 134 residue molecule which is further proteolytically processed into a 76 residue fragment termed N-terminal proBNP (NT-proBNP) and the active portion of this hormone, a 32-residue disulfide-linked peptide (BNP-32). The active hormone regulates cardiac hemodynamic output while as yet no biological function has been attributed to NT-proBNP. Some solution properties of synthetically generated NT-proBNP in benign media are known. The protein is monomeric, elutes aberrantly on size-exclusion chromatography as an apparent larger molecular species, and possesses little global secondary structure as assessed by circular dichroism. To explore the solution structure of NT-proBNP in greater detail, we use 2D-NOESY and 2D-TOCSY NMR on recombinant NT-proBNP to obtain a high resolution solution conformation at the alpha-carbon level. Importantly, NH(i)-NH(i+1) coupling is virtually absent at room temperature implying that large stretches of primary sequence are unordered. Together, the results of these physicochemical measurements classify NT-proBNP as a naturally unfolded protein referred to as an Intrinsically Unstructured Protein (IUP). The calculations of FoldIndex, a computer program which predicts disorder, were compared to the experimental results described here for NT-proBNP in addition to proBNP. NT-proBNP thus appears to be an ideal candidate for the study of native, unfolded proteins.     

Structural and posttranslational analysis of human calcium-binding protein,spermatid-associated 1

Recently, we detected a novel biomarker in human saliva called calcium-binding protein, spermatid-associated 1 (CABS1). CABS1 protein had previously been described only in testis, and little was known of its characteristics other than it was considered a structurally disordered protein. Levels of human CABS1 (hCABS1) in saliva correlate with stress, whereas smaller sized forms of hCABS1 in saliva are associated with resilience to stress. Interestingly, hCABS1 also has an anti-inflammatory peptide sequence near its carboxyl terminus, similar to that of a rat prohormone, submandibular rat 1. We performed phylogenetic and sequence analysis of hCABS1. We found that from 72 CABS1 sequences currently annotated in the National Center for Biotechnology Information protein database, only 14 contain the anti-inflammatory domain “TxIFELL,” all of which are primates. We performed structural unfoldability analysis using PONDER and FoldIndex and discovered three domains that are highly disordered. Predictions of three-dimensional structure of hCABS1 using RaptorX, IonCom, and I-TASSER software agreed with these findings. Predicted neutrophil elastase cleavage density also correlated with hCABS1 regions of high structural disorder. Ligand binding prediction identified Ca²⁺, Mg²⁺, Zn²⁺, leucine, and thiamine pyrophosphate, a pattern observed in enzymes associated with energy metabolism and mitochondrial localization. These new observations on hCABS1 raise intriguing questions about the interconnection between the autonomic nervous system, stress, and the immune system. However, the precise molecular mechanisms involved in the complex biology of hCABS1 remain unclear. We provide a detailed in silico analysis of relevant aspects of the structure and function of hCABS1 and postulate extracellular and intracellular roles.     

Identification of PDZ5, a candidate universal minicircle sequence binding protein of Trypanosoma cruzi

The dodecamer universal minicircle sequence is a conserved sequence present in minicircles of trypanosomatid kinetoplast DNA studied so far. This sequence is recognised by a protein named universal minicircle sequence binding protein, described for Crithidia fasciculata, involved in minicircle DNA replication. We have identified a Trypanosoma cruzi gene homologue of the Crithidia fasciculata universal minicircle sequence binding protein. Similar to the Crithidia fasciculata universal minicircle sequence binding protein, the Trypanosoma cruzi protein, named PDZ5, contains five zinc finger motifs. Pulsed field gel electrophoresis indicated that the pdz5 gene is located in the chromosomal band XX of the Trypanosoma cruzi genome. The predicted amino acid sequence of PDZ5 shows a high degree of similarity with several trypanosomatid zinc finger proteins. Specific antibody raised against Crithidia fasciculata universal minicircle sequence binding protein recognises both the recombinant and endogenous PDZ5. The complete pdz5 coding sequence cloned in bacteria expresses a recombinant PDZ5 protein that binds specifically to the universal minicircle sequence dodecamer. These data strongly suggest that PDZ5 represents a Trypanosoma cruzi universal minicircle sequence binding protein.     

DNA sequence of a human Sm autoimmune antigen. The multigene family contains a processed pseudogene

The sequence of a complementary DNA clone coding for a human autoimmune antigen has been determined. This DNA sequence predicts the amino acid sequence of a small protein ("E") which is associated with small nuclear RNA in human cells. Analysis of the predicted protein sequence suggests that the E protein is not closely related to other nucleic acid binding proteins. Screening of a human genomic DNA library has led to the isolation of several members of the E protein multigene family. Sequence analysis of one member of this family reveals that it is flanked by direct repeats and contains several mutations. One of these mutations, an insertion, terminates the long open reading frame. These features are compatible with the designation of this sequence as a processed pseudogene.     

Complete cDNA sequence of a Dictyostelium ubiquitin with a carboxy-terminal tail and identification of the protein using an anti-peptide antibody

The complete sequence of a Dictyostelium discoideum cDNA is presented that codes for monoubiquitin extended at its C-terminus by a 52 amino acid tail. The sequence of both the ubiquitin portion and the tail is highly homologous to the one of Saccharomyces cerevisiae and to a partial mouse sequence. The highly basic tail sequence contains a putative metal and nucleic acid-binding motif. The gene encoding the 0.6 kb mRNA of the C-terminally extended ubiquitin is represented only once in the haploid genome. The 0.6 kb mRNA as well as its translation product, a 15 kDa protein, is expressed in exponentially growing cells and remains present for at least 5 h of development. Using antibodies against a synthetic peptide that corresponds to the C-terminal amino acid sequence, a 15 kDa protein containing the extension a synthetic peptide that corresponds to the C-terminal amino acid sequence, a 15 kDa protein containing the extension was also detected in yeast.     

Identification of a single-copy gene encoding a Type I chlorophyll a/b-binding polypeptide of photosystem I in Arabidopsis thaliana

We have isolated and sequenced cDNA and genomic clones from Arabidopsis thaliana which specify a 241 residue protein with 84% sequence identity to a photosystem I Type I chlorophyll a/b -binding (CAB) protein from tomato. The open reading frame is interrupted by three introns which are found at equivalent positions as the corresponding introns in the tomato gene. Comparison to the amino acid sequence of other CAB proteins confirms that all CAB proteins share two regions of very high similarity. However, near the N-terminus and between the conserved regions this light-harvesting complex I (LHCI) protein, as other LHCI proteins from other plant species, has sequence motifs which appear to be PSI-specific. Restriction analysis of genomic DNA shows that the Arabidopsis protein is encoded by a single-copy gene.     

Isolation of a human cDNA encoding a 25 kDa FK-506 and rapamycin binding protein.

Recently, the nearly complete peptide sequence of a 25 kDa rapamycin and FK-506 binding protein that had been isolated from calf thymus, brain, and spleen was reported (1). Based upon the amino acid sequence of this bovine protein, bFKBP25, we have isolated from a JURKAT cDNA library the cDNA encoding the human homolog, hFKBP25. Translation of the open reading frame contained within this cDNA clone yields a sequence that, in its C-terminal half, is 41% identical to the major human FK-506 binding protein, hFKBP12, and 43% identical to hFKBP13. The N-terminal half of hFKBP25 is unrelated to any known protein.     

Characterization of a region in mature LamB protein that interacts with a component of the export machinery of Escherichia coli

It has been shown that the synthesis of an export-defective protein can interfere with the normal export process in Escherichia coli by limiting the availability of SecB protein, a component of the export apparatus (Collier, D.N., Bankaitis, V.A., Weiss, J.B., and Bassford, P.J. (1988) Cell 53, 273-283). Consistent with this observation, we find that the interference elicited by an export-defective LamB protein is a titratable response resulting from the limitation of a single ligand. We have mapped the interfering region in LamB to between amino acids 320 and 380 of the mature protein. Expression of this sequence in the form of a LacZ-LamB-LacZ fusion protein elicits the export interference phenotype. Deletion of the sequence from an export-defective LamB protein eliminates the ability of this protein to interfere with the export of other secreted proteins. Together, these findings show that this sequence is both necessary and sufficient to cause export interference. Surprisingly, deletion of this sequence from an otherwise wild-type LamB protein does not cause the mutant LamB product to exhibit any obvious export defect. Based on our results, we propose that SecB interacts with both amino acids 320-380 of mature LamB and the LamB signal sequence during initiation of the export process.     

Identification and characterization of a SET/NAP protein encoded by a brain-specific gene, MB20

A new member of the NAP/SET gene family, named MB20, was isolated from a mouse brain cDNA library by virtue of its CAG trinucleotide repetitive sequence and a brain-specific gene expression pattern. The complementary DNA sequence predicted an open reading frame of 545 amino acids, with four copies of an 11-amino-acid direct repeat. The consensus sequence for these repeats, PKE-P--K-EE, is present in the largest subunit of murine neurofilament (NF-H). The MB20 protein sequence is homologous to nucleosome assembly proteins of several species, and its C-terminus is homologous to SET proteins. Immunoblot analysis revealed that MB20 protein is expressed in the brain. Transient transfection and immunofluorescence microscopy demonstrated that MB20 is distributed in the cytoplasm as well as in the nucleus. Deletion of the N-terminal end imparts the complete localization of MB20 protein to the nucleus. The ability of MB20 to bind histone proteins was analyzed by sucrose gradient sedimentation and by retention of histone proteins by immobilized MB20 protein. On the basis of its expression pattern, predicted sequence, and protein properties, we propose that MB20 plays a unique role in modulating nucleosome structure and gene expression during brain development.     

Protein U, a late-developmental spore coat protein of Myxococcus xanthus, is a secretory protein.

Protein U is a spore coat protein produced at the late stage of development of Myxococcus xanthus. This protein was isolated from developmental cells, and its amino-terminal sequence was determined. On the basis of this sequence, the gene for protein U (pru) was cloned and its DNA sequence was determined, revealing an open reading frame of 179 codons. The product from this open reading frame has a typical signal peptide of 25 amino acid residues at the amino terminal end, followed by protein U of 154 residues. This result indicates that protein U is produced as a secretory precursor, pro-protein U, which is then secreted across the membrane to assemble on the spore surface. This is in sharp contrast to protein S, a major spore coat protein produced early in development, which has no signal peptide, indicating that there are two distinct pathways for trafficking of spore coat proteins during the differentiation of M. xanthus.     

Isolation and DNA sequence of a gene encoding alpha-trichosanthin, a type I ribosome-inactivating protein

alpha-Trichosanthin (alpha-TCS) is a ribosome-inactivating protein that has recently been shown to inhibit the replication of human immunodeficiency virus. We have isolated a gene encoding alpha-TCS and have determined its DNA sequence. The data indicate that alpha-TCS is synthesized as a preproprotein consisting of 289 amino acids, the first 23 residues of which comprise a putative secretory signal peptide. The last 19 residues comprise a carboxyl extension that has not been reported to be associated with the mature protein and that may be processed in the endoplasmic reticulum or Golgi apparatus of cells producing alpha-TCS. The mature protein consists of 247 amino acids. The sequence predicted by translation of the DNA sequence agrees with and confirms the primary sequence determined recently on the protein. The molecular clone for alpha-TCS will facilitate directed mutational analyses that may provide information on how this peptide, and other ribosome-inactivating proteins, function. These studies may also lead to the development of therapeutic agents with altered activities and/or improved properties for in vivo use.     

Characterization of a gene which encodes a mannosyltransferase homolog of Paracoccidioides brasiliensis

We screened an expression library of the yeast form of Paracoccidioides brasiliensis with a pool of human sera that was pre-adsorbed with mycelium, from patients with paracoccidioidomycosis (PCM). A sequence (PbYmnt) was obtained and characterized. A genomic clone was obtained by PCR of P. brasiliensis total DNA. The sequence contained a single open reading frame (ORF) encoding a protein of 357 amino acid residues, with a molecular mass of 39.78 kDa. The deduced amino acid sequence exhibited identity to mannosyl- and glycosyltransferases from several sources. A DXD motif was present in the translated gene and this sequence is characteristic of the glycosyltransferases. Hydropathy analysis revealed a single transmembrane region near the amino terminus of the molecule that suggested a type II membrane protein. The PbYmnt was expressed preferentially in the yeast parasitic phase. The accession number of the nucleotide sequence of PbYmnt and its flanking regions is AF374353. A recombinant protein was generated in Escherichia coli. Our data suggest that PbYmnt encodes one member of a glycosyltransferase family of proteins and that our strategy was useful in the isolation of differentially expressed genes.     

Molecular cloning and characterization of a cDNA encoding a novel cuticle protein from the Chinese oak Silkmoth, Antheraea pernyi.

In our research to identify gene involved in the cuticle protein, we cloned a novel cuticle protein gene, ApCP13, from the Chinese oak silkmoth, Antheraea pernyi, larvae cDNA library. The ApCP13 gene encodes a 120 amino acid polypeptide with a predicted molecular mass of 13 kDa and a pI of 4.01, and is intron-less gene. The ApCP13 contained a type-specific consensus sequence identifiable in other insect cuticle proteins and the deduced amino acid sequence of the ApCP13 cDNA is most homologous to another wild silkmoth, A. yamamai CP12 (86% protein sequence identity), followed by Bombyx mori LCP18 (35% protein sequence identity). Northern blot analysis revealed that the ApCP13 showed the epidermis-specific expression. This is the first report of cuticle protein gene in the wild silkmoth, A. pernyi.     

Characterization of a cDNA encoding a protein involved in formation of the skeleton during development of the sea urchin Lytechinus pictus.

In order to investigate the role of proteins in the formation of mineralized tissues during development, we have isolated a cDNA that encodes a protein that is a component of the organic matrix of the skeletal spicule of the sea urchin, Lytechinus pictus. The expression of the RNA encoding this protein is regulated over development and is localized to the descendents of the micromere lineage. Comparison of the sequence of this cDNA to homologous cDNAs from other species of urchin reveal that the protein is basic and contains three conserved structural motifs: a signal peptide, a proline-rich region, and an unusual region composed of a series of direct repeats. Studies on the protein encoded by this cDNA confirm the predicted reading frame deduced from the nucleotide sequence and show that the protein is secreted and not glycosylated. Comparison of the amino acid sequence to databases reveal that the repeat domain is similar to proteins that form a unique beta-spiral supersecondary structure.