Beta-sheet secondary structure of an LDL receptor domain from complement factor I by consensus structure predictions and spectroscopy

Functional and Expressional Analyses Reveal the Distinct Role of Complement Factor I in Regulating Complement System Activation during GCRV Infection in Ctenopharyngodon idella

Complement factor I (CFI), a complement inhibitor, is well known for regulating the complement system activation by degrading complement component 3b (C3b) in animal serum, thus becoming involved in innate defense. Nevertheless, the functional mechanisms of CFI in the complement system and in host-pathogen interactions are far from being clarified in teleost fish. In the present study, we cloned and characterized the CFI gene, CiCFI, from grass carp (Ctenopharyngodon idella) and analyzed its function in degrading serum C3b and expression changes after grass carp reovirus (GCRV) infection. The open reading frame of CiCFI was found to be 2121 bp, encoding 706 amino acids with a molecular mass of 79.06 kDa. The pairwise alignments showed that CiCFI shared the highest identity (66.9%) with CFI from Carassius gibelio and the highest similarity (78.7%) with CFI from Danio rerio. The CiCFI protein was characterized by a conserved functional core Tryp_SPc domain with the catalytic triad and substrate binding sites. Phylogenetic analysis indicated that CiCFI and the homologs CFIs from other teleost fish formed a distinct evolutionary branch. Similar with the CFIs reported in mammals, the recombinant CiCFI protein could significantly reduce the C3b content in the serum, demonstrating the conserved function of CiCFI in the complement system in the grass carp. CiCFI mRNA and protein showed the highest expression level in the liver. After GCRV infection, the mRNA expressions of CiCFI were first down-regulated, then up-regulated, and then down-regulated to the initial level, while the protein expression levels maintained an overall downward trend to the late stage of infection in the liver of grass carps. Unexpectedly, the protein levels of CiCFI were also continuously down-regulated in the serum of grass carps during GCRV infection, while the content of serum C3b proteins first increases and then returns to the initial level, suggesting a distinct role of CiCFI in regulating complement activation and fish-virus interaction. Combining our previous results that complement factor D, a complement enhancer, shows continuously up-regulated expression levels in grass carps during GCRV infection, and this study may provide the further essential data for the full picture of complex complement regulation mechanism mediated by Df and CFI of the grass carp during pathogen infection.

Insect vitellogenin/lipophorin receptors: molecular structures, role in oogenesis, and regulatory mechanisms

Insect vitellogenin and lipophorin receptors (VgRs/LpRs) belong to the low-density lipoprotein receptor (LDLR) gene superfamily and play a critical role in oocyte development by mediating endocytosis of the major yolk protein precursors Vg and Lp, respectively. Precursor Vg and Lp are synthesized, in the majority of insects, extraovarially in the fat body and are internalized by competent oocytes through membrane-bound receptors (i.e., VgRs and LpRs, respectively). Structural analysis reveals that insect VgRs/LpRs and all other LDLR family receptors share a group of five structural domains: clusters of cysteine-rich repeats constituting the ligand-binding domain (LBD), epidermal growth factor (EGF)-precursor homology domain that mediates the acid-dependent dissociation of ligands, an O-linked sugar domain of unknown function, a transmembrane domain anchoring the receptor in the plasma membrane, and a cytoplasmic domain that mediates the clustering of the receptor into the coated pits. The sequence analysis indicates that insect VgRs harbor two LBDs with five repeats in the first and eight repeats in the second domain as compared to LpRs which have a single 8-repeat LBD. Moreover, the cytoplasmic domain of all insect VgRs contains a LI internalization signal instead of the NPXY motif found in LpRs and in the majority of other LDLR family receptors. The exception is that of Solenopsis invicta VgR, which also contains an NPXY motif in addition to LI signal. Cockroach VgRs still harbor another motif, NPTF, which is also believed to be a functional internalization signal. The expression studies clearly demonstrate that insect VgRs are ovary-bound receptors of the LDLR family as compared to LpRs, which are transcribed in a wide range of tissues including ovary, fat body, midgut, brain, testis, Malpighian tubules, and muscles. VgR/LpR mRNA and the protein were detected in the germarium, suggesting that the genes involved in receptor-endocytotic machinery are specifically expressed long before they are functionally required.

PROTEUS2: a web server for comprehensive protein structure prediction and structure-based annotation

PROTEUS2 is a web server designed to support comprehensive protein structure prediction and structure-based annotation. PROTEUS2 accepts either single sequences (for directed studies) or multiple sequences (for whole proteome annotation) and predicts the secondary and, if possible, tertiary structure of the query protein(s). Unlike most other tools or servers, PROTEUS2 bundles signal peptide identification, transmembrane helix prediction, transmembrane β-strand prediction, secondary structure prediction (for soluble proteins) and homology modeling (i.e. 3D structure generation) into a single prediction pipeline. Using a combination of progressive multi-sequence alignment, structure-based mapping, hidden Markov models, multi-component neural nets and up-to-date databases of known secondary structure assignments, PROTEUS is able to achieve among the highest reported levels of predictive accuracy for signal peptides (Q2 = 94%), membrane spanning helices (Q2 = 87%) and secondary structure (Q3 score of 81.3%). PROTEUS2's homology modeling services also provide high quality 3D models that compare favorably with those generated by SWISS-MODEL and 3D JigSaw (within 0.2 Å RMSD). The average PROTEUS2 prediction takes ∼3 min per query sequence. The PROTEUS2 server along with source code for many of its modules is accessible a http://wishart.biology.ualberta.ca/proteus2.

Improving the accuracy of protein secondary structure prediction using structural alignment

Background

The accuracy of protein secondary structure prediction has steadily improved over the past 30 years. Now many secondary structure prediction methods routinely achieve an accuracy (Q3) of about 75%. We believe this accuracy could be further improved by including structure (as opposed to sequence) database comparisons as part of the prediction process. Indeed, given the large size of the Protein Data Bank (>35,000 sequences), the probability of a newly identified sequence having a structural homologue is actually quite high.

Results

We have developed a method that performs structure-based sequence alignments as part of the secondary structure prediction process. By mapping the structure of a known homologue (sequence ID >25%) onto the query protein's sequence, it is possible to predict at least a portion of that query protein's secondary structure. By integrating this structural alignment approach with conventional (sequence-based) secondary structure methods and then combining it with a "jury-of-experts" system to generate a consensus result, it is possible to attain very high prediction accuracy. Using a sequence-unique test set of 1644 proteins from EVA, this new method achieves an average Q3 score of 81.3%. Extensive testing indicates this is approximately 4–5% better than any other method currently available. Assessments using non sequence-unique test sets (typical of those used in proteome annotation or structural genomics) indicate that this new method can achieve a Q3 score approaching 88%.

Conclusion

By using both sequence and structure databases and by exploiting the latest techniques in machine learning it is possible to routinely predict protein secondary structure with an accuracy well above 80%. A program and web server, called PROTEUS, that performs these secondary structure predictions is accessible at . For high throughput or batch sequence analyses, the PROTEUS programs, databases (and server) can be downloaded and run locally.

Molecular cloning and characterization of LR3, a novel LDL receptor family protein with mitogenic activity

We report molecular cloning and initial functional characterization of a novel member of the low density lipoprotein receptor (LDLR) gene family. The cDNA was isolated from a human osteoblast cDNA library and encoded a 1,615 amino acids protein designated as LR3. It has, in the extracellular region, a cluster of three LDLR ligand binding repeats at a juxtamembrane position and four EGF precursor homology domains separated by YWTD spacer repeats. The entire ectodomain shares the same modular organization with the middle portion of the extracellular regions of two LDLR family members, LDLR-related protein (LRP), and gp330/megalin. LR3 mRNA was expressed in most of the adult and fetal tissues examined. The highest expression level was seen in aorta. In human osteosarcoma cells examined, LR3 mRNA was highly enriched in TE85 cells, moderately expressed in MG63 cells and primary human osteoblasts, and undetectable in SaOS-2 cells. NIH 3T3 cells transfected with either full length LR3 or its ectodomain showed significantly increased proliferation, whereas transfection of intracellular domain had no proliferative effect. We predict that LR3 is a multi-functional protein with potential mitogenic activity.

Possible arrangement of the five domains in human complement factor I as determined by a combination of X-ray and neutron scattering and homology modeling

Human factor I is a multidomain plasma serine protease with one factor I-membrane attack complex (FIMAC) domain, one CD5 domain, two low-density lipoprotein receptor (LDLr) domains, and one serine protease (SP) domain and is essential for the regulation of complement. The domain arrangement in factor I was determined by X-ray and neutron scattering on serum-derived human factor I (sFI) and recombinant insect cell factor I (rFI). While the radii of gyration of both were the same at 4.05 nm and both had overall lengths of 14 nm, the cross-sectional radii of gyration were different at 1.70 nm for sFI and 1.57 nm for rFI. This difference was attributed to their different means of glycosylation which is complex-type for sFI and high-mannose-type for rFI. Homology models were constructed for the FIMAC, LDLr, and SP domains of factor I using related crystal structures, and CD5 was represented as a globular protein by referencing its electron microscopy dimensions. In these models, 38 of the 40 Cys residues in factor I were predicted to form internal disulfide bridges. The two remaining Cys residues at the N terminus of the FIMAC domain and at the center of the first LDLr domain were potentially not bridged. It was postulated that, if these two Cys residues were bridged to each other, the FIMAC, CD5, and LDLr-1 domains would form a compact triangular arrangement. This hypothesis was tested by automated scattering curve fit searches based on 9600 bilobal models, setting the FIMAC, CD5, and LDLr-1 domains as one lobe and the large SP domain as the other lobe. The searches gave a single small family of similar structures with a separation of 5.9 nm between the centers of the lobes which gave similar good X-ray and neutron fits for both sFI and rFI, despite the different glycosylations of sFI and rFI. These best-fit structures for factor I showed that this domain model is plausible, and suggested that the SP and the CD5 and LDLr-1 domains may present exposed surfaces in factor I whose roles are to interact separately with its substrates C3b and C4b and with cofactor proteins.

Analogy and solution scattering modelling: new structural strategies for the multidomain proteins of complement, cartilage and the immunoglobulin superfamily

Many immunologically relevant proteins possess multidomain structures. Molecular structures both at the level of the individual domain and that of the intact protein are required for a full appreciation of function and control. Two recently developed structural approaches are reviewed here. Analogy modelling methods are based on the current understanding of many protein structures, and make possible the identification of folds for superfamilies of unknown structures. An integrated multidisciplinary predictive approach has been successfully applied to the von Willebrand factor type A, proteoglycan tandem repeat and factor I/membrane attack complex domains. The available experimental and predictive evidence is assembled in order to identify a known three-dimensional structure related to the unknown one of interest. Neutron and X-ray scattering curve modelling provides information on the full multidomain structure in solution. As scattering curves can be calculated from known atomic structures, the present availability of structures for many domains in conjunction with tight constraints based on these structures and the covalent connections between them results in a small family of allowed best-fit structures for a given scattering curve. The curve-fit procedure can be automated, and whole multidomain structures can be determined to a positional precision of the order of 0.2-1 nm. Such models are informative on the steric accessibility of each domain and their functional activity, and this is illustrated for antibody, cell-surface and complement proteins.

Human complement factor I: its expression by insect cells and its biochemical and structural characterisation

Factor I is a five-domain plasma serine protease which is essential for the regulation of the complement system. In order to express this, the factor I coding sequence was cloned into a recombinant baculovirus system, which was used to infect Trichoplusia ni cells. Using the native factor I leader sequence, recombinant factor I (rFI) was secreted into the culture medium. Purified rFI was recognised by polyclonal antisera and by the factor I-specific monoclonal antibody MRC-OX21. SDS PAGE showed that rFI was processed into two chains with molecular weights of 48,000 and 36,000. Amino acid sequence analysis showed that the N-terminal sequences of the rFI chains were the same as those of serum-derived factor I (sFI), confirming that processing was correct. Since both molecular weights were less than those observed for sFI, this is attributed to the replacement of complex-type oligosaccharides by high mannose ones in rFI. C3(NH,) cleavage assays showed that rFI had 55% the activity of sFI. Circular dichroism and Fourier transform infrared spectroscopy showed that the protein folding of rFI and sFI were very similar. Both had a secondary structure low in alpha-helix and high in beta-sheet, as expected from crystal structure and multiple sequence alignment analyses. It is inferred that the reduced activity of rFI is attributable to its changed glycosylation. The availability of rFI and structures for the domains in factor I makes possible new approaches to determine the molecular basis of its interactions with factor H and C3b.

Ligand-binding domains in vitellogenin receptors and other LDL-receptor family members share a common ancestral ordering of cysteine-rich repeats

Insect vitellogenin and yolk protein receptors (VgR/YPR) are newly discovered members of the low-density lipoprotein receptor (LDLR) family, which is characterized by a highly conserved arrangement of repetitive modular elements homologous to functionally unrelated proteins. The insect VgR/YPRs are unique in having two clusters of complement-type cysteine-rich (class A) repeats or modules, with five modules in the first cluster and seven in the second cluster, unlike classical LDLRs which have a single seven-module cluster, vertebrate VgRs and very low density lipoprotein receptors (VLDLR) which have a single eight-module cluster, and LDLR-related proteins (LRPs) and megalins which have four clusters of 2-7, 8, 10, and 11 modules. Alignment of clusters across subfamilies by conventional alignment programs is problematic because of the repetitive nature of the component modules which may have undergone rearrangements, duplications, and deletions during evolution. To circumvent this problem, we "fingerprinted" each class A module in the different clusters by identifying those amino acids that are both relatively conserved and relatively unique within the cluster. Intercluster reciprocal comparisons of fingerprints and aligned sequences allowed us to distinguish four cohorts of modules reflecting shared recent ancestry. All but two of the 57 modules examined could be assigned to one of these four cohorts designated A, B, C, and D. Alignment of clusters based on modular cohorts revealed that all clusters are derived from a single primordial cluster of at least seven modules with a consensus arrangement of CDCADBC. All extant clusters examined are consistent with this consensus, though none matches it perfectly. This analysis also revealed that the eight-module clusters in vertebrate VgRs, insect VgR/YPRs, and LRP/megalins are not directly homologous with one another. Assignment of modules to cohorts permitted us to properly align 32 class A clusters from all four LDLR subfamilies for phylogenetic analysis. The results revealed that smaller one-cluster and two-cluster members of the family did not originate from the breakup of a large two-cluster or four-cluster receptor. Similarly, the LRP/megalins did not arise from the duplication of a two-cluster insect VgR/YPR-like progenitor. Rather, it appears that the multicluster receptors were independently constructed from the same single-cluster ancestor.