|Year : 2021 | Volume
| Issue : 1 | Page : 7-15
Variability in linear polypeptide stabilizes proteoglycan than zinc finger protein in vascular smooth muscle cells: An In Silico approach
Karthikeyan D Rajamani
V ClinBio Labs Private Limited, Central Research Facility, Sri Ramachandra Institute of Higher Education and Research; Department of Environmental Health Engineering, Sri Ramachandra Institute of Higher Education and Research, Porur-600 116, Chennai, Tamil Nadu, India
|Date of Submission||01-Nov-2020|
|Date of Acceptance||24-Dec-2020|
|Date of Web Publication||13-Mar-2021|
Dr. Karthikeyan D Rajamani
ICMR Center for Air Quality, Climate and Health, Department of Environmental Health Engineering, Sri Ramachandra Institute of Higher Education and Research, Porur, Chennai - 600 116, Tamil Nadu
Source of Support: None, Conflict of Interest: None
Background: Structural and physicochemical topologies of proteins play a considerable role in differentiating the functional properties of the biological system. We aimed to study the physicochemical similarities, structural and functional differences of versican (VCAN) and early growth response (Egr) proteins involved in vascular injuries. Methods: For the primary structure prediction, the proteomic tools Expasy's Protparam is used, likewise, for secondary structure and content prediction SOPM and SOPMA tool is used. The transmembrane regions in VCAN and EGR proteins are predicted through SOSUI (Classification and Secondary Structure Prediction of Membrane Proteins) server. The CYSREC tool is used to identify the presence of disulphide bonds in all the VCAN and EGR proteins, additionally through homology modelling the disulphide bonds are visualized and structure of the modelled proteins are validated through Rampage (Ramachandran plot), ProQ (Protein Quality Server) and ProSA (Protein Structure Analysis) server. Results: VCAN and Egr proteins resemble hydrophilic in nature, similarly negative score of the grand average of hydropathicity index confirms hydrophilic nature. The maximum molecular weight for VCAN is observed as 39265 and 61623 Dalton for EGR protein. VCAN proteins showed a higher level of basic residues except Q86W61, while all the Egr proteins were acidic residues. The extinction coefficient (EC) has unique absorbance at 280 nm wavelength. Based on the aliphatic index (AI ≥ 45) and instability index (II ≥ 40) most of the VCAN and Egr proteins were unstable. The Classification and Secondary Structure Prediction of Membrane Proteins server classifies all Egr and few VCAN and proteins are soluble nature. Secondary structure content prediction and SOPM server show most of the VCAN proteins are beta sheets and many Egr proteins are alpha-helical, while few with mixed structures. Besides these differences, the VCAN protein stability was identified by most probable disulfide (SS) bridges using CYS_REC tool and confirmed by homology modeling in tertiary structure. Whereas the probable disulfide bonds in Egr proteins were not identified. Conclusion: The findings with these functional and structural properties will add an extra room in understanding their dual role.
Keywords: Disulfide bonds, early growth response, in silico, smooth muscle cells, versican
|How to cite this article:|
Rajamani K. Variability in linear polypeptide stabilizes proteoglycan than zinc finger protein in vascular smooth muscle cells: An In Silico approach. Biomed Biotechnol Res J 2021;5:7-15
|How to cite this URL:|
Rajamani K. Variability in linear polypeptide stabilizes proteoglycan than zinc finger protein in vascular smooth muscle cells: An In Silico approach. Biomed Biotechnol Res J [serial online] 2021 [cited 2021 Jun 22];5:7-15. Available from: https://www.bmbtrj.org/text.asp?2021/5/1/7/311092
| Introduction|| |
Vascular smooth muscle cells (VSMCs) occupy a highly specialized role in cardiovascular pathologies such as restenosis, atherosclerosis, and allograft vasculopathy., The extracellular matrix (ECM) is also equally involved in cellular rigidity and smooth muscle cells (SMC) mobility during the pathological conditions., While there is a different class of ECM involved in SMC replication, the role of proteoglycans found to play a major role on it. The proteoglycans have a characteristic distribution pattern such as aggrecan found in cartilages, nerocan, and brevican was more prominent in the central nervous system, and versican (VCAN) widely accumulated in soft tissues.,, VCAN is a chondroitin sulfate proteoglycan being principle gene have been upregulated during vascular injury through their ability to bind with lipoproteins in the arterial wall. The structural pattern of VCAN gene and protein followed by the amino terminal domain binds hyaluronan to the carboxy-terminal globular domain; it also contains lectin-like domain adjacent to epidermal growth factor domains.,
In addition to the structural and functional properties of VCAN in SMCs, early growth response (Egr) is another ECM macromolecule widely expressed in SMCs. Similar to VCAN, the Egr-an extracellular stimuli play a critical role in the pathogenesis of atherosclerosis, coronary artery disease, and allograft vasculopathy. Egr is a zinc finger transcriptional factor and a product of immediate early gene being upregulated after vascular injury localized to macrophages in the internal margins of atherosclerotic lesions.,, To date, the activation of Egr is considered master regulator, as it controls the expression of various genes facilitating cell proliferation and neointimal hyperplasia. However, there are many factors which known to induce Egr-1, like cytokines, shear stress, extracellular stimuli consists of transactivation and repression domains, through DNA binding zinc fingers that recognize GC rich sequences present in target gene promoters.
Although there are number of studies are being reported based on in vitro and in vivo studies to elucidate their involvement in vascular SMCs, the structural properties of both of these transcriptional proteins have not been studied in detail. It is important to know the arrangement of linear polypeptides, structural alignments, and bonding patterns that define the stability of these proteins. The arrangement of each amino acid residues, hydrophilic, and hydrophobic profile may confer functional difference in molecular mechanisms underlying in SMC proliferation. Since there is a variation in the linear arrangement of residues that determines the transcriptional control and regulation of ribosomes in understanding the causes of diseases. Hence, considering the functional similarities of VCAN and Egr in vascular SMCs, we aimed to study the physicochemical, structural, and functional differences in 16 sequences (8 VCAN + 8 Egr) from Homosapiens through proteomic tools.
| Methods|| |
Protein sequence retrieval and selection
Human proteins VCAN and Egr are retrieved from the UniProtKB/Swiss-Prot release 57.0 (http://www.expasy.org/sprot) protein sequence database. The Swiss-Prot database is scanned for the keyword “Versican” and “Early growth response” the protein of Homo sapiens (Human) alone were chosen, and all these protein sequences were downloaded in FASTA format. Eight sequences in each protein were used for this study. The details of VCAN and Egr protein for comparative physico-chemical, structural, and functional characterization are given in [Supplementary Table 1].
Proteomic tools and servers
The amino acid composition of human VCAN and EGR proteins was computed using the tool BioEdit 5.0.9. Percentages of hydrophobic and hydrophilic residues were computed using the primary structural data. The physicochemical parameters such as theoretical isoelectric point, molecular weight, extinction coefficient (EC), half-life, instability index,, aliphatic index, antigenic site, and grand average hydrophathy (GRAVY) values were computed using the Expasy's ProtParam prediction server. The correlation between the number of acidic and basic residues are calculated in this server. The tools SOPM and SOPMA were used for the secondary structure prediction., Secondary Structural Content Prediction (http://coot.embl.de/SSCP/) server is used for the computation of percentages of α-helical, β-strand and coiled regions and secondary structure class identification. The SOSUI server performed the identification of transmembrane regions in VCAN and Egr human proteins.
Prediction of disulfide bridges-SS bound cysteines
Disulfide bridges and cysteine residues in VCAN and Egr proteins were predicted by two different methods. In the first method, the presence of disulfide bonds (SS) with a total number of cysteines were identified through the protein sequences (FASTA format) submitted in the CYS_REC tool. In the second method, the SS bonds were identified through visualization of three-dimensional (3D) structure of protein; the 3D structure of six VCAN proteins was predicted by homology modeling through Esypred server and visualized in RasMol tool.
Structure analysis and validation
For comparative modeling, the six VCAN protein sequences are selected based on disulfide bridges predicted in the CYS_REC tool. The templates for 3D structures were predicted in Protein Data Bank (www.rscb.org) through BLASTP analysis with the expectation value of 0.01 for D6RGZ6, E9PF17, Q86W61, Q59FG9, and O14594 proteins. The modeled 3D structures were validated using servers Rampage (Ramachandran plot), Protein Quality (ProQ) server, and protein structure analysis.
| Results and Discussion|| |
Primary structure analysis-amino acid transporters in endothelial cells
The linear arrangement of amino acids in VCAN and Egr protein showed that all the proteins are hydrophilic in nature [Table 1]. The amino acids asparagine (N), lysine (K), aspartic acid (D), glutamine (Q), histidine (H), arginine ® and glutamic acid (E) determine the hydrophilic property of proteins. The maximum molecular weight of VCAN protein observed as 39264.6 and minimum of 99,118 Dalton; similarly for Egr 61,623 calculated as maximum and minimum of 24,195 Dalton. The negative score of the GRAVY index profile confirms the hydrophilic nature of VCAN and Egr protein. Though, both the proteins are hydrophilic in nature, it has distinct features of acidic and basic amino acid residues. The examined VCAN proteins showed a higher level of basic residues, except Q86W61, since it shows a high content of acidic residue. In the case of EGR proteins, every protein sequence has a higher level of basic amino acid residues. The amino acids arginine®, histidine (H), and lysine (K) attached with an amino group are called basic hydrophilic; higher level of aspartic acid (D) and glutamic acid (E) attached with carboxyl group are known as acidic hydrophilic. These hydrophilic signaling molecules travel freely through the bloodstream, but require membrane proteins to influence their target cells.
In general, the hydrophilic protein molecules are unable to cross the plasma membrane directly; instead, they bind to cell surface receptors (transmembrane) of proteins to generate one or more signals in the target cell. As reported earlier, several forms of hypercholesterolemia are due to genetic defects in the synthesis of abnormal low-density lipoprotein receptors. The outcome is an elevation in serum cholesterol levels and increases the chances for the development of atherosclerosis.
Physico-chemical biology of versican and early growth response proteins
The EC of VCAN and Egr proteins had unique absorbance at 280 nm wavelength; where VCAN protein EC ranges from 5200 to 294470 and EGR protein ranges from 8700 to 38,150 M−1 cm−1 with respect to the concentration of Cys, Trp, and Tyr. The protein absorbs ultraviolet at maximum 280 and 200 nm, the amino acid with aromatic rings are the primary reasons for the absorbance peak at 280 nm, particularly tryptophan (W). ProtParam tool classifies most of the VCAN and Egr human proteins as unstable on the basis of the aliphatic index (AI ≥45) and instability index (II ≥40) proteins in room temperature, except the VCAN protein Q86W61 as stable (II <40). The aliphatic side chains in amino acids alanine, isoleucine, leucine, and valine are responsible for high AI (77.4) than usual proteins. The proteins with half-life of 30 h have a greater instability index from 50 to 77. However, the electrostatic interactions are caused by the charged residues at physiologic pH (7.4), it includes positively charged lysine (K) and arginine®, as well as negatively charged aspartic acid (D) and glutamic acid (E) as given in [Table 2]. The importance of these charged residues was widely known for the characteristic feature of protein folding and stability because these positive and negatively charged residues have differential effects on cell membranes. As reported, Asp and Glu residues cause re-positioning of the C-terminal end of the transmembrane helix; while Arg and Lys residues have little effect on the C-terminal end.
Secondary structural properties of versican and early growth response
The secondary structure is determined through protein folding includes hydrogen bonding, disulfide bonds, and the presence of proline residues in a protein. Secondary structure content prediction server classified most of the VCAN proteins as beta-sheet structure, except O14594 and Q96GW7. Similarly, the Egr proteins were classified as either mixed structure or alpha, while two proteins are irregular in structure [Table 3]. In alpha and beta structures, hydrogen bonding interactions are involved. Considering the secondary structure of proteins, the frequency of proline residues with higher helix occurs frequently in Egr than VCAN proteins. The information from the secondary structure of VCAN and Egr proteins is instrumental in understanding their role in tertiary structure.
|Table 3: Deviation in secondary structure content of versican and Egr proteins|
Click here to view
In accordance with previous reports, the Egr proteins with helix structure showed lower chances of crosslinks, may be due to steric hindrances involved in the formation of disulfides between the cysteine residues. In such case, the occurrence of disulfide bridges is high in VCAN protein with beta-sheets, which stabilize and highlight their preferences on stabilization of VCAN protein in SMC.
Prediction of transmembrane regions for versican and early growth response proteins
Based on the physicochemical characteristics, the protein sequences are subjected to identify the transmembrane region in proteins. The SOSUI server classifies the VCAN and all EGR proteins as soluble, except few VCAN proteins P13611, D6RGZ6, Q861N61, and O14594. The transmembrane or insoluble region in all these proteins are classified as primary and secondary [Table 4]; the hydrophobic residues, polar residues and positively charged residues are plotted in transmembrane regions using Pepwheel tool [Figure 1]. The membrane topology was found to have more leucine, isoleucine, and methionine residues. Whereas all Egr were soluble proteins as influenced by the acidic property (Asp and Glu) with a very high instability index.
|Figure 1: (a) P13611-primary, (b) D6RGZ6-primary, (c) Q86W61-primary, (d) O14594-Primary, and (e) O14594-Secondary. Hydrophobic residues are represented in blue square (V, L, M, I) and violet color letters (A, C, F, G, P, W); polar residues-in red diamonds (E, N, Q, S, T); positively charged residues-octagons (H, K, R) identified through Pepwheel tool|
Click here to view
|Table 4: Difference in membrane topology between versican and Egr proteins|
Click here to view
Disulfide bonds as a switch for versican protein stabilization
Considering the importance of disulfide bonds in structural stability, folding, and function of proteins, we analyzed the presence of disulfide bridges in VCAN and Egr proteins by two different methods. First, the presence of total cysteine residues and probable bonding patterns were identified using the protein sequences submitted in the CYS_REC tool. The presence of cysteine residues in submitted sequences of VCAN and Egr proteins were identified. Where, the most probable bonding (SS) patterns were not identified in all Egr proteins.
In the second method, the structure of all VCAN and Egr proteins was predicted by homology modeling. The SS bonds predicted in the three-dimensional structure of six VCAN proteins are shown in [Figure 2]a,[Figure 2]b,[Figure 2]c,[Figure 2]d,[Figure 2]f, the presence of unbounded cysteine residues in Egr proteins are identified, as shown in [Figure 3]. The consequence of DS bonds add stability to proteins; when there is incorrect disulfide linkages between the conserved sequence, then the protein cannot find its native state; rather it will aggregate. The protein von Willebrand factor adheres to collagen, cross-link with platelets during arterial bleeding, it has force-sensing domain which is unfolded through vascular injury. This results increased adherence to the neighboring domains, and to collagen. As shown in [Table 5], the topology of SS bonds residues are same with all these proteins, which indicate similarity in their kinetic roles. Although the topologies between 2 and 3 SS bonds are same, the position of amino acid residues, which they interconvert between the intermediates, can accumulate the changes in protein function. Because the stabilization of VCAN protein may be considered to be entropy through destabilization that contributes through interactions with hydrophobic residues. The identified SS thiol reside more on negatively charged (Aspartate and glutamate) binding repeats in the C-terminal sequence with different arrangements of Thr-Lys-Arg-Leu-Tyr-Pro residues.
|Figure 2: Homology modeling and three dimensional structure of versican proteins predicted using Esypred server. Versican proteins with disulfide bonds predicted in (a) P13611, (b) D6RGZ6, (c) E9PF17, (d) Q86W61, (e) Q59FG9, (f) O14594|
Click here to view
|Figure 3: Three dimensional structure of early growth response protein with cysteine residues|
Click here to view
|Table 5: Position of unbound cysteine residues and disulfide bond (SS) residues predicted through CYS_REC tool and 3D structure viewed by RasMol tool|
Click here to view
It becomes necessary to postulate the modeled protein structures to understand the number of allowable initiating events in the folding process. The modeled proteins are validated through the Ramachandran Plot and ProQ server, as shown in [Figure 4]. It shows the majority of the VCAN protein residues about 75%–95% in favored regions, which indicates the accuracy of predicted structure [Table 6]. This referred to as nucleations, most likely occur in polypeptide chain, which confirm equilibrium between linear arrangements of residues in VCAN protein.
|Figure 4: Validation of three dimensional structure of versican proteins through Ramachandran Plot.(A) P13611, E9PF17 and Q86W61, (B) D6RGZ6 and Q59FG9, (C) O14594; and Protein Quality Server-(a) P13611, E9PF17 and Q86W61, (b) D6RGZ6 and Q59FG9, (c) O14594|
Click here to view
The level of structural and functional similarity among VCAN and EGR proteins overwhelm their involvement in the population of SMCs. But, in some way, it differs due to the presence of one or more disulfide bonds in VCAN protein than EGR protein. This necessitates the comprehensive study on the occurrence of DS bond and its variations in transcriptional proteins (VCAN and EGR). In another way, ligand-bound cysteines are more similar to DS bonded cysteine than free cysteines. It is believed that DS bonds are conserved proteins and the connectivity pattern with other amino acid residues may be used as a diagnostic tool. The disulfide bond is the most common covalent bond link between amino acids formed due to reduction-oxidation between thiol groups.
Importance of SS bonds in antigen selection
The human VCAN protein found to have disulfide bridges with few unbound cysteines residues. Mostly, amino-terminal cysteines residues located in the cytoplasmic domain of glycoprotein, while the next cysteine located to the transmembrane domain is responsible for the intermolecular disulfide bonds. As shown in [Table 7], the antigenic site for each VCAN protein identified close to disulfide bonds, this shows the interaction of immunoglobulin chains (heavy and light) occur through disulfide bridges, which confer greater stability to the immunoglobulin. The distribution of different disulfide bond isoforms is dependent on the type of light chains (κ, λ). However, there are few unpaired or unbound cysteine residues in VCAN and Egr proteins may have a detrimental effect during pathological conditions.
|Table 7: Resemblance of disulfide bridges and antigenic position in versican proteins|
Click here to view
| Conclusion|| |
The findings are interpreted in terms of the linear arrangement of amino acids, hydrogen bonding, and disulfide bonding patterns connecting their involvement in VSMCs. Few insoluble VCAN protein tends to have more hydrophobic stretches, lower glutamine content, fewer negatively charged residues with a higher percentage of aromatic amino acid than soluble EGR proteins. Identified functionally important charged residues (+ve and-ve) at either carboxy or amino-terminal end are important consequences for ribosomal occupancy. The role of disulfide bridges and their link to the antigenic site is more likely to elucidate the importance during normal and pathological conditions. The information about secondary structure (α-helix and β-sheet) is instrumental in understanding their role in the tertiary and quaternary structure of proteins. The consistency of these findings with proteomic analysis will be fruitful to understand their pathological features while working with several experimental studies that outline the importance of these proteins add an extra room in understanding their dual role in SMC.
It is the concept of understanding two different biomolecules that play a similar role but has distinct structural differences in SMC. The linear arrangement of amino acid residues and their transport in blood vessels has provided essential features in endothelial cells. The findings from this study will be an overlay for further studies in molecular modeling and targeted drug design in inhibiting the role of VCAN and Egr for the mechanisms that has been played. In addition, it makes possible to understand the pathological features when more data are generated on similar patterns of disulfide bridges, which directs toward the development of biomarkers and diagnostic kits.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Björkerud S. Agglomeration to nodules modulates human arterial smooth muscle cells to distinct postinjury phenotype via foam cell transition. Am J Pathol 1987;127:485-98.
Orlandi A, Bochaton-Piallat ML, Gabbiani G, Spagnoli LG. Aging, smooth muscle cells and vascular pathobiology: Implications for atherosclerosis. Atherosclerosis 2006;188:221-30.
Malhotra R, Tyson DW, Rosevear HM, Brosius FC. Hypoxia-inducible factor-1alpha is a critical mediator of hypoxia induced apoptosis in cardiac H9c2 and kidney epithelial HK-2 cells. BMC Cardiovasc Disord 2008;8:9.
Wang CC, Sharma G, Draznin B. Early growth response gene-1 expression in vascular smooth muscle cells: Effects of insulin and oxidant stress. Am J Hypertens 2006;19:366-72.
Isogai Z, Shinomura T, Yamakawa N, Takeuchi J, Tsuji T, Heinegård D, et al
. 2B1 antigen characteristically expressed on extracellular matrices of human malignant tumors is a large chondroitin sulfate proteoglycan, PG-M/versican. Cancer Res 1996;56:3902-8.
De Toledo OM, Marquezini MV, Bin JK, Pinheiro MD, Mora OA. Biochemical and cytochemical characterization of extracellular proteoglycans in the inner circular smooth muscle layer of dog small intestine. IUBMB Life 2002;54:19-25.
Farb A, Kolodgie FD, Hwang JY, Burke AP, Tefera K, Weber DK, et al
. Extracellular matrix changes in stented human coronary arteries. Circulation 2004;110:940-7.
Rahmani M, Read JT, Carthy JM, McDonald PC, Wong BW, Esfandiarei M, et al
. Regulation of the versican promoter by the β-catenin-T-cell factor complex in vascular smooth muscle cells. J Biol Chem 2005;280:13019-28.
Heydarkhan-Hagvall S, Esguerra M, Helenius G, Soderberg R, Johansson BR, Risberg B. Production of extracellular matrix components in tissue-engineered blood vessels. Tissue Eng 2006;12:831-42.
Nagai R, Shindo T, Manabe I, Suzuki T, Kurabayashi M. KLF5/BTEB2, a Kruppel-like zinc-finger type transcription factor, mediates both smooth muscle cell activation and cardiac hypertrophy. Adv Exp Med Biol 2003;538:57-65; discussion 66.
Khachigian LM, Lindner V, Williams AJ, Collins T. Egr-1-induced endothelial gene expression: A common theme in vascular injury. Science 1996;98:186-91.
Holm AM, Andersen CB, Haunsø S, Hansen PR. ACE-inhibition promotes apoptosis after balloon injury of rat carotid arteries. Cardiovasc Res 2000;45:777-82.
Gousseva N, Kugathasan K, Chesterman CN, Khachigian LM. Early growth response factor-1 mediates insulin-inducible vascular endothelial cell proliferation and regrowth after injury. J Cell Biochem 2001;81: 523-34.
Khachigian LM. Early growth response-1 in cardiovascular pathobiology. Circ Res 2006;271:1427-31.
Yamagata K, Kaufmann WE, Lanahan A, Papapavlou M, Barnes CA, Andreasson KI, et al
. Egr3/Pilot, a zinc finger transcription factor, is rapidly regulated by activity in brain neurons and colocalizes with Egr1/zif268. Learn Mem 1994;1:140-52.
Lipman DJ, Pearson WR. Rapid and sensitive protein similarity searches. Science 1985;227(4693):1435-41.
Gill SC, von Hippel PH. Calculation of protein extinction coefficients from amino acid sequence data. Anal Biochem 1989;182:319-26.
Bachmair A, Finley D, Varshavsky A. In vivo
half-life of a protein is a function of its amino-terminal residue. Science 1986;234:179-86.
Varshavsky A. The N-end rule pathway of protein degradation. Genes Cells 1997;2:13-28.
Guruprasad K, Reddy BV, Pandit MW. Correlation between stability of a protein and its dipeptide composition: A novel approach for predicting in vivo
stability of a protein from its primary sequence. Protein Eng Des Sel 1990;4:155-61.
Ikai A, Yanagita Y. A cross-linking study of apo-low density lipoprotein. J Biochem 1980;88:1359-64.
Kyte J, Doolittle RF. A simple method for displaying the hydropathic character of a protein. J Mol Biol 1982;157:105-32.
Geourjon C, Deléage G. Sopm: A self-optimized method for protein secondary structure prediction. Protein Eng Des Sel 1994;7:157-64.
Geourjon C, Deléage G. Sopma: Significant improvements in protein secondary structure prediction by consensus prediction from multiple alignments. Bioinformatics 1995;11:681-4.
Eisenhaber F, Frömmel C, Argos P. Prediction of secondary structural content of proteins from their amino acid composition alone. II. The paradox with secondary structural class. Proteins Struct Funct Genet 1996;25:157-68.
Hirokawa T, Boon-Chieng S, Mitaku S. SOSUI: Classification and secondary structure prediction system for membrane proteins. Bioinformatics 1998;14:378-9.
Lovell SC, Davis IW, Arendall WB 3rd
, de Bakker PI, Word JM, Prisant MG, et al
. Structure validation by Calpha geometry: Phi, psi and Cbeta deviation. Proteins 2003;50:437-50.
Cristobal S, Zemla A, Fischer D, Rychlewski L, Elofsson A. A study of quality measures for protein threading models. BMC Bioinform 2001;2:5.
Wiederstein M, Sippl MJ. ProSA-web: Interactive web service for the recognition of errors in three-dimensional structures of proteins. Nucleic Acids Res 2007;35:W407-410.
Alberts B, Johnson A, Lewis J. Molecular biology of the cell. In: The Self-Assembly and Dynamic Structure of Cytoskeletal Filaments. 4th
ed. New York: Garl Sci; 2002.
Monné M, Nilsson IM, Johansson M, Elmhed N, Von Heijne G. Positively and negatively charged residues have different effects on the position in the membrane of a model transmembrane helix. J Mol Biol 1998;284:1177-83.
Thangudu RR, Manoharan M, Srinivasan N, Cadet F, Sowdhamini R, Offmann B. Analysis on conservation of disulphide bonds and their structural features in homologous protein domain families. BMC Struct Biol 2008;08:55.
Pawar AP, DuBay KF, Zurdo J, Chiti F, Vendruscolo M, Dobson CM. Prediction of “aggregation-prone” and “aggregation- susceptible” regions in proteins associated with neurodegenerative diseases. J Mol Biol 2005;350:379-92.
Hogg PJ. Disulfide bonds as switches for protein function. Trends Biochem Sci 2003;28:210-14.
[Figure 1], [Figure 2], [Figure 3], [Figure 4]
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6], [Table 7]