PREDICTED STRUCTURE AND BINDING MOTIFS OF COLLAGEN α1(XI).

NIH Public Access Author Manuscript GSTF Int J Bioinforma Biotechnol. Author manuscript; available in PMC 2014 October 10.

NIH-PA Author Manuscript

Published in final edited form as: GSTF Int J Bioinforma Biotechnol. 2011 December ; 1(1): 43–48.

PREDICTED STRUCTURE AND BINDING MOTIFS OF COLLAGEN α1(XI) Owen M. McDougal1,2, Lisa R. Warner2,3,†, Chris Mallory1,2, and Julia Thom Oxford2,4,* 1Department

of Chemistry and Biochemistry, Boise State University, Boise, ID 83725

2Biomolecular

Research Center, Boise State University, Boise, ID 83725

3Department

of Materials Science and Engineering, Boise State University, Boise, ID 83725

4Department

of Biological Sciences, Boise State University, Boise, ID 83725

Abstract NIH-PA Author Manuscript NIH-PA Author Manuscript

The amino propeptide of collagen α1(XI) (NPP) has been shown to bind glycosaminoglycans and to form a dimer. While these are independent biochemical events, it is likely that dimerization facilitates the interaction with glycosaminoglycans or alternatively, that glycosaminoglycan interaction facilitates the formation of an NPP:NPP dimer. The computer program MODELLER was used to generate a homology model of the collagen α1(XI) NPP monomer using the crystal structure of the closely related noncollagenous-4 (NC4) domain of collagen α1(IX) (PDB:2UUR) as the template. Additionally, a dimer model of collagen α1(XI) NPP domain was created based upon the thrombospondin dimer template (PDB:1Z78). The structure of the dimer created in MODELLER was validated by comparison to a dimer model generated by docking two monomers of PDB:2UUR using ClusPro. Calculations of relative binding energy for the interaction between each collagen α1(XI) NPP model and glycosaminoglycans as ligands was performed using AutoDock4. Computational results support a higher affinity between heparan sulfate and a dimer compared to a monomer. These findings are supported by affinity chromatography experiments in which distinct monomer and dimer peaks were observed. Sequential point mutation studies of the putative binding site (147-KKKITK-152) indicated the importance of the basic lysine residue for binding to heparan sulfate. Two orders of magnitude change in binding affinity was predicted when comparing wild type to the mutation K152A. Experimental data supports the predicted change in affinity.

Keywords heparin; heparan sulfate; collagen; molecular interaction; glycosaminoglycan; protein

* Corresponding Author Telephone: 208-426-2395. [email protected]. †Present Addresses Lisa Warner, Department of Chemistry and Biochemistry, University of Colorado, Boulder, Boulder, CO 80309 [email protected]

Author Contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

McDougal et al.

Page 2

INTRODUCTION NIH-PA Author Manuscript

Collagen is a triple helical protein comprising approximately 25% of the protein contained in the human body. The triple helix of collagen is unique, due to its formation from three left-handed helical strands to compose the right-handed triple helix.1 The strands that make up this triple helix have the sequence Gly-Xxx-Yyy; where approximately 30% of the Xxx and Yyy are proline and hydroxyproline, respectively.2-4 To date, more than 27 different collagens have been reported in the literature, of which 16 have non-collagenous domains attached to the extended collagen triple helix.5,6 Collagens type V and XI are minor fibrillar collagens involved in regulating the nucleation, assembly, and ultimate diameter of collagen fibrils found ubiquitously.2,7 Collagens type V and XI form triple helical molecules composed of three different left-handed helical strands, where each of the alpha chains contain non-collagenous domains. The non-collagenous α1(XI) amino propeptide (NPP) of the gene product Col11a1 is found resident on the surface of collagen fibrils for an extended period of time in tissues.2-4


The minor fibrillar collagens of blood vessel walls include type V/XI heterotypic hybrid collagen molecules. Because the minor fibrillar collagens provide surface domains available for interaction, the NPP domain may play a role in the structural integrity of the vessel wall. Currently, a protein database structure file is not available for the NPP domain of collagen α1(XI), but a recently published structure for the NC4 domain of collagen α1(IX) is available that demonstrates remarkable sequence homology to the NPP domain of collagen α1(XI).8 The NPP domain of collagen α1(XI) belongs to a family of laminin, neurexin, sex hormone binding globulin (LNS) domains, in which a crystal structure of the thrombospondin-1 heparin binding domain has also recently appeared in the literature in a monomer and dimer form.9


Our previous affinity chromatography evidence supports the hypothesis that the NPP domain of collagen α1(XI) interacts with glycosaminoglycans such as heparan sulfate.10 This interaction is proposed to be significant in determining the thickness of the collagen fibril as it forms.6 These interactions occur at a glycosaminoglycan binding site within collagen α1(XI) consisting of 147-KKKITK-152.2 Our previous experimental evidence indicates the formation of an NPP α1(XI) dimer.11 Using a combination of homology modeling and protein-protein docking, a computational model for the collagen α1(XI) NPP domain was created. Molecular docking simulations were performed in AutoDock4 to calculate the energy of interaction between collagen α1(XI) NPP (monomer and dimer) with the heparan sulfate ligand.11,12 The AutoDock4 output includes the estimated free energy for binding that is calculated based upon predicted intermolecular forces of attraction, i.e. hydrophobic, electrostatic, and hydrogen bonding. In addition, a three dimensional visual depiction of bound ligand provides insight into binding paradigms consistent with experimental data.

GSTF Int J Bioinforma Biotechnol. Author manuscript; available in PMC 2014 October 10.

McDougal et al.

Page 3

MATERIALS AND METHODS NIH-PA Author Manuscript

Collagen α1(XI) NPP homology model monomer


The sequence for collagen α1(XI) NPP was submitted to BLAST against available PDB structures for template identification.13 BLAST returned two possible templates; thrombospondin-1 (PDB:1Z78) and the NC4 domain of collagen α1(IX) (PDB:2UUR).8,9,13 The sequence for each template was aligned independently to the collagen α1(XI) NPP domain, and then collaboratively using the ‘salign’ command in MODELLER.14 Homology models for the collagen α1(XI) NPP domain using both 2UUR and 1Z78 as templates were then created using the MODELLER method with disulfide bonding specified between Cys 25–Cys 207 and Cys 146–Cys 200.2,8,9,14 The returned models were evaluated using PROCHECK for areas of high energy, restricted points on the Ramachandran plot, and residue clashes.15,16 Loop rebuilding was performed using MODELLER. Corrections to amino acids occupying restricted areas of the Ramachandran plot and energy minimization were amended in Chimera.17 An RMSD was calculated for the homology model generated from each template (2UUR & 1Z78), using Chimera.8,9,17 The homology model created from 2UUR (HM1) was selected as the best model based on sequence alignment, query coverage, E value, and an RMSD of 0.785Å from template structure. Chimera was used to make the following single-point mutations into the collagen α1(XI) NPP domain putative binding site: K147A, K148A, K149A, and K152A.17 Collagen α1(XI) homology model dimer A dimer model of the collagen α1(XI) NPP was created with MODELLER and ClusPro using the thrombospondin-1 dimer (PDB:2ES3) as a template.9,14,18-21 The homology model was created by the same process as the monomer model with the exception that a repeated sequence of collagen α1(XI) NPP was used. Model evaluation was performed using PROCHECK, Verify3D and Ramachandran plot analysis.15,16 A second dimer model was created by submitting two identical homology model monomers to the ClusPro server.18-21 The overall lowest energy model was selected from the balanced interaction cluster. Heparin affinity chromatography


The recombinant protein derived from the NPP α1(XI) collagen sequences from rat was expressed in BL21 DE3 E. coli cells (Novagen, Madison, WI) using the pET11a bacterial expression vector (Stratagene, La Jolla, CA). Inclusion body proteins were solubilized in 50 mM sodium phosphate, pH 7.6 containing 6 M guanidine hydrochloride, 1 mM 2mercaptoethanol and 80 mM NaCl. Solubilized proteins were clarified by centrifugation at 18,000 × g, 4°C for 30 minutes and applied to a nickel-NTA-agarose (QIAGEN, Inc, Valencia, CA) column at a flow rate of 1 ml/minute for purification by chelation chromatography. Absorbance at 280 nm and conductivity were monitored throughout purification. Recombinant proteins were dialyzed against 0.05 M Tris-HCl, pH 7.5, containing 0.15 M NaCl. Samples were applied to a heparin-agarose (Sigma Chemicals, St. Louis, MO) column at low ionic strength. Protein was eluted from the column during a linear gradient of NaCl concentration to determine the NaCl concentration required for


McDougal et al.

Page 4

displacement of the protein. Absorbance at 280 nm and conductivity were monitored continuously throughout the experiment.


Docking Studies Docking studies between the collagen α1(XI) NPP monomer homology model, single-point putative binding site mutants, MODELLER dimer, and the ClusPro dimer with the glycosaminoglycan heparan sulfate (disaccharide, decasaccharide, and Arixtra, a low molecular weight heparin) as the ligand were performed using AutoDock4.12 The AutoDock standard conditions for a large run were used with a grid spacing of 0.379 Å. Binding energies were evaluated using WordPad, with residue interactions visualized in Chimera.17

RESULTS Homology model


BLAST search results indicated that the NC4 domain of collagen IX (PDB:2UUR) was the best template with an E value of 5.0 × 10-5.8,13 Alignment and model creation was performed using MODELLER with PDB:2UUR as the template. The RMSD between the computational model of the collagen α1(XI) NPP monomer and 2UUR was 0.785 Å (Figure 1A).8,14,17 Point mutations were introduced into the putative binding site sequentially using Chimera.17 Docking interactions between heparan sulfate disaccharide and collagen α1(XI) NPP monomer models indicated that Lys152 plays the largest role in binding as evidenced by the largest change in free energy between the wild type and the K152A mutant (Table 1). Binding Studies


Docking results for heparan sulfate and each dimer model supports the hypothesis that dimerization increases affinity for glycosaminoglycans relative to the protein monomer (see Table 2). Heparan sulfates chosen for this study include a heparan sulfate disaccharide, heparan sulfate decasaccharide, and Arixtra. The di- and decameric heparan sulfates were chosen, because it is this form that is found on proteoglycans that plays a role in cell adhesion, migration, proliferation and cellular differentiation. Arixtra is a low molecular weight heparin that was chosen to investigate the molecular interactions that may be relevant in the preventative treatment for clotting problems such as deep vein thrombosis. Heparin affinity chromatography studies Affinity chromatograms show two distinct peaks eluting from the heparin agarose column corresponding to the NaCl concentration required to disrupt the protein-heparin interaction (Figure 2). The right-hand peak shows an increased affinity for heparin, eluting at approximately 0.55 M NaCl, compared to the left-hand peak, which requires only 0.4 M NaCl to displace the protein from the column. The proportion of protein eluting at higher NaCl concentration increased with increasing recombinant NPP concentration within the concentration range of 0.25 mg/mL (Figure 2A) to 3.0 mg/mL (Figure 2E). All data are shown as an overlay in panel 2F of Figure 2.


McDougal et al.

Page 5

DISCUSSION NIH-PA Author Manuscript

Point mutation docking studies revealed that K148A had the lowest estimated binding energy, providing a preliminary hypothesis that K148 is the least critical residue involved in binding of heparan sulfate and heparin to collagen α1(XI) NPP protein. Furthermore, K152 appears to be the most important residue in the putative binding site involved in glycosaminoglycan interaction as evidenced by a predicted increase in binding energy, i.e. less negative value. Experimental results suggest that a greater affinity for glycosaminoglycans occurs upon dimerization or oligomerization of collagen α1(XI) NPP. Docking results for heparan sulfate disaccharide, decasaccharide, and Arixtra support the experimental results of increasing affinity for glycosaminoglycans upon dimerization of the collagen α1(XI) NPP. While it is clear that they are coordinated, it is still unclear whether dimerization is facilitated by interactions with glycosaminoglycans or alternatively, if dimerization occurs prior to binding, and subsequently facilitates interaction with glycosaminoglycans. This study provides information that furthers our understanding of the molecular interactions that may play a role in tissue integrity and extracellular matrix organization in tissue such as cartilage, blood vessel walls, and others.


Minor fibrillar collagens provide surface domains available for interaction; the NPP domains described in our study may play a role in molecular interactions. The minor fibrillar collagens of blood vessel walls include type V/XI heterotypic hybrid collagen molecules. Mutations in the genes encoding minor fibrillar collagens have been associated with platelet or coagulation dysfunctions and prolonged bleeding time associated with some diseases.22-27 Interaction between Arixtra and the NPP domain of minor fibrillar collagens like types V and XI may provide additional information regarding the mechanism of action. Additionally, pathologic chronic fibrosis can be caused by those proteins that promote adhesion, including the heparan sulfate proteoglycans. Thus, the results from this study can inform efforts toward targeting heparan sulfate–dependent molecular interactions that may yield new antifibrotic methods to prevent scar tissue formation in chronic fibrotic disease including systemic sclerosis, characterized by microvascular injury, autoimmune inflammatory responses and progressive, life-threatening fibrosis.


CONCLUSION To date there is no structural information available for the collagen α1(XI) NPP domain. Using computational modeling, we have created a homology model based on the NC4 domain of collagen IX (PDB:2UUR) as a template. The resulting model was used to investigate glycosaminoglycan binding and provided insight into protein:glycosaminoglycan interactions. In silico prediction provided preliminary insight into the importance of K152 in the binding interactions with glycosaminoglycans, from a 100-fold decrease in binding affinity of K152A compared to that of the wild type. Furthermore, dimerization of the collagen α1(XI) NPP domain has been observed experimentally to increase the binding affinity to glycosaminoglycans. This experimental result was successfully replicated through the modeling of a dimer and docking interactions


McDougal et al.

Page 6


of heparan sulfate and its derivatives. Additional studies are being conducted to determine if glycosaminoglycan binding induces dimerization, or alternatively, if increased affinity for glycosaminoglycans is a result of collagen α1(XI) NPP dimerization.

Acknowledgments Funding Sources Authors wish to thank Barbara Jibben for editorial assistance. This work was supported in part by grants from the National Institutes of Health/National Center for Research Resources Grant P20RR16454, NIH/NICHD R15HD059949, NASA (NNX10AN29A), Arthritis Foundation, Research Corporation Cottrell College Scholars, and Mountain States Tumor and Medical Research Institute, National Institutes of Health/NIAMS RO1AR047985 and KO2AR48672, the M. J. Murdock Charitable Trust, and Lori and Duane Stueckle Professorship.

References

NIH-PA Author Manuscript NIH-PA Author Manuscript

1. Brodsky B, Ramshaw JAM. The Collagen Triple-Helix Structure. Matrix Biol. 1997; 15:545–554. [PubMed: 9138287] 2. Fallahi A, Kroll B, Warner LR, Oxford RJ, Irwin KM, Mercer LM, Shadle SE, Oxford JT. Structural model of the amino propeptide of collagen XI α1 chain with similarity to the LNS domain. Prot Sci. 2005; 14:1526–1537. 3. Morris NP, Bachinger HP. Type XI Collagen is a Heterotrimer with the Composition (1α, 2α, 3α) Retaining Non-triple-helical Domains. J Biol Chem. 1987; 262:11345–11350. [PubMed: 3112157] 4. Thom JR, Morris NP. Biosynthesis and Proteolytic Processing of Type XI Collagen in Embryonic Chick Sterna. J Biol Chem. 1991; 266:7262–7269. [PubMed: 2016327] 5. Kadler KE, Baldock C, Bella J, Boot-Handford RP. Collagens at a Glance. J Cell Sci. 2007; 120:1955–1958. [PubMed: 17550969] 6. Gordon MK, Hahn RA. Collagens. Cell Tissue Res. 2010; 339:247–257. [PubMed: 19693541] 7. Keene DR, Oxford JT, Morris NP. Ultrastructural Localization of Collagen Types II, IX, and XI in the Growth Plate of Human Rib and Fetal Bovine Epiphyseal Cartilage: Type XI Collagen is Restricted to Thin Fibrils. J Histochem Cytochem. 1995; 43:967–979. [PubMed: 7560887] 8. Leppanen V-M, Tossavainen H, Permi P, Lehtio L, Ronnholm G, Goldman A, Kilpelainen I, Pihlajamaa T. Crystal Structure of the N-terminal NC4 Domain of Collagen IX, a Zinc Binding Member of the Laminin-neurexin-sex Hormone Binding Globulin (LNS) Domain Family. J Biol Chem. 2007; 282:23219–23230. [PubMed: 17553797] 9. Tan K, Duquette M, Liu JH, Shanmugasundaram K, Joachimiak A, Gallagher JT, Rigby AC, Wang JH, Lawler J. Heparin-induced cis- and trans-dimerization Modes of the Thrombospondin-1 Nterminal Domain. J Biol Chem. 2008; 283:3932–3941. [PubMed: 18065761] 10. Vaughan-Thomas A, Young RD, Phillips AC, Duance VC. Characterization of Type XI CollagenGlycosaminoglycan Interactions. J Biol Sci. 2001; 276:5303–5309. 11. Oxford JT, DeScala J, Morris N, Gregory K, Medeck R, Irwin K, Oxford R, Brown R, Mercer L, Cusack S. Interaction Between Amino Propeptides of Type XI Pro-collagen Alpha 1 Chains. J Biol Chem. 2004; 279:10939–10945. [PubMed: 14699108] 12. Morris GM, Goodsell DS, Halliday RS, Huey R, Hart WE, Belew RK, Olson AJ. Automated Docking Using a Lamarckian Genetic Algorithm and an Empirical Binding Free Energy Function. J Comput Chem. 1998; 19:1639–1662. 13. Johnson M, Zaretskaya I, Raytselis Y, Merezhuk Y, McGinnis S, Madden TL. NCBI BLAST: A Better Web Interface. Nucleic Acids Res. 2008; 36:W5–W9. [PubMed: 18440982] 14. Marti-Renom MA, Stuart AC, Fiser A, Sanchez R, Melo F, Sali A. Comparative Protein Structure Modeling of Genes and Genomes. Annu Rev Biophys Biomol Struct. 2000; 29:291–325. [PubMed: 10940251] 15. Laskowski RA. PROCHECK: A Program to Check the Stereochemical Quality of Protein Structures. J Appl Cryst. 1993; 26:283–291.


McDougal et al.

Page 7


16. Arnold K, Bordoli L, Kopp J, Schwede T. The SWISS-MODEL Workspace: A Web-based Environment for Protein Structure Homology Modeling. Bioinformatics. 2006; 22:195–201. [PubMed: 16301204] 17. Pettersen EF, Goddard TD, Huang CC, Couch GS, Greenblatt DM, Meng EC, Ferrin TE. UCSF Chimera-A Visualization System for Exploratory Research and Analysis. J Comput Chem. 2004; 25:1605–1612. [PubMed: 15264254] 18. Kozakov D, Hall DR, Beglov D, Brenke R, Comeau SR, Shen Y, Li K, Zheng J, Vakili P, Paschilidis IC, Vajda S. Achieving Reliability and High Accuracy in Automated Protein Docking: ClusPro, PIPER, SDU, and Stability Analysis in CAPRI Rounds 13-19. Proteins: Struct Funct Bioinf. 2010; 78:3124–3130. 19. Kozakov D, Brenke R, Comeau SR, Vajda S. PIPER: An FFT-Based Protein Docking Program with Pairwise Potentials. Proteins: Struct Funct Bioinf. 2006; 65:392–406. 20. Comeau SR, Gatchell DW, Vajda S, Camacho CJ. ClusPro: An Automated Docking and Discrimination Method for the Prediction of Protein Complexes. Bioinformatics. 2004; 20:45–50. [PubMed: 14693807] 21. Comeau SR, Gatchell DW, Vajda S, Camacho CJ. ClusPro: A Fully Automated Algorithm for Protein-Protein Docking. Nucleic Acids Res. 2004; 32:W96–W99. [PubMed: 15215358] 22. Cunniff C, Williamson-Kruse L. Ehlers–Danlos syndrome, type VIII presenting with periodontitis and prolonged bleeding time. Clinical Dysmorphology. 1995; 4:145–149. [PubMed: 7606321] 23. Kashiwagi H, Riddle JM, Abraham JP, Frame B. Functional and ultrastructural abnormalities of platelets in Ehlers–Danlos syndrome. Annals of Internal Medicine. 1965; 63:249–254. [PubMed: 14322851] 24. Uden A. Collagen and bleeding diathesis in Ehlers Danlos syndrome. Scandinavian. Journal of Haematology. 1982; 28:425–430. 25. Onel D, Ulutin SB, Ulutin ON. Platelet defect in a case of Ehlers–Danlos syndrome. Acta Haematologica. 1973; 50:238–244. [PubMed: 4201749] 26. Estes JW. Platelet size and function in the heritable disorders of connective tissue. Annals Internal Medicine. 1968; 68:1237–1249. 27. Anstey A, Mayne K, Winter M, Van de Pette J, Pope FM. Platelet and coagulation studies in Ehlers–Danlos syndrome. British Journal of Dermatology. 1991; 125:155–163. [PubMed: 1911298]

ABBREVIATIONS


NPP

amino propeptide

NC4

noncollagenous domain 4

Col11a1

collagen α1(XI)

LNS

laminin-neurexin-sex hormone binding protein

RMSD

root mean square deviation

BLAST

Basic Local Alignment Search Tool

PDB

Protein data bank


McDougal et al.

Page 8

NIH-PA Author Manuscript NIH-PA Author Manuscript NIH-PA Author Manuscript

Figure 1.

Collagen α1(XI) NPP homology models of: A) monomer built from the template 2UUR; B) lowest energy balanced interaction dimer from ClusPro.8,18-21


McDougal et al.

Page 9


Figure 2.

Heparin agarose affinity chromatography. Recombinant Col11a1 NPP was applied to a heparin-agarose column at low ionic strength and then eluted by increasing the NaCl concentration. Absorbance at 280 nm and conductivity were monitored continuously, and the data were plotted as the absorbance at 280 nm as a function of NaCl concentration. Protein concentrations used were A. 0.25 mg/mL; B. 0.50 mg/mL; C. 0.75 mg/mL; D. 1.5 mg/mL; E. 3.0 mg/mL; F. Overlay of data from panels A through E. Two peaks are observed: one eluting at 0-.4 mM and the other at 0.55 mM NaCl. Residuals are presented on the corresponding graph.

NIH-PA Author Manuscript GSTF Int J Bioinforma Biotechnol. Author manuscript; available in PMC 2014 October 10.

McDougal et al.

Page 10

Table 1


Collagen α1(XI) NPP monomer homology model and mutant interactions between heparan sulfate disaccharide.12 Model

Estimated Free Energy (kcal/mol)

Hydrogen bond residue interactions

Polar residue interactions

Wild Type

-8.08

ARG119, LYS149

ARG119, LYS149

K147A

-8.37

PHE118, LYS152

PHE118, LYS152

K148A

-8.41

LYS152

LYS152

K149A

-7.42

LYS148

LYS148

K152A

-5.34

-

LYS149

Submission of the collagen α1(XI) NPP structure file to ClusPro resulted in a cluster of dimer models with favorable desolvation and electrostatic energies between monomer units. ClusPro scoring of dimer models provided the most energetically favored collagen α1(XI) NPP:NPP dimer, with a weighted score of -918.2 (see Figure 1B).18-21

NIH-PA Author Manuscript NIH-PA Author Manuscript GSTF Int J Bioinforma Biotechnol. Author manuscript; available in PMC 2014 October 10.


NIH-PA Author Manuscript -21.13 -7.93

Arixtra

-17.53

Heparan sulfate disaccharide

Heparan sulfate decasaccharide

Est. Free Energy of Binding (kcal/mol)

Ligand

-5.07

-22.22

-10.57

Electrostatic Energy (kcal/mol)

-7.92

-23.18

-12.65

Total Intermolecular Energy (kcal/mol)

LYS73

LYS148

-

Hydrogen bond residue interactions

Interactions between ClusPro dimer and a heparan sulfate disaccharide, decasaccharide, and Arixtra.12,18-21

LYS72, ARG97, ASP125

LYS73, ARG97, ARG119, LYS148, TYR197

LYS73, ASN122, LYS149

Polar residue interactions


Table 2 McDougal et al. Page 11


Protein-specific prediction of mRNA binding using RNA sequences, binding motifs and predicted secondary structures.

Collagen XXII binds to collagen-binding integrins via the novel motifs GLQGER and GFKGER.

The predicted structure of the calcium-binding component of troponin.

GntR Family of Bacterial Transcription Factors and Their DNA Binding Motifs: Structure, Positioning and Co-Evolution.

Uncommon structural motifs dominate the antigen binding site in human autoantibodies reactive with basement membrane collagen.

PreDREM: a database of predicted DNA regulatory motifs from 349 human cell and tissue samples.

ATtRACT-a database of RNA-binding proteins and associated motifs.

Refined solution structure and ligand-binding properties of PDC-109 domain b. A collagen-binding type II domain.

Synthesis and structure of oxetane containing tripeptide motifs.

CD74 complexes.

Predicted structure of rabbit N-terminal, calcitonin and katacalcin peptides.

Immunochemical and collagen-binding properties of fibronectin.

Binding of betamethasone and dexamethasone by collagen.

Spectrin binding motifs regulate Scribble cortical dynamics and polarity function.

Zinc-binding cysteines: diverse functions and structural motifs.

Primary structure of the heparin-binding site of type V collagen.

Varying levels of complexity in transcription factor binding motifs.

Identification of novel sphingolipid-binding motifs in mammalian membrane proteins.

A database of protein structure families with common folding motifs.

Binding properties of SUMO-interacting motifs (SIMs) in yeast.

Automatic definition of recurrent local structure motifs in proteins.

Functional diversity of CTCFs is encoded in their binding motifs.

Predicted structure model of Bungarotoxin from Bungarus fasciatus snake.

Predicting transcription factor binding motifs from DNA-binding domains, chromatin accessibility and gene expression data.