PDBsum entry 1kjd

Cell adhesion protein

PDB id

1kjd

Contents

			Protein chain

					120 a.a.

			Metals

					_CA

References listed in PDB file

Key reference

Title

Knowledge-Based model building of proteins: concepts and examples.

Authors

J.Bajorath, R.Stenkamp, A.Aruffo.

Ref.

Protein Sci, 1993, 2, 1798-1810. [DOI no: 10.1002/pro.5560021103]

PubMed id

7505680

Abstract

We describe how to build protein models from structural templates. Methods to identify structural similarities between proteins in cases of significant, moderate to low, or virtually absent sequence similarity are discussed. The detection and evaluation of structural relationships is emphasized as a central aspect of protein modeling, distinct from the more technical aspects of model building. Computational techniques to generate and complement comparative protein models are also reviewed. Two examples, P-selectin and gp39, are presented to illustrate the derivation of protein model structures and their use in experimental studies.



	Figure 1. Fig. 1. A: Structure-basedsequencealignment of C-typelectin domains fP-selectin andthemannose-bindingprotein(MBP). Secondary-structureelments in MBP arelabeled.Residuescon- servedin MBP and P-slectin are boxedandresiduesconsrved in all selectins are shaded. This sequence-structurealignment provided the basis for comparative model building of the P-selectinligand-binding domain. B: Model structure of li- gand-bindng domain of P-selectin. The model,represented as a solid was ased on identified structural similar- ity to crystal structure of the C-typelectindomain of the rat mannose-bindingdomain, whichrevealed a previouslyunknown roteinfold.The iewis longa-helix 2, locatedbelow theloop regioncoloredinyellow. A conservedcalciumpositionisshwn in red.Analyisof the modelsuggeste a shallowdepression proximal to the conservedcalcium as a potentialligad-binding ite.Thisregion is flankedn the left by loop, colored yel- ow,with a five-residueinsertionreltive to the mannose-binding Residues the roposedligand-bindingregion of P-selectinweresubjected to ite-specificmutagenesisanalysis, the hypothesisregarding the location ofrsidues in P-selectincriticl inding to its cellularligand.Theresidues are shown in a colorcoded ashion:magenta,crucialfor binding; lavender,significant contribution to binding; blue, minorcon- tribution to binding. C: Assessentof the P-selectinmodelby comparison f 3D-profils of the MBPcrystal structure (rela- tive to the MBPsequence)andoftheP-selectinmodelstructure (relative to its sequence). The profiles were alculatedusing a 21-residuewindow for scoreaveraging. The calculatedZ-score for the MBPsequence and crystal structure is30.8, the Z-score for the P-selectinsequence and model is 34.9. No eg- ative valueswereobserved that ould ocalin- consistenciesin the structural odels. The analysissuggests n equivalentcompatibiltyofsequenceandstructure he model and the X-ray structure.		Figure 2. Fi. 2. A: Model structure of he extracellular domain of gp39. The structureofthehomotrimer is in green. The y-axis is approximately to thethreefold molecular symmetry axis. he locations f site-specific natural gp39 mutants in three patients (C.D., A.Y. and J.W.) are These mutations im- pair thefunction of gp39 in these patients.One of thethree symmetry-related region n gp39 that correspond o the receptor- binding sites in tumor necrosis factor beta (TNFP) is in Locations of the three mutations shown heresuggests that residue changes may interfere with the binding ofgp39 to receptor CD40. B: Comparison of calculated 3D-profiles, a 21-residue window, forthe trimeric tumor necrosis fac- orapha (TNFa) crystal andthe trimeric gp39 model structure elative to their sequences. The calculated Z-scores the TNFa structure and sequence and forthe gp39 model and se- were 30.8 and 32.8,respectively. The analysis supports he proposed structral similarity of gp39 and