PDBsum entry 1ie5

Cell adhesion

PDB id: 1ie5

Contents

Protein chain

107 a.a. *

* Residue conservation analysis

References listed in PDB file

Key reference

Title

Solution structure of the third immunoglobulin domain of the neural cell adhesion molecule n-Cam: can solution studies define the mechanism of homophilic binding?

Authors

A.R.Atkins, J.Chung, S.Deechongkit, E.B.Little, G.M.Edelman, P.E.Wright, B.A.Cunningham, H.J.Dyson.

Ref.

J Mol Biol, 2001, 311, 161-172. [DOI no: 10.1006/jmbi.2001.4861]

PubMed id

11469865

Abstract

Homophilic binding of the neural cell adhesion molecule (N-CAM) mediates the calcium-independent cell-cell adhesion that is involved in neuronal development. Two hypotheses have been advanced for the mechanism of homophilic binding. Cell-based experiments have implicated each of the five extracellular immunoglobulin (Ig) domains of N-CAM in the homophilic adhesion interaction, and have predicted that the third domain (Ig III) self-associates. The alternative hypothesis is based on solution observations, which implicate a specific antiparallel interaction between the first two Ig domains (Ig I and Ig II). In order to test these hypotheses, we have determined a high-resolution solution structure of recombinant Ig III (sequence derived from chicken N-CAM) and examined the aggregation behavior of isolated Ig domains in solution. The structure shows that Ig III adopts a canonical Ig fold, in which the beta strands ABED and A'GFCC' form two beta sheets that are linked by a disulfide bond. In contrast to the demonstrated aggregation of Ig III on solid supports, we were unable to demonstrate self-association of Ig III under any of a variety of solution conditions. The structure shows that the surface of Ig III is dominated by two large acidic patches, which may explain our failure to observe self-association in solution. To evaluate the involvement of the Ig I-Ig II interaction in cell-cell adhesion, we designed a point mutation in Ig I (F19S) that proved sufficient to abrogate the Ig I-Ig II interaction seen in solution. However, the introduction of this mutation into full-length N-CAM expressed in COS-7 cells failed to affect N-CAM-mediated cell-cell adhesion. The inability to observe Ig III self-association in solution, combined with the failure of the F19S mutation to affect N-CAM-mediated cell-cell adhesion, suggests that, although solution studies can give important insights into the structures of individual domains, the interactions observed in solution between the domains may not be representative of the interactions that occur on the cell surface.

Figure 4.
Figure 4. (a) Stereoview of the 20 final AMBER structures with the lowest restraint violations. For clarity, only residues 7-107 are shown. Structures were superimposed on the backbone heavy atoms of the residues in the b strands and the turn of helix using the program MOLMOL.[50] (b) Ribbon diagram depiction of the mean structure of Ig III showing b strands in blue and the turn of 3[10] helix in red. Only residues 7-107 are shown. The Figure was prepared with the program MOLMOL.[50]

Figure 6.
Figure 6. Analytical ultracentrifugation data on the first three Ig domains from N-CAM, collected on a Beckman XL-I centrifuge. (a) Sedimentation data recorded at 14,000 rpm on an 11 µM solution of wild-type Ig I-III in PBS (o) fitted to an associating model of monomer-dimer equilibrium (continuous line). (b) Sedimentation data recorded at 20,000 rpm on a 12 µM solution of F19S Ig I-III in PBS (o) fitted to a single species (continuous line). The residual differences between the experimental data and the fit for each point are shown above for each experiment.

The above figures are reprinted by permission from Elsevier: J Mol Biol (2001, 311, 161-172) copyright 2001.