Type I and II classical cadherins help to determine the adhesive specificities of animal cells. Crystal-structure determination of ectodomain regions from three type II cadherins reveals adhesive dimers formed by exchange of N-terminal beta strands between partner extracellular cadherin-1 (EC1) domains. These interfaces have two conserved tryptophan side chains that anchor each swapped strand, compared with one in type I cadherins, and include large hydrophobic regions unique to type II interfaces. The EC1 domains of type I and type II cadherins appear to encode cell adhesive specificity in vitro. Moreover, perturbation of motor neuron segregation with chimeric cadherins depends on EC1 domain identity, suggesting that this region, which includes the structurally defined adhesive interface, encodes type II cadherin functional specificity in vivo.

Vertebrate genomes encode nineteen "classical" cadherins and about a hundred non-classical cadherins. Adhesion by classical cadherins depends on binding interactions in their amino terminal EC1 domains, which swap N-terminal β-strands between partner molecules from apposing cells. However, strand swapping sequence signatures are absent from non-classical cadherins, raising the question of how these proteins function in adhesion. Here we show that T-cadherin, a GPI-anchored cadherin, forms dimers through an alternative non-swapped interface near the EC1-EC2 calcium binding sites. Mutations within this interface ablate the adhesive capacity of T-cadherin. These nonadhesive T-cadherin mutants also lose the ability to regulate neurite outgrowth from T-cadherin expressing neurons. Our findings reveal the likely molecular architecture of the T-cadherin homophilic interface, and reveal its requirement for axon outgrowth regulation. The adhesive binding mode employed by T-cadherin may also be used by other non-classical cadherins.

Summary Cadherin-mediated cell adhesion is achieved through dimerization of cadherin N-terminal extracellular (EC1) domains presented from apposed cells. The dimer state is formed by exchange of N-terminal β-strands and insertion of conserved tryptophan indole side chains from one monomer into hydrophobic acceptor pockets of the partner molecule. The present work characterizes individual monomer and dimer states and the monomer-dimer equilibrium of the mouse Type II cadherin-8 EC1 domain using NMR spectroscopy. Limited picosecond-to-nanosecond timescale dynamics of the tryptophan indole moieties for both monomer and dimer states are consistent with well-ordered packing of the N-terminal β-strands intramolecularly and intermolecularly, respectively. However, pronounced microsecond-to-millisecond timescale dynamics of the side chains are observed for the monomer, but not the dimer, state, suggesting that monomers transiently sample configurations in which the indole moieties are exposed. Dimer formation is favored at low pH and by the presence of calcium, indicating a role for calcium in the strand swapping mechanism. The results are discussed in terms of possible kinetic mechanisms for EC1 dimerization.

Bullous Pemphigoid Antigen 1 (BPAG1) is a member of the plakin family of proteins. The plakins are multi-domain proteins that have been shown to interact with microtubules, actin filaments and intermediate filaments, as well as proteins found in cellular junctions. These interactions are mediated through different domains on the plakins. The interactions between plakins and components of specialized cell junctions such as desmosomes and hemidesmosomes are mediated through the socalled plakin domain, which is a common feature of the plakins. In this study, we report the crystal structure of a stable fragment from BPAG1, residues 226-448, defined by limited proteolysis of the whole plakin domain. The structure, determined by single-wavelength anomalous diffraction (SAD) phasing from a selenomethionine-substituted crystal at 3.0 Å resolution, reveals a tandem pair of triple helical bundles closely related to spectrin repeats. Based on this structure and analysis of sequence conservation, we propose that the architecture of plakin domains is defined by two pairs of spectrin repeats interrupted by a putative Src-Homology 3 (SH3) domain.

No abstract