Figure 1.
Evolutionary relationship among the LGI vertebrate amino acid sequences.
The figure shows the best likelihood tree (−lnL = −21148.01332) obtained using the PHYML program. The length of the branches represents the number of reconstructed change of state over all sites (bar represents 0.2 substitutions per site), bootstrap values are reported at the nodes. Blue squares indicate the fish sequences whereas the green and red arrows respectively the amphibian and bird sequences. An asterisk indicates the Ornithorhynchus anatinus protein.
Figure 2.
Alignment of LGI family members and domain organization.
Multiple alignment of representative homologs in the LGI family. Species are abbreviated as follows: Hs = Homo sapiens; Mm = Mus musculus; Rn = Rattus norvegicus; Dr = Danio rerio; Xt = Xenopus tropicalis; Cf = Canis familiaris. The LGI1 domains and secondary structure are shown on the top part. Missense mutations analyzed in this paper (triangles) and putative glycosylation sites (stars) are indicated on the bottom of the alignment. Red lines are used to connect cysteine residues that form disulphide bridges in the structural model. acc: accessibility level from DSSP (black = high and white = low).
Figure 3.
A. Consensus sequence repeat pattern of the LRR domain. Secondary structure is drawn on the top part of the alignment: an arrow represents the β-strand and a ribbon the α-helix connected by curved lines (loops). B. Schematic diagram of repetitive structural units in LGI1 protein. Conserved positions of the consensus pattern are reported on the diagram. Coloured pink spheres for buried residues and blue spheres for exposed residues.
Figure 4.
LRR model, structural analysis.
A Cartoon of the LRR model coloured from N-terminal (blue) to C-terminal (red); B. Electrostatic surface (negative charge in red and positive charge in blue); C. Position of missense mutations, mutated residues are shown as spheres with structural mutations indicated in red; D. Conserved surface with ConSurf colour code from unconserved (cyan) to strictly conserved (magenta).
Figure 5.
A. Consensus sequence repeat pattern of EPTP domain. h = hydrophobic residue; p = polar; a = aromatic residue; t = tiny residue. Secondary structure is drawn on the top part of the alignment. Arrows represent β-strands connected by curved lines (loops). Loops forming the top surface are coloured in green, while those forming the bottom surface are coloured in blue. B. Schematic diagram of repetitive structural units in the LGI1 protein. Conserved positions of the consensus pattern are reported on the diagram. Pink and blue spheres indicate buried and exposed residues respectively.
Figure 6.
EPTP model, structural analysis.
A. Top (up) and bottom (down) view of electrostatic surface of EPTP model (negative charge in red and positive charge in blue); B. Top (up) and bottom (down) view of the conserved surface of EPTP model with ConSurf colouring from unconserved (cyan) to strictly conserved (magenta). C. Cartoon of the EPTP model in top and lateral view with ConSurf colouring. Spheres indicate residues found mutated in ADTLE patients with structural mutations indicated in red.
Figure 7.
Top (A) and lateral (B) view of the hypothetical peptide binding site on the EPTP model. The position of a hypothetical peptide (green spheres) was obtained by superimposition of the EPTP model with the WDR5 structure (PDB code 3EMH). Note that the insertion specific for LGI1 ( in yellow) maps on the bottom face of the domain.
Figure 8.
Hypothetical structural assembly and interactions.
A. LGI1 is represented as the association of LRR (green arc) and EPTP (violet trapezoid) domains. LGI1 interactions with ADAM proteins likely occur on the top surface of the EPTP domain. B. The two hypothetical ways by which LGI1 could mediate the trans-synaptic interaction between presynaptic ADAM23 and postsynaptic ADAM22.
Table 1.
Missense mutations overview for the LGI1 protein.