Models for the C1 complex determined by physical techniques

The C1 complex is an association of C1q and C1r2C1s2. Neutron scattering and ultracentrifugation provide a valuable means of understanding the solution structure of the subcomponents and their complex, and these can be supplemented by protein structure prediction techniques. C1q is constructed from six globular heads connected by collagen-like arms. The solution data for C1q show that the arms are of length 14.5 nm and not 11.5 nm as proposed from electron microscopy, the average arm-axis angle is 40 degrees, and that the structure is flexible in solution at the junction of the six arms. The sequences of C1r and C1s show that each is constructed from six protein domains. C1r and C1s are elongated macromolecules of lengths 18-20 nm. Their solution properties are best described as the lengthwise arrangement of a protease domain of diameter 4 nm, two "short consensus repeat" domains, each of length 4 nm, and an N-terminal globular entity of length 6 nm containing the first three protein domains. Solution data on the C1r2 dimer is interpreted as an X-shaped association of the two C1r monomers as proposed from electron microscopy. Six criteria are enumerated for constructing models of C1 from these two structures, and four distinct models for the C1 complex are reviewed. While further evidence is required to make this choice unequivocal, the W-model is favoured. This places each monomer of C1r and C1s on four adjacent arms of C1q, and offers the most reasonable explanation of the known properties of the C1 complex.