Surfactant protein D is a terrific molecule, and the assorted arrangements of multimers has been described somewhat but I have only found one publication which actually tries to figure out details about the unique shape of the collagen-like domain. The carbohydrate recognition domain and the coiled-coil neck section has been studied – ad nauseum and the RCSB has countless molecular models. The N-terminal has been a little less studied, even less the collagen-like domain.  I am not really sure why that is since there are dozens of images of SP-D trimers, dodecamers and multimers (fuzzyballs, fuzzy balls) from which to determine length and curvature of that domain.  At anyrate, there are some images which show kinks at the neck (a comment typically included in descriptions of SP-D, so that is not new) and at least one paper suggests there might be something going on at the glycosylation site (closer to the N terminal).  Here is a group of just for trimeric arms (cut and rearranged from dodecamers (images from manuscripts that I did NOT write – thank you for those) and it becomes clear that an assessment of angles in this kind of order might be useful. One kink, maybe two kinks (see blue arrows on left).

It is pretty clear that the length of the collagen-like segment that show up right at the crotch of the bifurcation of the arms of the dodecamer is not going to end up the exactly length of the whole collagen-like portion. It seems at this point that the there is N terminal association of the four trimers and then more close binding for a greater length up the collagen-like sequence.  The proportions of the micrographs just don’t add up (especially visually) to the dimensions given in the literature for the collagen-like sequence. In particular too, the diagrams are way off. This begs the question, where does the central ly bound portion end? and does this make a difference in the shape of the center of the fuzzyball made up of 6 or 8 dodecamers? as i mentioned weeks ago on this blog.