I found this article by Martin van Eijk et al which had rotary shadowed images of surfactant protein D.  One of these (which might be a dodecamer…with perhaps one of the trimers (on the left of the image) that is a little unwound? perhaps, or it also may be a structure with more than the four trimers of a dodecamer) but it has a small center area which presumably might indicate a different type of N terminus binding than human SP-D.  I would like to find AFM images of porcine SP-D to compare since this image (shadowed) and AFM produce such different views of SP-D that they are not comparable.

It will be fun to count the arms on the fuzzyballs in this image.  It appears that the porcine SP-D dodecamer (which i have measured as the diameter of a circle (dotted red line) that intersects the CRD) is just over 100nm, actually about 110nm according to their bar marker (red below). The ratio of CRD to length and width of the arms of the collagen-like and N terminus domains “looks” at first glance to be similar to hSP-D.