I have measured the size of the SP-D images in the article by Perino et al, and their assessment of whether SP-D assists in controlling infection with vaccinia virus. The purpose was to assess whether there was a difference in the size of the molecules from one figure to the next, (which there ws not : Fig 4c-molecules 1-2 =118.71 nm +/ 7.8 nm; Figure 4b 1-11 =118.73 nm +/ 8.76 nm; Figure 4b2 molecules 1-9; =117.27 nm +/ 5.94 nm; Figure 4b3 molecules 1-4; 115.52 nm +/ 4.74 nm. They provided bar markers for each of the images and this was used to determine the arm lengths and total diameter. (SEE MY figure below).
This is the first negative stained group of SP-D dodecamers and lower number multimers that I have measured using the line-node method and the diameter method. While the text appears to suggest that this is recombinant human SP-D (just like Arroyo et al says their dodecamers are recombinant human SP-D) the difference in size between the AFM (latter paper) and negative staining (Perrino et al is huge) i.e. 30 nm approximately.

Figure below shows exactly what molecules looked like, and exactly where i anticipated the molecules to begin and end.

RhSP-D negatively stained and pictured in a publication by Perino et al I have used to map using LUT the peaks along the collagen like domain.  There are now three techniques which provide info on the possible configuration of that portion of the molecule.  The peak smoothness, low noise, and visual appearance make the LUT plots resulting at least as informative as AFM images.

red bar marker is that size provided in the original figure (Perino et al, Fig 4b, central multimer from the middle panel. The image was resampled in photoshop to 700ppi, but no changes in contrast were made.  The image was rotated and the arm within the white box, cropped and exported to tiff and opened in ImageJ2 (FIJI). The image below this shows that area plotted (half of the 110 nm diameter found for the whole multimerm) going to the center of the Ntermini junction with the other arms (in this case on the right hand side of the image). Plot is in blue, the left side two peaks are the CRD (often showing a valley where the CRD are just beside or adjacent or overlapping each other), a little slow slope which would most likely be the neck, then moving to the right, 3 very obvious peaks, a lesser obvious peak and the Nterminus peak.  The only problem here is that the 3 peaks in the collagen like domain usually are reversed in height, the highest peak being nearest the Nterminus…. so thats a little bit of a problem.

I have tried quite hard to figure out why there is a discrepancy (about 122% larger) in the size of the surfactant protein D dodecamers (well all species of SP-D molecules) in the cover image (which is stated by JMB cover description to be 1micron as the bar marker – total relative width of the micrograph) but which when calculated on individual selected out dodecamers is off by considerable amounts…. much larger. I redid this at least three times thinking i was making an error but i am pretty sure that the cover image micron marker (in the text) is not correct.  This makes it a little harder to explain when I return all dodecamers to a standard size (to be determined yet, but somewhere around 145 – 150nm from the tip of the CRD on one arm to the tip of the CRD on the other side.  If anyone out there is listening or has an explanation… please advise. Image below (top dodecamer) shows a dodecamer from their figure 1 measured with the bar marker for that image, and the bottom dodecamer is excised from their cover image and sized with the magnification given in the cover text.

GIRTGIQ – sequence in C1q to which is attributed the “kink” in the collagen like domain of C1q (Pfliegers et al) . They write “The two following interruptions in the Gly-Y-Z triplet repeats are involved in the kink of the collagen-like region,i.e.the insertion of a threonine at position 39 in the A chain and the replacement of a glycine by an alanine at position 36 in the C chain”. Also they suggest three lysines (58,59 and 61in A,B and C chains) along the collagen-like domain are likely where interaction sites with C1r and C1s proteases becoming the C1 complex. Their pictures are wonderful.