Osmium tetroxide reaction with cell membrane

Background: $\ce{OsO4}$ is a widely used staining agent used in transmission electron microscopy (TEM) to provide contrast to the image. In the staining of the plasma membrane, osmium(VIII) oxide binds phospholipid head regions, thus creating contrast with the neighbouring protoplasm (cytoplasm).The osmium tetroxide molecule is tetrahedral and therefore non-polar. This nonpolarity helps $\ce{OsO4}$ penetrate charged cell membranes.

Question:

What would be the product formed by reaction of $\ce{OsO4}$ with polar head of phospholipid of cell membrane?

My attempt , since $\ce{OsO4}$ is non-polar and phospholipid is polar, therefore it must be some sort of covalent interaction, I couldn't find any literature regarding further at wikipedia.

• Its tetrahedral in shape, so all dipoles cancel each other. Could you explain further? – JM97 Oct 19 '17 at 10:23
• True, but that's irrelevant. – Ivan Neretin Oct 19 '17 at 10:25
• @IvanNeretin OsO4 is very reactive but it isn't polar. It's a symmetrical tetrahedron and can't be polar. It is very reactive with certain organic functional groups, which is why it stains tissues and polymers. – matt_black Oct 19 '17 at 12:23
• sigh Not that again... If I've never again heard of zero dipole molecule as "non-polar" I'd be happier. Substances are polar and this has not much to do with any dipoles. Also that is irrelevant to the question. – Mithoron Oct 19 '17 at 14:16
• I really don't understand the reason for OsO4 being polar, all 4 O atoms are at equidistant from central metal atom, so how can it be polar? – JM97 Oct 19 '17 at 15:33

According to the paper The chemical nature of osmium tetroxide fixation and staining of membranes by x-ray photoelectron spectroscopy[1] by White et al. (1976),

The results support a scheme for the reaction of osmium tetroxide with tissues in which the initial reaction site is the double bonds of unsaturated lipids to form $\ce{Os(VI)}$ derivatives. Subsequent hydrolysis and further reduction yield complexes of $\ce{Os(IV)}$ and $\ce{Os(III)}$. A mixture of these three states is present in membrane specimens during microscopic observation. $\ce{Os(VI)}$ and $\ce{Os(IV)}$ could be present as osmate esters and osmium dioxide, respectively; $\ce{Os(III)}$ could be present as an oxo- or amino complex(es).

However, since $\ce{OsO4}$ reacts with biological material, the above list is probably incomplete.

Riemersma[2] suggests that negatively charged colloidally dispersed osmium dioxide hydrate is deposited at the polar ends of lipids that contain a positively charged (phosphoryl)choline moiety (lecithin, phosphatidylcholines, etc). In this case we would have a pure Coulomb interaction.

References

1. White DL, Andrews SB, Faller JW, Barrnett RJ, Biochim Biophys Acta, 436(3):577-92 (1976).

2. Jelle C. Riemersma, J. Histochem. Cytochem, 11, 436-442 (1963)