Reader Comments

Post a new comment on this article

Grow factor diffusivity

Posted by andrea_gamba on 13 Jan 2014 at 11:33 GMT

Dear Authors,

I happened to notice the following imprecision: according to Ref. http://dx.doi.org/10.1039...
the diffusion coefficient of SDF-1/CXCL12 is of the order of

10^(-6) cm^2/s = 10^(-10) m^2/s

not 10^(-13) m^2/s.

See also http://dx.doi.org/10.1158...

This is somewhat larger than the diffusion coefficient of VEGF-165, compatibly with the Stokes-Einstein relation and the fact that SDF-1/CXCL12 is a smaller molecule (74 vs. 165 aa).

Actually, if one starts from the Stokes-Einstein relation, it is difficult to imagine that a grow factor may have the hypothesized 10^(-13) m^2/s diffusivity.

No competing interests declared.

RE: Grow factor diffusivity

RoelandMerks replied to andrea_gamba on 19 Jan 2015 at 14:00 GMT

Dear Andrea,

Thank you for bringing this to our attention. This was a unit conversion mistake from the value in cm^2/s used by Lin and Butcher (Lab Chip 2006)), which was estimated using the Stokes-Einstein equation and a 10kDa protein. The correct diffusivity of SDF-1 in a watery solution is 1.7 * 10^10 m^2/s . The mistake will be formally corrected to set the publication record straight.

However, note that in an in vivo or in vitro culture setting the diffusion coefficient rate of SDF-1/CXCL12 will likely be much lower than the diffusivity predicted by the Stokes-Einstein relation, because SDF-1, like many other small chemokines, is immobilized along the endothelial cell surface by the heparin sulfate GAG chains of proteoglycans (Ziarek et al. (2013) J Biol. Chem. 288:737). Such immobilization of chemokines is thought to induce sharp local gradients with the highest concentration near the cell surface, and is for example required for inducing cell migration in vivo in response to the chemokines RANTES/CCR5, MIP-1β/CCR4 and MCP-1/CCR2 (reviewed in Proudfoot (2006) Biochem. Soc. Transact. 34:422). Other secreted growth factors, including a number of VEGF-isoforms, also have heparin-binding domains allowing for similar mechanisms.

For these reasons we argue that assuming low diffusivities for the secreted growth factor is justified. For a more recent cell-based model of vasculogenesis that includes explicit immobilization of secreted growth factors see Köhn-Luque et al. (2011) PLoS ONE 6:e24175).

Best wishes,

Roeland Merks

No competing interests declared.