Recent comments by AA Spector are interesting and deserve further discussion. Earlier elegant work by Maniotis and Ingber demonstrated the interconnectedness between the cell surface (via integrins) and the nucleus through the cytoskeleton. Coffey also promoted the importance of cytoskeleton in mechanical signal transduction in normal cells and the differences in tumor cells. There ideas are not well received, however, by the field. An important issue is the magnitude of the surface deformation: if it is large, then one expects the nucleus to be deformed. A finite element analysis by SM Mijailovich et al (J Appl Physiol, 2003) showed that a localized surface load decays rapidly in space-as a function of distance squared, suggesting that a physiologic load may not be able to deform structures inside the nucleus directly. This is consistent with St Venant principle that states a local force causes only a local deformation. A recent review by Vogel and Sheetz also highlighted the importance of local deformation leading to local biochemical activities.
The recent experimental evidence from our group are not consistent with the above modelings or predictions. We showed that even for the surface deformation as small as ~300 nm, one can still observe significant (and concentrated) strains in the nucleolus and in the cytoskeleton as far as 30 um away from the load (Hu et al, 2003, 2004, 2005). The deformation and strains in the nucleolus are observed in the absence of detectable nuclear membrane deformation! The model(s) by Zhigang Suo and me is an attempt to explain these observations.
Much more work needs to be done in this area.