We welcome Dr Biggin’s critique of our study. The paper in PlosOne in our opinion deserves attention as it addresses an issue with this frequently-used technology that is often overlooked.

Our research began with a glaring discrepancy between the results of live cell imaging and imaging of the same cells after fixation with paraformaldehyde. This is not a subtle effect and therefore demanded an explanation. We hypothesized that time-of-occupancy was at the root of it, and did FRAP and ChIP to test this. The results supported our hypothesis, as explained in the paper.

Dr Biggin’s critique focuses on ChIP, whereas our analysis, as explained above, is primarily about the reason why live-cell imaging and paraformaldehyde fixation give completely different results by microscopy. He questions our interpretation that a threshold rather than increased time spent unbound is responsible for the results. The reason for invoking a threshold is that live cell imaging shows that all the MeCP2 mutants without exception are well-localized to heterochromatic dots, whereas after fixation some of the mutants are localized whereas others give uniform nuclear labeling. In our view the live cell images cannot be lying (in other words all of these mutant forms of MeCP2 must be spending most of their time bound to heterochromatic dots). Therefore fixation must be responsible for the in vitro artifact. Whether or not mutant MeCP2 localizes to the dots after fixation matches well with the residence time by FRAP, so we propose that residence time is the key.

To explain the core microscopic observations that prompted our study, Dr Biggin proposes that some difference in the way we treated the cells is responsible. To us this explanation seems highly unlikely to be correct. In the course of this work we ruled out that temperature had any effect on localisation of MeCP2 to chromatin. Furthermore, there is no evidence or precedent for “subtle changes in cell physiology” altering protein binding on such a scale. In our opinion, the limitation of Dr Biggins' critique is that it offers no plausible explanation for the large and reproducible distinction between MeCP2-mutant localisation in live versus fixed cells. Our experiments, on the other hand, provide experimental evidence in favour of a residence time hypothesis.

We speculate that the threshold seen by microscopy could arise because of competition between intra-molecular cross-links that destroy the DNA binding ability of MeCP2 and MeCP2-DNA crosslinks that freeze its interaction with DNA. Intra-molecular cross-links probably occur at a constant rate whereas the mutations severely reduce the time spent in contact with DNA. Thus protein denaturation starts to predominate over DNA cross-linking when DNA binding becomes very transient. Viewed in this way, the threshold is due to competition between these processes. To test this, and to advance our understanding of formaldehyde crosslinking, further experimentation will undoubtedly be required.