Transcription is a process that depends on many elements such as transcription factors, DNA methylation and chromatin structure. Histones are essential for the proper regulation of transcription, and the posttranslational modifications on their tails have been related both to activation (H3K4me3, H3K9ac and H3K36me3) and silencing (H3K27me3 and H3K9me3) of gene expression.
Using the fruit fly as model organism, in our lab at the CRG and in collaboration with Montserrat Corominas’ lab in the Universitat de Barcelona, we have found a set of genes (defined as developmentally regulated genes by their restricted expression to a short period of time) that are expressed without the canonical histone marks associated to activation of transcription. These findings have been published in the October’s issue of Nature Genetics, the cover of which also refers to our work and to Dalí’s butterflies –designed by Luisa Lente and Hagen Tilgner.
Why these developmentally regulated genes show a different pattern of histone modifications is not known, but we speculate that the need of rapid activation–deactivation of gene expression during development may be easier to achieve in absence of the histone posttranslational modifications. In this context, transcription factors would play a more important role. This hypothesis is reinforced by the fact that the binding of transcription factors is different between developmentally regulated and stable genes (those expressed throughout development).
There are previous reports claiming that some genes are transcribed in the absence of the canonical histone marks for gene expression (Hödl and Basler –2012–, Chen and collaborators –2013– and Zhang and collaborators –2014–), but our main contribution to the field is that we have seen that the lack of chromatin marks is a general feature of genes that are expressed for a short period of time, being the expression of these genes likely to be regulated mainly by the action of transcription factors.
We are often asked how come nobody had described these patterns before. Actually, the answer is probably because our approach was different from the beginning. When we started the project our main goal was to analyze the chromatin marks and the splicing of tissue- and time-specific genes in the fruit fly. The settings used to define the tissue- and time-specific genes were, then, very astringent, being our developmentally regulated genes only expressed in one time point throughout development. However, to our surprise, these particular genes did not show the expected histone marks when they were expressed. It was then that we focused our attention in these particular genes.
This work was presented in the 11th Transcription and Chromatin Conference (2014) at the EMBL in Heidelberg – where they were already introduced as being controversial- and was afterwards highlighted in the EpiGenie webpage and in the Epigenetics journal, in both cases by Sascha H. C. Duttke.
As we knew that these results are controversial and somehow challenge the classical association of histone marks with transcription, we put a great deal of effort in generating and analyzing all available data to produce as solid results as possible to demonstrate that our observations do not arise from a detection artifact. Besides, we think that with this work we open a new line of research as the results, so far observed in fly and worm, need to be further validated in mammalian systems. In this sense, our results so far with the ENCODE mouse tissues and human cell lines point out that this lack of chromatin marks in regulated genes may be a general feature along metazoans.