Epigenetics and DNT

Evidence exists that environmental chemicals as well as drugs can influence the epigenome (Landrigan et al. 2005; Onishchenko et al. 2008; Skinner et al. 2011) but until now no clear common effects can be described (e.g. a compound up-regulates only active histone PTMs).

This could be due to the epigenome being regulated very dynamically and the chromatin states being strongly depended on the cell type and on the developmental stage. Therefore, if a compound can only affect open chromatin regions, it might affect an actively transcribed gene in one cell type but not in another, where the gene is silenced and packaged in a dense chromatin state. Also, histone modifications could be the consequence of altered gene expression, which was recently suggested as opposing hypothesis to the histone code (Henikoff and Shilatifard 2011).

An offset between early insult and late effect has also been seen in human diseases (described in Chapter 1.1.3.2 and 1.22). In these late-onset diseases epigenetic mechanisms have been shown to play a major role. We have shown that exposure to DNT compounds can alter neural differentiation depending on the exposure period, i.e. short exposure can be compensated for, while prolonged exposure leads to an adverse effect. Thus, we wondered if

and the adverse outcome after prolonged exposure as it has been proposed for some late-onset disorders.

Indeed, we showed that short, transient exposure increase histone acetylation levels transiently without altering the overall differentiation or the histone methylation pattern at the promoters of marker genes (Fig. 6.3 A). Prolonged early exposure also increased histone acetylation levels transiently, but additionally altered histone methylation patterns and neural differentiation (Fig. 6.3 B). This could indicate that changes in histone methylation patterns could represent the persistence detector for toxicant exposure and only if this pattern is changed the differentiation track is disturbed. We postulate that two changes are necessary after toxicant exposure to alter histone methylation patterns and the neural differentiation of hESC: altered histone acetylation levels, which can happen very fast and transient, and changes in the transcriptome, which occur slower (Fig. 6.3 C). Only if both positions are changed (+), the output is affected (+). We found that a short impulse (exposure to toxicant) can alter acetylation levels but not the transcriptome at the end of differentiation (Fig. 6.3 C, 1). Only prolonged early exposure affects the acetylation levels (transiently) and the

Figure 6.3. Epigenetic changes as persistence detector of toxic insults. (A) Short early exposure to a VPA or TSA transiently increases histone acetylation but does not alter the histone methylation pattern or the differentiation track. (B) Increased early exposure to VPA or TSA also transiently increases acetylation levels.

Subsequently histone methylation patterns are changed and neural differentiation is disturbed. (C) Exposure to a toxicant might fast alter acetylation and seems to alter the transcriptomic fingerprint in a slower manner. Only if both changes occur (input + and +), the signal seems to be mediated towards an altered differentiation track (output purple +). (1) Altered acetylation without altered transcriptome does not seem to affect histone methylation and differentiation. (2) Altered acetylation plus altered transcriptome change histone methylation and differentiation.

A

B

C

transcriptome and therefore changes the differentiation (Fig. 6.3 C,2). In summary, it seems as if alterations in histone methylation patterns of the promoters of marker genes accumulate during prolonged exposure, even in the absence of the primary effect of the DNT compound (increase in histone acetylation) (see also Chapter 6).

Yet, the altered histone methylation pattern could be secondary due to an altered differentiation track (Fig.6.4). To investigate which explanation is more likely, shorter time periods in the beginning of the differentiation need to be investigated. However, the method of chromatin immunoprecipitation has a higher variability compared to qPCR or whole genome analysis on Affymetrix chips.

As we detected secondary alterations on histone methylation when the primary increase in histone acetylation was absent again, it could be that altered chromatin marks precede alterations in marker gene expression. The interconnection of epigenetic mechanisms has indeed been shown to work in both directions. As mentioned above, DNA methylation can recruit HDACs via MeCP2, thereby decreasing acetylation levels of histones and enhancing the silent state of chromatin. VPA, in turn, inhibits HDACs and was shown to induce replication-independent demethylation of DNA (Szyf 2009).

Another reason for the lack of evidence for common effects by epigenetic modifying chemicals could have more technical reasons. Most studies either investigate global levels of PTMs or perform chromatin immune precipitation (ChIP) analysis on limited regions of the genome. Also, in this thesis the enrichments of histone PTMs are investigated only at or around the transcription start site of a gene, but the regulatory sequences of a eukaryotic gene can be several thousand base pairs long that cannot be covered by manual primer design

Figure 6.4. Question of cause and consequence in disturbed neural development and altered epigenetic marks. Altered patterning indicated by an altered expression pattern (see Fig. 6.2) is associated with altered chromatin marks (histone methylation patterns at the promoters of marker genes). Alterations in the transcriptome and in chromatin marks can be caused in three ways. First, the marks can be altered simultaneously by toxicant interference with both mechanisms. Second, altered gene expression can lead to a different cellular fate, having different chromatin marks. Third, the toxicant can alter the chromatin marks which leads to alteration of gene expression, finally leading to altered differentiation.

anymore. Therefore, to investigate broader chemical effects on histone PTMs and the link between epigenetic mechanisms and DNT, studies will need to utilize the technology of ChIP-sequencing (ChIP-seq).

Even though we are aware of the shortcomings and needs for refinement, our data is amongst the first to propose that epigenetic mechanisms mediate the switch from adaptive responses after short exposure to a DNT compound to an adverse effect after prolonged treatment.