Andrew Earl and Dr. Steven Johnson, Microbiology & Molecular Biology

Introduction

Nucleosomes are the fundamental unit of chromatin organization. They consist of an octamer of histone proteins (two of each H2A, H2B, H3, and H4) wrapped around by 147 base pairs of DNA. Their positioning and occupancy around important genetic elements such as enhancers and promoters are some of the most important means of epigenetic regulation: changes in expression of genes that reflect a change in something other than DNA sequence. For example, highly-expressed genes are generally associated with low nucleosome occupancy, whereas inactive genes are associated with high nucleosome occupancy. It has also been shown that the positioning of nucleosomes is well-regulated around promoters and enhancers. In addition, different post-transcriptional modifications (PTMs) to amino acid residues on histone tails are correlated with different chromatin states. Among the most well-characterized PTMs are acetylation to lysine residues on the histone protein H3. The precise mechanisms of how PTMs regulate chromatin states are still under investigation.

A significant portion of research done on histone PTMs has been to characterize them in three ways 1) by determining what genomic elements PTMs tend to be associated with in vivo 2) the molecular pathways leading to PTM addition or removal and 3) how transcription factors interact with PTMs. Even though differential nucleosome positioning and occupancy have been found to be powerful aspects of nucleosomes’ epigenetic regulation, until recently, no lab has yet investigated how histone PTMs affect the DNA sequence preference of their nucleosomes. This project sought to answer how acetylation at selected lysine residues on the histone protein H3 (some of the most well-understood PTMs) affects the positioning of the modified nucleosomes on the DNA sequence. Solving this will be a significant step in understanding the regulatory effects of PTMs and could potentially add a new layer to the histone code hypothesis.

Methodology

To determine how specific lysine acetylation on H3 affects nucleosome positioning in vitro, we used the amber codon unnatural amino acid incorporation. Specifically, we transformed E. coli with plasmids that expressed the following: a tRNA and aminoacyl-tRNA synthetase to recognize to add an acetylated lysine rather than stopping when encountering the TAG amber stop codon, modified H3 genes to have the amber stop codon in place of lysine codons at specific residues, and a T7 RNA polymerase to control timing of protein production. Next, we induced expression of each modified (and control) H3 as well as the other histone proteins and dialyzed them to form histone octamers after purification on a gel-filtration column. Another round of gel-filtration chromatography was performed for further purification. Then, C. elegans genomic DNA sheared by restriction enzymes or sonication was reconstituted with the histone octamers to form nucleosomes. We then digested these nucleosomes with MNase to isolate only the DNA that bound to histone octamers.

These DNA fragments were sequenced using Illumina sequencing and mapped to the C. elegans genome. I then wrote a program that extrapolated the 147 base pair sequence based on the center of the read and counted the number of each possible combination of 1, 2, and 4 base pairs (e.g. A, AC, and ACGT) at each position within the 147 base pair sequence. Last, I compared the relative counts at each position for each type of modified nucleosomes to control, unmodified nucleosomes.

Results

Though more rigorous statistical methods need to be employed to identify the most significant differences between unmodified nucleosomes and nucleosomes with acetylated H3 lysines, initial results are promising. Many of the differences appear to be significant. Notably, all modified nucleosomes show a decreased preference for sequences of adenine and thymine repeats, particularly at positions 12-14 base pairs from the edges of the 147 base pair sequence. Another striking observation is that each type of modified nucleosome displays a decreased preference for motifs involving GTAC on the flanking regions of the sequence. There are several other sequences that are either more or less common in specific regions between the modified nucleosomes and control nucleosomes.

Discussion

Our data suggest that histone modifications do play a role in nucleosome positioning in vitro, though perhaps that role is minor. There are many more steps that need to be taken to verify this role and measure its impact. First, more thorough statistical analysis needs to be run on this data to determine what is statistically significant. Next, an analysis of what types of genetic elements most commonly contain the sequences that display a significant difference between modified and control nucleosomes should be performed. Validation of this data by repeating the experiments and comparing these results against in vivo CHIP-seq datasets would help quantify how important the role of histone modifications in nucleosome positioning in living organisms is. In addition, more histone PTMs can be studied using this method, including phosphorylation, crotonylation, and ubiquitination. Future research could also investigate how combinatorial patterns of histone modifications affect nucleosome positioning.

Conclusion

This project investigated how acetylation at specific lysine residues on H3 affects nucleosomes’ DNA sequence preference in vitro by making modified nucleosomes and comparing the sequences of the C. elegans DNA that they bound to in vitro to control, unmodified nucleosomes. We found that there are differences at specific positions of these 147 base pair fragments of DNA, especially when it comes to regions of adenine and thymine repeats close to the flanking regions of the nucleosomal DNA. Future analysis of this data is required to identify what initial observations are most significant, and future experiments can use this method to determine how other histone PTMs affect nucleosome positioning in vitro.