There are >100 lysine residues within the nucleosome and greater than 90% have been shown to have post-translational modifications (PTMs) (Figure 1). If you considered PTMs as a chemical indexing system and each PTM creates a unique nucleosome you could make >10¹⁸ distinct nucleosomes. Not all of these modifications are observed and coordinated mechanisms of crosstalk suggest the potential that they may make up a learning network.

Mass spectrometry coupled enzyme kinetics in multiproduct systems (Figure 2). We were the first group to use validated quantitative and targeted mass spectrometry (UPLC-QqQ) to study lysine acetyltransferases (KATs) under steady state conditions, allowing us to monitor one substrate turning over into multiple products or acetylation states. This was a technical challenge due to the 100’s of samples that have to be quantitated for this analysis (typically 300 samples for one steady-state curve). We also needed to develop a new theoretical framework for the analysis of one substrate to multiple product turnover.

Specificity and selectivity of p300 and CBP are different and regulated by acetyl-CoA concentration.  These two KATs share ~90% identity in their functional domains but are often correlated with different diseases and developmental responses. Thus, understanding their unique, site-specific activity has provided insight into how the function of the two proteins differ. The findings of our initial study were highlighted in F1000 for its importance in identifying unique differences in these two homologous enzymes. From this starting point we were then able to link acetyl-CoA concentration with the residue preferences of these enzymes (Figure 2). We proposed these enzymes potentially function through a monomeric cooperative model (manifesting as hysteresis), which predicts that at certain concentrations competitive inhibitors could activate the enzyme, resulting in a paradoxical activity. We were able to demonstrate this effect both in vitro and in vivo.

Histone chaperones can recognize histone PTMs, and direct crosstalk (Figure 3). Histone chaperones both interact with KATs and assemble and disassemble nucleosomes, strongly suggesting that they could regulate which histone residue could be modified. Rtt109 is a yeast KAT which is an active site structural homolog of human p300 and CBP and forms a complex with the histone chaperone Vps75 (Nap1 family). In vivo, Rtt109 functions with a second histone chaperone Asf1, to facilitate the acetylation of H3K56, which is critical for DNA damage in yeast. Early biochemical data was able to show that Asf1 could increase the activity of Rtt109-Vps75 but did not result in a change in selectivity. One of the major goals was to determine the mechanism of this change in selectivity observed in vivo. To test our hypothesis that pre-existing acetylation influences the selectivity of Rtt109-Vps75 we made a library of singly acetylated histones using a strain of E. coli containing an orthogonal N^ε-acetyllysyl-tRNA synthetase/tRNA_CUA pair. Using this library with Rtt109-Vps75 under steady-state conditions in the presence and absence of Asf1 we found that only the combination of Asf1 and H3K14ac/H4 could completely change the selectivity of Rtt109-Vps75. The ability of Asf1 to alter the selectivity of Rtt109-Vps75 only in the presence of H3K14ac, suggest that Asf1 was not only a histone chaperone but an acetyl reader. We identified an acidic patch on the backside of Asf1, which we propose ‘reads’ the acetylation state of the H3 tail (Figure 3) and from these data we proposed a new model that reconciles previously conflicting genetic and biochemical data.

The regulation of hydroxylated estrogens in cancer. The oxidative metabolite of estrogen, 4-hydroxyestrogen, could be oncogenic, while 2-hydroxyestrogen is suggested to be anti-proliferative. Both of these products could be further broken down by the P450 enzymes CYP1A1 and 1B1 to produce semiquinones and/or quinones causing DNA damage. Preventing the formation of quinones or semiquinones is the enzyme catechol-O-methyltransferase (COMT) can convert the catechol estrogens to methoxyestrogens. To begin to understand the role of these enzymes in oncogenesis we recently developed a UPLC-QqQ based assay for estrogens and their derivatives. Our assay improves on the sensitivity and speed of existing assays allowing for both in vivo (human urine and cell culture) and in vitro quantitation (Figure 4).