Cells are able to regulate the activity of their genes in response to different cues. Genetic information is encoded in DNA and one way to regulate gene activity is to modify the DNA by attaching chemical “epigenetic” markers to it. When a cell divides, these epigenetic markers can be inherited by the daughter cells so that they share the same patterns of gene activity as the parent cell. When the DNA of the parent cell is copied prior to cell division, the epigenetic markers are also copied onto the new DNA. Mistakes in this process are linked to a wide range of diseases in humans, such as cancer and neurological disorders.

One type of epigenetic marker is known as a methyl tag and it is added to DNA by certain enzymes in a process called DNA methylation. A protein called UHRF1 is required for human cells to inherit patterns of DNA methylation through cell division. This protein binds to newly copied DNA that lacks some methyl tags as well as to another protein associated with DNA called histone H3. UHRF1 modifies histone H3 by attaching a small protein molecule called ubiquitin to it. This helps to recruit a DNA methylation enzyme to place methyl tags on the newly copied DNA. However, it was not clear how the various properties of UHRF1 allow it to control how DNA methylation is inherited.

Harrison et al. addressed this question by studying purified proteins and DNA fragments outside of living cells. The results show that UHRF1 binding to DNA and histone H3 work together to bring UHRF1 to the sites on DNA that require methylation. Further experiments revealed that the methylation pattern on newly copied DNA is able to activate the ability of UHRF1 to place ubiquitin on histone H3.

The findings of Harrison et al. reveal a new mechanism by which dividing cells control how DNA methylation is inherited by their daughter cells. A future challenge will be to find out how attaching ubiquitin to histone H3 activates DNA methylation.

Epigenetic regulation of chromatin architecture and gene expression is driven, in large part, by proteins that write, erase, and read histone post-translational modifications (PTMs) and DNA methylation. These proteins and their complexes are often comprised of multiple regulatory domains, permitting intricate mechanisms that govern allosteric control of enzymatic activity and multivalent engagement of chromatin through one or more reader modules (Du et al., 2015; Musselman et al., 2012; Noh et al., 2016; Rothbart and Strahl, 2014; Ruthenburg et al., 2007; Su and Denu, 2016). The E3 ubiquitin ligase UHRF1 (ubiquitin-like, containing PHD and RING finger domains 1) is one such multi-domain epigenetic regulator (Figure 1A) that plays a central role in DNMT1-directed DNA methylation maintenance during DNA replication (Bostick et al., 2007; Sharif et al., 2007). It does so in part through the reader activity of its linked TTD-PHD (tandem Tudor and plant homeodomain) towards the N-terminus of histone H3 when it is di- and tri-methylated at lysine 9 (H3K9me2/me3) (Arita et al., 2012; Rothbart et al., 2013, 2012), and through RING (really interesting new gene) domain-mediated catalysis of H3K18 and H3K23 ubiquitylation that promotes DNMT1 association with H3 (Nishiyama et al., 2013; Qin et al., 2015).

Figure 1 with 2 supplements see all

Download asset Open asset

The SRA (SET and RING-associated domain) of UHRF1 binds DNA with modest selectivity towards hemi-methylated CpG dinucleotides (HeDNA) (Arita et al., 2008; Avvakumov et al., 2008; Hashimoto et al., 2008) and has also been implicated in DNA methylation regulation. However, as previously studied mutations and deletions of the SRA disrupt DNA interaction regardless of DNA methylation status (Liu et al., 2013; Sharif et al., 2007), the specific contribution of HeDNA recognition to this epigenetic regulatory process has not been defined. We therefore sought to gain insight into the function of HeDNA recognition through the UHRF1 SRA domain and determine the relationship between the enzymatic and histone- and DNA-binding activities of this multi-domain epigenetic regulator.

We first produced recombinant full-length human UHRF1 and quantified the interaction of this protein with double-stranded DNA oligonucleotides containing a single unmodified (UnDNA), hemi-methylated (HeDNA), or symmetrically methylated (SyDNA) CpG dinucleotide by fluorescence polarization (FP). UHRF1 displayed a 10- to 20-fold preference for HeDNA over UnDNA, and a 5- to 10-fold preference for HeDNA over SyDNA (Figure 1B; left panel). We also confirmed the binding preferences of two previously characterized single amino acid substitutions to the SRA domain (Avvakumov et al., 2008). G448D (DNA^mut) disrupts all DNA-binding (Figure 1B; middle panel) by installing a negatively charged residue at a position that contacts the DNA backbone (Figure 1—figure supplement 1), and N489A (HeDNA^mut), harbored within the NKR finger that contacts the unmethylated cytosine opposite the methylated base (Figure 1—figure supplement 1), disrupts only HeDNA-sensing (Figure 1B; right panel). We also observed that the DNA binding affinity of UHRF1 was exquisitely sensitive to small perturbations in salt concentration; we measured a nearly 500-fold affinity difference for HeDNA between 50 mM and 150 mM NaCl (Figure 1—figure supplement 2).

We next used a previously developed genetic complementation system in HeLa cells (Rothbart et al., 2012) to determine the contribution of DNA-binding and HeDNA-sensing to UHRF1 function in DNA methylation maintenance. Consistent with our previous observations (Rothbart et al., 2013, 2012), global DNA methylation levels were significantly reduced following stable knockdown of endogenous UHRF1 by shRNA (Figure 1C). DNA methylation was restored by reintroduction of a wild-type UHRF1 transgene, but like mutations that disrupt histone interaction through the PHD finger (H3^mut; Figure 1—figure supplement 1), E3 ubiquitin ligase activity (RING^mut) and DNA binding (DNA^mut), HeDNA^mut could not rescue DNA methylation loss in cells despite retaining its ability to bind DNA (Figure 1B–C). These results demonstrate that in addition to the well-appreciated roles of histone-binding and ubiquitin ligase activity to the DNA methylation regulatory function of UHRF1 (Nishiyama et al., 2013; Rothbart et al., 2013, 2012), hemi-methylated DNA sensing is critical for DNA methylation maintenance.

Notably, unlike H3^mut and DNA^mut, wild-type and HeDNA^mut bound to bulk chromatin biochemically fractionated from HeLa cells (Figure 1D). Collectively, these findings suggest that the histone- and DNA-binding domains of UHRF1 are performing complementary functions to target UHRF1 to chromatin, and that HeDNA recognition provides an additional regulatory layer in the DNA methylation program.

To test this hypothesis, we first sought to determine whether the independently characterized DNA- and histone-binding activities of UHRF1 might function in concert. In agreement with previous analyses of the isolated TTD-PHD (Rothbart et al., 2013, 2012), full-length UHRF1 displayed a preference for H3K9me3 peptides over unmodified H3 peptides (H3K9un) (Figure 2A; top panel, see also Supplementary file 1). No binding was observed for H3^mut (Figure 2—figure supplement 1A) or for wild-type protein binding to peptides containing an N-terminal 5-carboxyfluorecin (FAM) probe to block PHD engagement (Figure 2A; bottom panel). Performing these assays with full-length UHRF1 allowed us to ask whether DNA binding affects H3 peptide binding and vice versa. Histone binding measurements in the presence of 10 μM unlabeled HeDNA, SyDNA, or UnDNA enhanced the interaction with C-terminal FAM-labeled H3K9me3 and H3K9un peptides (Figure 2A; top panel). HeDNA did not enhance binding to N-terminal FAM-labeled peptides, indicating that the multivalent interaction of the TTD-PHD with a single H3 peptide (Rothbart et al., 2013, 2012) remained intact. Reciprocally, DNA-binding measurements in the presence of 10 μM unlabeled H3K9me3 peptide enhanced DNA binding affinity 5–10 fold irrespective of the methylation status on DNA (Figure 2B–C and Figure 2—figure supplement 1B-C,E). Collectively, these experiments demonstrate that the histone- and DNA-binding modules of UHRF1 are regulated by reciprocal positive allostery.

Figure 2 with 1 supplement see all

Download asset Open asset

In agreement with the multivalent histone engagement model of the UHRF1 TTD-PHD, the extent to which H3 peptides augmented the interaction of UHRF1 with DNA was dependent on the epigenetic signature on H3. H3K9me3 peptide showed a three-fold enhancement of DNA binding over H3K9un, and asymmetric di-methylation of arginine 2 (H3R2me2a), which blocks the UHRF1 PHD interaction with H3 (Rajakumara et al., 2011), did not enhance DNA binding (Figure 2B). Consistently, H3^mut completely perturbed the ability of an H3K9me3 peptide to positively regulate DNA binding, a previously characterized double mutation to the linker connecting the TTD-PHD that uncouples multivalent engagement to H3 (Linker^mut) (Arita et al., 2012; Rothbart et al., 2013) exhibited a weaker enhancement of DNA binding in the presence of peptide than wild-type, and DNA^mut remained unable to bind DNA in the presence of H3K9me3 (Figure 2C and Figure 2—figure supplement 1D–E). Conversely, the DNA binding affinity of HeDNA^mut was still enhanced by H3K9me3, although HeDNA^mut could not discriminate between UnDNA and HeDNA (Figure 2C), providing a biochemical basis for HeDNA^mut retention on chromatin (Figure 1D).

The observed positive allostery between DNA- and histone-binding suggested the possibility of a direct physical interaction between the SRA and TTD-PHD domains of UHRF1. Consistent with this hypothesis, the UHRF1 TTD-PHD associated with the SRA and SRA-RING in pull-down experiments (Figure 3A), and this association was perturbed in the presence of DNA, irrespective of methylation status (Figure 3B, left). SRA-RING DNA^mut maintained interaction with the TTD-PHD in the presence of DNA (Figure 3B, right). However, an H3K9me2 peptide did not inhibit the interaction between the SRA and the TTD-PHD (Figure 3C). These results suggest that the DNA-binding surface of the SRA contributes to an intramolecular interaction in a manner non-competitive with histone binding. To ensure that the allostery observed was due to an intramolecular rearrangement and not through oligomerization, we characterized UHRF1 in the presence and absence of ligands with several biophysical techniques. Indeed, UHRF1 remained monomeric and in good agreement with the expected molecular weight as measured by analytical size exclusion chromatography, dynamic light scattering, and atomic force microscopy (Figure 3D–F).

Figure 3

Download asset Open asset

Collectively, these results show that the histone- and DNA-binding domains of UHRF1 interact and that general DNA binding releases this physical association. The data further suggest that ligand-induced intramolecular rearrangement of UHRF1 domain connectivity results in high-affinity retention of UHRF1 on chromatin through positive regulation of histone- and DNA-binding activities. This model is generally consistent with a recent report published during the preparation of this manuscript (Fang et al., 2016), which shows that HeDNA enhances histone interaction and suggests a closed-to-open conformational change in UHRF1 intramolecular architecture upon ligand binding. However, two key differences between our findings are that we show reciprocal positive allostery between the histone- and DNA-binding domains of UHRF1, and that general DNA interaction (regardless of methylation state) can displace the TTD-PHD domain and enhance histone interaction. We note that the conclusions from Fang et al. relied upon the interpretation of qualitative in-solution pull-down experiments conducted with a 2:1 DNA:UHRF1 ratio (see Fang et al., 2016), whereas we used quantitative FP to measure relative binding affinities and included unlabeled ligands in our experiments at concentrations at least five-fold over their measured K_d values to ensure saturation.

Since UHRF1 ubiquitin ligase activity is required to support DNA methylation yet is dispensable for bulk chromatin interaction (see Figure 1C–D and Nishiyama et al., 2013; Qin et al., 2015), we hypothesized that there may be a functional link between HeDNA-binding and UHRF1 ligase activity. To begin testing this hypothesis, we in vitro reconstituted UHRF1-mediated ubiquitylation using recombinant UHRF1, H3 peptides, Flag-tagged ubiquitin, and the ubiquitin conjugation enzymes E1 (Uba1) and E2 (UbcH5c) (for a review of the mechanism of ubiquitin activation see Schulman, 2011). Surprisingly, we observed robust ubiquitylation of an H3_1-32K9me2 peptide (Supplementary file 1) in a 20 min end-point assay in the presence of HeDNA (Figure 4A). Neither apo-UHRF1, SyDNA, nor UnDNA could stimulate this activity at concentrations well above their measured K_d values (Figure 4A), despite the ability of these DNAs to positively regulate histone binding (Figure 2A). Consistent with our measured K_d for HeDNA (Figure 1B), we observed reduced ubiquitylation activity as HeDNA concentration fell below 300 nM (Figure 4A). To our knowledge, this is the first demonstration that a DNA-protein interaction, and in particular an epigenetic modification, directly regulates enzymatic activity of an E3 ubiquitin ligase.

Figure 4 with 1 supplement see all

Download asset Open asset

To further characterize HeDNA-stimulated UHRF1 ubiquitylation, we compared the rate of UHRF1 enzymatic activity on itself (auto-ubiquitylation measurements are often used as a proxy to monitor E3 ligase activity) and an H3_1-32K9me2 peptide substrate. We measured a 2.5-fold rate enhancement of UHRF1 auto-ubiquitylation in the presence of HeDNA vs. UnDNA, both in the presence or absence of H3_1-32K9me2 peptide (Figure 4B and Figure 4—figure supplement 1A). In sharp contrast, the rate of HeDNA-stimulated H3_1-32K9me2 mono-ubiquitylation was stimulated by more than 100-fold over the rate obtained with UnDNA (Figure 4B). Comparing the rate of activity on UHRF1 substrates (peptide vs. self), the rate of auto-ubiquitylation was 20-fold faster than the rate of peptide ubiquitylation in the presence of UnDNA. Conversely, the rate of peptide ubiquitylation was seven-fold faster than the rate of auto-ubiquitylation in the presence of HeDNA (Figure 4B). Based on these observations, we propose that HeDNA-binding acts as an allosteric switch to enhance ubiquitylation of histone substrates.

Similar to peptide substrates, UHRF1 mono-, di-, and tri-ubiquitylation of purified HeLa mononucleosomes was stimulated by HeDNA (Figure 4—figure supplement 1B), confirming that the enhanced ubiquitylation activity of UHRF1 is relevant in the context of chromatin. In addition, when excess H3_1-15K9me2 peptide (Supplementary file 1) (which harbors the TTD-PHD binding site but not the published ubiquitin target lysines) was added to UHRF1 mononucleosome ubiquitylation assays, H3_1-15K9me2 effectively inhibited enzymatic activity towards mononucleosome substrates (Figure 4C). In contrast, HeDNA concentrations as high as 40 μM did not block mononucleosome ubiquitylation, consistent with its role as an activator of UHRF1 E3 ligase activity. These results suggest that the N-terminus of H3 is the primary binding site for substrate recognition through the TTD-PHD and that DNA interaction can occur in trans to the nucleosome being targeted for ubiquitylation.

To further investigate the role of UHRF1 reader domain functions in ligase activity, we tested the previously described H3^mut, Linker^mut, DNA^mut, HeDNA^mut (Figure 1A) and D469G (Avvakumov et al., 2008) mutants in ubiquitylation assays using H3_1-32K9me2 (Figure 4—figure supplement 1C) and H3_1-43K9un peptides as substrate (Figure 4D). Reacting UHRF1 with peptide substrates, we observed low ubiquitin ligase activity in the absence of HeDNA, while HeDNA binding permitted robust formation of mono-, di-, and tri-ubiquitylated H3 peptides (Figure 4D and Figure 4—figure supplement 1C). Characterizing DNA-, HeDNA-, and histone-binding loss-of-function UHRF1 mutants in ubiquitylation assays revealed defects in HeDNA-dependent H3 ubiquitylation, with the exception of the previously reported SRA loss-of-function mutant (D469G) (Avvakumov et al., 2008), that exhibited wild-type binding to HeDNA in our assays (Figure 4—figure supplement 1D). Ubiquitylation defects observed for H3^mut and Linker^mut confirmed a critical role for the TTD-PHD as the substrate-binding domain for HeDNA-dependent H3 ubiquitylation and demonstrated that multivalent cis engagement of H3K9me3 by the UHRF1 TTD-PHD is required for proper ubiquitylation (Figure 4D and Figure 4—figure supplement 1C). In addition, complete loss of HeDNA-dependent ubiquitylation for DNA^mut UHRF1 and the absence of multi-ubiquitylated H3 for HeDNA^mut UHRF1 further support the role of DNA binding and HeDNA recognition to fully activate UHRF1 ubiquitin ligase activity. The histone ubiquitylation defects observed for these loss-of-function mutants further highlights the interplay between UHRF1 functional domains to support proper UHRF1 ubiquitin ligase activity.

There is a growing appreciation for the role of allosteric regulation of RING E3 ubiquitin ligase activity in the field of ubiquitin biology (Vittal et al., 2015). Most often, the regulation of RING E3 ligases is accomplished through modulation of the E3 affinity for an E2-ub (thioesterified E2-ubiquitin) conjugate. Auto-inhibition release is the primary mechanism observed to date, in which steric occlusion of the RING domain prevents E3 interaction with the E2-ub until the E3 receives an appropriate release signal. Examples of RING auto-inhibition release include neddylation of Cullins (Duda et al., 2008; Saha and Deshaies, 2008), phosphorylation of Cbl (Dou et al., 2012), and substrate/peptide mimetic binding to inhibitor of apoptosis 1 (Dueber et al., 2011). Taking into consideration the above-described allosteric regulatory mechanism of E3 ligase activity, we first sought to determine whether the interaction with HeDNA could affect association with E2-ub. Using isothermal titration calorimetry (ITC) and nuclear magnetic resonance (NMR) spectroscopy, we monitored the interaction between UHRF1 and E2-N-ub (isopeptide-linked C85K E2-ubiquitin conjugate) or ¹⁵N-E2-O-ub (oxyesterified C85S E2-ubiquitin conjugate), respectively. Surprisingly, neither ITC nor NMR spectrum intensity loss measurements indicated a change in affinity for the E2 conjugates in the absence or presence of HeDNA (Figure 5A–B and Figure 5—figure supplement 1). In addition, we readily observed UHRF1 auto-ubiquitylation in the presence of HeDNA and UnDNA (Figure 4B and Figure 4—figure supplement 1A), indicating the RING domain of UHRF1 could productively interact with E2-ub regardless of the methylation status of the bound DNA. Notably, upon E3 binding to the conjugated E2, NMR resonances belonging to the conjugated ubiquitin did not suffer as great a loss in intensity (Figure 5B and Figure 5—figure supplement 1B–C), indicating that ubiquitin retained its dynamics when bound to the E3. This observation suggests that UHRF1 binding does not promote closed E2-ub states as strongly as other canonical RING domain E3’s (Christensen et al., 2007; Pruneda et al., 2011a, 2012).

Figure 5 with 1 supplement see all

Download asset Open asset

Recently, another mechanism of allosteric regulation of RING activity has been described where a ligand, Poly-ADP-ribose (PAR), induces a conformational change directly in the E3 RNF146 RING domain. This alternative RING conformation stabilizes the E2-ub/RING complex, thereby enhancing ubiquitin discharge from the conjugated E2 (DaRosa et al., 2015). To test whether HeDNA-induced UHRF1 ubiquitin ligase activity enhanced ubiquitin discharge from the conjugated E2, we performed single turnover ubiquitylation assays where purified E2-ub served as the ubiquitin donor and excess free lysine was present as a proxy ubiquitin substrate. The rate of ubiquitin discharge from E2 in these assays (monitored by the loss of the E2-ub and the appearance of free E2) showed only a modest increase in the reactivity of the conjugate in the presence of HeDNA (Figure 5C). These results suggest that HeDNA-dependent activation of UHRF1 RING activity does not occur through enhancement of the intrinsic rate of ubiquitin discharge from the E2 to non-specific lysine sidechains.

Remarkably, when an H3_1-20 peptide (Supplementary file 1) was added to ubiquitin discharge reactions, we observed rapid conversion of E2-ub to E2 (Figure 5C). Additionally, we observed the appearance of a band corresponding to ubiquitylated H3_1-20 (Figure 5C). Notably, we also observed a decrease in the amount of free ubiquitin and UHRF1 auto-ubiquitylation formed when peptide was present, presumably because more ubiquitin was being transferred to H3 (Figure 5C). Thus, even under conditions where free lysine was in great excess, ubiquitin was transferred rapidly and preferentially to H3 substrate in the presence of HeDNA.

To determine whether activation of UHRF1 occurs upon substrate binding (i.e., substrate-assisted activation), we performed single turnover assays in the presence of H3_1-20, H3_1-20K14acK18ac, and H3_1-20K9acK14acK18ac (Figure 5D, see also Supplementary file 1). We previously demonstrated the interaction of the UHRF1 TTD-PHD with these potential substrates by peptide microarray (Rothbart et al., 2013) and reasoned that the acetylated peptides would maintain interaction with the TTD-PHD but would be unable to accept ubiquitin. Neither the H3_1-20K14acK18ac nor H3_1-20K9acK14acK18ac were capable of being modified with ubiquitin (Figure 5D). Additionally, rapid E2-ub depletion was only observed in the sample containing H3_1-20 (Figure 5D), indicating that substrate binding alone does not enhance the E3 ligase activity of UHRF1 in the presence of HeDNA. Consistent with our previous single turnover results (Figure 5C), reactions that contained unblocked lysines on H3_1-20 accumulated less auto-ubiquitylated UHRF1 and free ubiquitin compared to assays with no peptide or with the acetylated H3 peptides (Figure 5D), supporting a model where HeDNA alters the substrate specificity of UHRF1 ubiquitylation. It is worth noting that in all single and multiple turnover assays performed, virtually no ubiquitylated species of H3 were observed in assays that lacked HeDNA. Taken together, these data demonstrate that HeDNA stimulates UHRF1 ubiquitin ligase activity through a novel regulatory mechanism and suggest that HeDNA binding serves as an allosteric switch that directs the TTD-PHD bound H3 substrate to the E2-ub active site for transfer.

We next used high-resolution mass spectrometry to further characterize the histone lysine specificity of HeDNA-stimulated UHRF1 ubiquitylation. HeLa mononucleosomes were reacted with UHRF1 in the presence of HeDNA or UnDNA for 2 hr using the enrichment strategy depicted in Figure 6A. Since histone proteins are highly basic, propionic anhydride was used to chemically modify free lysines and facilitate the identification of peptide fragments (Garcia et al., 2007). We first identified peptides that were enriched in the HeDNA sample relative to the UnDNA sample, and the only histone peptides that were enriched greater than ten-fold were derived from H3 (Supplementary file 2) (see Materials and methods for information about normalization and quantification). Consistent with our findings that HeDNA stimulates UHRF1 ubiquitin ligase activity, ubiquitin remnants on H3 peptides were heavily enriched in samples reacted in the presence of HeDNA relative to UnDNA (Figure 6B and Figure 6—figure supplement 1,2). In addition to the previously identified sites H3K18 and H3K23, we also found H3K14ub, H3K27ub, and H3K36ub heavily enriched in the presence of HeDNA (Figure 6B, Figure 6—figure supplement 1,2). We further identified abundant H3K18ub and H3K23ub (multi-ubiquitylation) on the same peptide. These results are consistent with laddering observed for H3ub in our experiments with synthetic peptides (see Figure 4B) and recombinant and native mononucleosomes (see Figure 4C and Figure 4—figure supplement 1C), as well as immunoblots from HeLa cells in previous studies (Nishiyama et al., 2013; Qin et al., 2015). H3K18ub was the most abundant ubiquitylated peptide based upon spectral counts (greater than 15-fold more than any other site), and H3K23ub was only observed in the context of H3K18ub (Figure 6B). Collectively these results suggest that H3K18 is the preferred ubiquitylation site for UHRF1, but that UHRF1 can target a number of lysines on the H3 tail.

Figure 6 with 3 supplements see all

Download asset Open asset

Consistent with our model of HeDNA altering UHRF1 substrate preferences, we also observed changes to the sites of UHRF1 auto-ubiquitylation in the presence of HeDNA (Figure 6—figure supplement 2C–E). In particular, we identified a nine-fold enrichment of UHRF1 K303ub, a solvent exposed lysine in the PHD near the C-terminus of a bound H3 peptide (Figure 6—figure supplement 2D). This region may represent the target zone for HeDNA-dependent ubiquitylation where the RING domain would be in proximity to this region. Accordingly, we observed an additional auto-ubiquitylated UHRF1 band in the presence of HeDNA compared to that observed with UnDNA (Figure 6—figure supplement 2E), corroborating our mass spectrometry results.

We also identified several histone PTMs that co-occurred with ubiquitylated H3 peptides, including all three states of H3K9 methylation (Figure 6B). Additionally, several of the most enriched H3 peptides not containing ubiquitin remnants also contained H3K9me2 (Supplementary file 2). We confirmed this epigenetic link by immunoblotting HeLa mononucleosomes ubiquitylated by UHRF1. Ubiquitylated H3 was detected in the presence of HeDNA (but not UnDNA) on nucleosomes marked with H3K9me3, but not on nucleosomes marked with H3K9acK14ac (Figure 6C). In addition, titrating recombinant human histones H3, H2A, and H2B into ubiquitylation assays revealed that while UHRF1 could modify H2A and H2B at concentrations above 5 μM, H3 could be modified at sub-micromolar concentrations, and H3K9me2 protein (synthesized by native chemical ligation) could be modified at even lower concentrations (Figure 6—figure supplement 3). Taken together, these findings strongly support the role of H3K9 methylation in directing UHRF1 ubiquitylation to adjacent lysine residues in the presence of HeDNA. Other histone PTMs co-occurring on ubiquitylated peptides were: H3K23ac, also identified in another study (Qin et al., 2015); H3K27me2, which often co-occurs with H3K9me2/me3 and is considered a hallmark of facultative heterochromatin (Boros et al., 2014); H3K36ac, and H3K37me3 (Figure 6B and Figure 6—figure supplement 1,2). However, future studies will be required to dissect the biological significance of these PTM combinations to ubiquitin ligase-dependent UHRF1 function.

Our studies define an orchestrated sequence of histone- and DNA-binding events targeting UHRF1 to chromatin and identify a key regulatory mechanism controlling DNA methylation inheritance through UHRF1 E3 ligase activation following recognition of HeDNA. This mechanism is consistent with the observation that UHRF1-dependent H3 ubiquitylation accumulates in S-phase when HeDNA intermediates are generated behind replicating DNA polymerase (Nishiyama et al., 2013; Qin et al., 2015). Building on recent studies connecting H3 ubiquitylation to DNMT1 recruitment (Nishiyama et al., 2013; Qin et al., 2015), we propose a model where UHRF1 is targeted to chromatin through its coordinated histone and DNA reading activities (Figure 7A). When UHRF1 encounters HeDNA, H3 ubiquitylation serves as a mechanism to facilitate the recruitment of DNMT1 to replicating regions of the genome to copy parental DNA methylation patterns (Figure 7B).

Figure 7 with 1 supplement see all

Download asset Open asset

Structural characterization of the UHRF1 SRA bound to HeDNA (Arita et al., 2008; Avvakumov et al., 2008; Hashimoto et al., 2008) and cellular localization of UHRF1 with DNMT1 and PCNA (proliferating cell nuclear antigen) at replicating heterochromatic foci (Bostick et al., 2007; Sharif et al., 2007) contribute to the model in which UHRF1 ubiquitylation of S-phase chromatin is mediated through HeDNA recognition. While it remains to be seen whether uncoupling UHRF1 from HeDNA recognition changes its residence genome-wide, our studies show HeDNA sensing is not required to target UHRF1 to bulk chromatin. Rather, coordinated recognition of H3 and DNA, independent of HeDNA discrimination, drives chromatin interaction. We propose that the avidity resulting from sub-μM affinities of UHRF1 for both DNA and H3 peptides through reciprocal positive allostery provides a biochemical basis by which UHRF1 is exclusively localized on chromatin. This may also explain why small perturbations to histone binding affinity through the TTD-PHD (e.g., Linker^mut and TTD aromatic cage mutation [Rothbart et al., 2013, 2012]) so dramatically affect chromatin targeting of this protein.

Our studies define the UHRF1 TTD-PHD as the substrate-binding domain for HeDNA-stimulated ubiquitylation, further demonstrating a functional role for the coordinated recognition of H3K9me2/me3 and HeDNA in UHRF1 ubiquitylation. UHRF1 appears to be versatile in targeting lysines for ubiquitylation on the H3 tail (Figure 6B): this ability may be related to the complexities of PTM patterning found on this region of the H3 tail (Young et al., 2009) and the necessity to promote efficient recruitment of DNMT1 to differentially modified chromatin environments. In addition, UHRF1 histone ubiquitylation may serve other roles in DNA related processes (i.e., DNA repair) (Liang et al., 2015; Tian et al., 2015; Zhang et al., 2016). Further studies are necessary to examine the biological consequence of different patterns of H3 ubiquitylation by UHRF1 and their relationship to pre-existing histone PTM signatures.

Important to note are studies on the enzymology of DNMT1 activity that show the enzyme has an intrinsic preference for HeDNA substrates (Goyal et al., 2006) and methylates in a processive manner (Bestor and Ingram, 1983; Hermann et al., 2004). Our studies define a major function for HeDNA, beyond direct stimulation of DNMT1 activity, in the regulation of UHRF1 histone ubiquitylation and DNA methylation inheritance. Considering DNMT1 behavior on oligonucleotide substrates, it is intriguing to speculate that UHRF1 functions to provide a nucleation event for DNMT1 recruitment to chromatin. Future studies mapping the genome-wide distribution of UHRF1-directed H3 ubiquitylation in relation to DNA methylation patterning will clarify the relationship between UHRF1 and DNMT1 activities.

How might HeDNA binding alter the substrate preference of UHRF1 directed ubiquitylation? We speculate that HeDNA is bound by the UHRF1 SRA in a manner that positions the RING in proximity to the H3 binding region of UHRF1. RING activity towards H3 may be conformationally restricted when UHRF1 binds SyDNA or UnDNA (Figure 7). Consistent with this hypothesis, structures of the SRA domain from mouse UHRF1 (Np95) show this domain can adopt different conformations bound to UnDNA and HeDNA (Figure 7—figure supplement 1) (Hashimoto et al., 2008). Also the NKR finger of the UHRF1 SRA, which harbors the HeDNA^mut, adopts a highly ordered conformation upon HeDNA binding (Avvakumov et al., 2008; Hashimoto et al., 2008). This allows for pseudo-base pairing to the exposed guanosine nucleotide (Figure 1—figure supplement 1), and we hypothesize this stable finger conformation is critical for HeDNA-stimulated H3 ubiquitylation. Unfortunately, efforts to crystallize the enzymatically active conformation of UHRF1 proved unsuccessful. Determining the active conformation of UHRF1 will be an important step in further understanding the regulation imparted by HeDNA.

Why might ubiquitin be an ideal PTM to accompany a temporally controlled process like replication-coupled DNA methylation? Ubiquitin itself is a functional protein domain capable of participating a wide variety of protein-protein interactions (Harrison et al., 2016) and can sterically occlude surfaces, as has been proposed for H2BK120ub in the formation of a productive complex with DOT1L (Zhou et al., 2016). Ubiquitin modifications are also dynamic and can be rapidly removed by deubiquitylases. In fact, recent analysis of ubiquitin turnover kinetics showed that the half-life of H2BK123ub in budding yeast is approximately one minute (Yumerefendi et al., 2016). Thus, discovering the identity of the deubiquitylase that removes H3 ubiquitylation may provide key insight into the dynamics of DNA methylation regulation at the level of histone ubiquitylation. The initial study implicating H3 ubiquitylation in the inheritance of DNA methylation indirectly suggested that USP7 may be responsible for this function through interaction with DNMT1 (Nishiyama et al., 2013). This is notable, as USP7 has also been shown to interact with UHRF1 (Zhang et al., 2015). Additionally, recent studies have tied USP7 to DNA replication (Lecona et al., 2016) and the maintenance of heterochromatin (Mungamuri et al., 2016), providing a biological link to replication-coupled inheritance of DNA methylation. However direct evidence of USP7 catalyzed deubiquitylation of H3 is lacking.

In conclusion, our study defines the relationship between UHRF1 histone-binding, DNA-binding, and ubiquitylation activities and connects HeDNA recognition to UHRF1 enzymatic function. Additionally, we characterize HeDNA as an active epigenetic mark that allosterically regulates UHRF1 ubiquitylation towards histone H3. More broadly, these finding provide a function for epigenetic patterning associated with UHRF1 beyond protein recruitment. We speculate that epigenetic mechanisms of multivalency and allostery are more widespread and add additional layers of complexity, specificity, and connectivity to chromatin recognition, modification patterning, and genome regulation.

1. 1. Joseph S Harrison
   2. Dennis Goldfarb
   3. Michael B Major
   4. Brian Kuhlman
   5. Scott B Rothbart

   (2016) Hemi-methylated DNA regulates DNA methylation inheritance through allosteric control of H3 ubiquitylation by UHRF1

   Publicly available at the Proteome Xchange (accession no: PXD003983).

   http://proteomecentral.proteomexchange.org/cgi/GetDataset?ID=PXD003983

1. 1. Arita K
   2. Ariyoshi M
   3. Tochio H
   4. Nakamura Y
   5. Shirakawa M

   (2008) Recognition of hemi-methylated DNA by the SRA protein UHRF1 by a base-flipping mechanism

   Nature 455:818–821.

   https://doi.org/10.1038/nature07249

   • Google Scholar
2. 1. Arita K
   2. Isogai S
   3. Oda T
   4. Unoki M
   5. Sugita K
   6. Sekiyama N
   7. Kuwata K
   8. Hamamoto R
   9. Tochio H
   10. Sato M
   11. Ariyoshi M
   12. Shirakawa M

   (2012) Recognition of modification status on a histone H3 tail by linked histone reader modules of the epigenetic regulator UHRF1

   PNAS 109:12950–12955.

   https://doi.org/10.1073/pnas.1203701109

   • Google Scholar
3. 1. Avvakumov GV
   2. Walker JR
   3. Xue S
   4. Li Y
   5. Duan S
   6. Bronner C
   7. Arrowsmith CH
   8. Dhe-Paganon S

   (2008) Structural basis for recognition of hemi-methylated DNA by the SRA domain of human UHRF1

   Nature 455:822–825.

   https://doi.org/10.1038/nature07273

   • Google Scholar
4. 1. Bestor TH
   2. Ingram VM

   (1983) Two DNA methyltransferases from murine erythroleukemia cells: purification, sequence specificity, and mode of interaction with DNA

   PNAS 80:5559–5563.

   https://doi.org/10.1073/pnas.80.18.5559

   • Google Scholar
5. 1. Boros J
   2. Arnoult N
   3. Stroobant V
   4. Collet JF
   5. Decottignies A

   (2014) Polycomb repressive complex 2 and H3K27me3 cooperate with H3K9 methylation to maintain heterochromatin protein 1α at chromatin

   Molecular and Cellular Biology 34:3662–3674.

   https://doi.org/10.1128/MCB.00205-14

   • Google Scholar
6. 1. Bostick M
   2. Kim JK
   3. Estève PO
   4. Clark A
   5. Pradhan S
   6. Jacobsen SE

   (2007) UHRF1 plays a role in maintaining DNA methylation in mammalian cells

   Science 317:1760–1764.

   https://doi.org/10.1126/science.1147939

   • Google Scholar
7. 1. Boutet E
   2. Lieberherr D
   3. Tognolli M
   4. Schneider M
   5. Bairoch A

   (2007) UniProtKB/Swiss-Prot

   Methods in Molecular Biology 406:89–112.

   https://doi.org/10.1007/978-1-59745-535-0_4

   • Google Scholar
8. 1. Branigan E
   2. Plechanovová A
   3. Jaffray EG
   4. Naismith JH
   5. Hay RT

   (2015) Structural basis for the RING-catalyzed synthesis of K63-linked ubiquitin chains

   Nature Structural & Molecular Biology 22:597–602.

   https://doi.org/10.1038/nsmb.3052

   • Google Scholar
9. 1. Christensen DE
   2. Brzovic PS
   3. Klevit RE

   (2007) E2-BRCA1 RING interactions dictate synthesis of mono- or specific polyubiquitin chain linkages

   Nature Structural & Molecular Biology 14:941–948.

   https://doi.org/10.1038/nsmb1295

   • Google Scholar
10. 1. DaRosa PA
    2. Wang Z
    3. Jiang X
    4. Pruneda JN
    5. Cong F
    6. Klevit RE
    7. Xu W

    (2015) Allosteric activation of the RNF146 ubiquitin ligase by a poly(ADP-ribosyl)ation signal

    Nature 517:223–226.

    https://doi.org/10.1038/nature13826

    • Google Scholar
11. 1. Delaglio F
    2. Grzesiek S
    3. Vuister GW
    4. Zhu G
    5. Pfeifer J
    6. Bax A

    (1995) NMRPipe: a multidimensional spectral processing system based on UNIX pipes

    Journal of Biomolecular NMR 6:277–293.

    https://doi.org/10.1007/BF00197809

    • Google Scholar
12. 1. Dou H
    2. Buetow L
    3. Hock A
    4. Sibbet GJ
    5. Vousden KH
    6. Huang DT

    (2012) Structural basis for autoinhibition and phosphorylation-dependent activation of c-Cbl

    Nature Structural & Molecular Biology 19:184–192.

    https://doi.org/10.1038/nsmb.2231

    • Google Scholar
13. 1. Du J
    2. Johnson LM
    3. Jacobsen SE
    4. Patel DJ

    (2015) DNA methylation pathways and their crosstalk with histone methylation

    Nature Reviews Molecular Cell Biology 16:519–532.

    https://doi.org/10.1038/nrm4043

    • Google Scholar
14. 1. Duda DM
    2. Borg LA
    3. Scott DC
    4. Hunt HW
    5. Hammel M
    6. Schulman BA

    (2008) Structural insights into NEDD8 activation of cullin-RING ligases: conformational control of conjugation

    Cell 134:995–1006.

    https://doi.org/10.1016/j.cell.2008.07.022

    • Google Scholar
15. 1. Dueber EC
    2. Schoeffler AJ
    3. Lingel A
    4. Elliott JM
    5. Fedorova AV
    6. Giannetti AM
    7. Zobel K
    8. Maurer B
    9. Varfolomeev E
    10. Wu P
    11. Wallweber HJ
    12. Hymowitz SG
    13. Deshayes K
    14. Vucic D
    15. Fairbrother WJ

    (2011) Antagonists induce a conformational change in cIAP1 that promotes autoubiquitination

    Science 334:376–380.

    https://doi.org/10.1126/science.1207862

    • Google Scholar
16. 1. Fang J
    2. Cheng J
    3. Wang J
    4. Zhang Q
    5. Liu M
    6. Gong R
    7. Wang P
    8. Zhang X
    9. Feng Y
    10. Lan W
    11. Gong Z
    12. Tang C
    13. Wong J
    14. Yang H
    15. Cao C
    16. Xu Y

    (2016) Hemi-methylated DNA opens a closed conformation of UHRF1 to facilitate its histone recognition

    Nature Communications 7:11197.

    https://doi.org/10.1038/ncomms11197

    • Google Scholar
17. 1. Garcia BA
    2. Mollah S
    3. Ueberheide BM
    4. Busby SA
    5. Muratore TL
    6. Shabanowitz J
    7. Hunt DF

    (2007) Chemical derivatization of histones for facilitated analysis by mass spectrometry

    Nature Protocols 2:933–938.

    https://doi.org/10.1038/nprot.2007.106

    • Google Scholar
18. 1. Goyal R
    2. Reinhardt R
    3. Jeltsch A

    (2006) Accuracy of DNA methylation pattern preservation by the Dnmt1 methyltransferase

    Nucleic Acids Research 34:1182–1188.

    https://doi.org/10.1093/nar/gkl002

    • Google Scholar
19. 1. Harrison JS
    2. Jacobs TM
    3. Houlihan K
    4. Van Doorslaer K
    5. Kuhlman B

    (2016) UbSRD: the Ubiquitin structural relational database

    Journal of Molecular Biology 428:679–687.

    https://doi.org/10.1016/j.jmb.2015.09.011

    • Google Scholar
20. 1. Hashimoto H
    2. Horton JR
    3. Zhang X
    4. Bostick M
    5. Jacobsen SE
    6. Cheng X

    (2008) The SRA domain of UHRF1 flips 5-methylcytosine out of the DNA helix

    Nature 455:826–829.

    https://doi.org/10.1038/nature07280

    • Google Scholar
21. 1. Hermann A
    2. Goyal R
    3. Jeltsch A

    (2004) The Dnmt1 DNA-(cytosine-C5)-methyltransferase methylates DNA processively with high preference for hemimethylated target sites

    Journal of Biological Chemistry 279:48350–48359.

    https://doi.org/10.1074/jbc.M403427200

    • Google Scholar
22. 1. Johnson BA
    2. Blevins RA

    (1994) NMR View: A computer program for the visualization and analysis of NMR data

    Journal of Biomolecular NMR 4:603–614.

    https://doi.org/10.1007/BF00404272

    • Google Scholar
23. 1. Kessner D
    2. Chambers M
    3. Burke R
    4. Agus D
    5. Mallick P

    (2008) ProteoWizard: open source software for rapid proteomics tools development

    Bioinformatics 24:2534–2536.

    https://doi.org/10.1093/bioinformatics/btn323

    • Google Scholar
24. 1. Lecona E
    2. Rodriguez-Acebes S
    3. Specks J
    4. Lopez-Contreras AJ
    5. Ruppen I
    6. Murga M
    7. Muñoz J
    8. Mendez J
    9. Fernandez-Capetillo O

    (2016) USP7 is a SUMO deubiquitinase essential for DNA replication

    Nature Structural & Molecular Biology 23:270–277.

    https://doi.org/10.1038/nsmb.3185

    • Google Scholar
25. 1. Liang CC
    2. Zhan B
    3. Yoshikawa Y
    4. Haas W
    5. Gygi SP
    6. Cohn MA

    (2015) UHRF1 is a sensor for DNA interstrand crosslinks and recruits FANCD2 to initiate the Fanconi anemia pathway

    Cell Reports 10:1947–1956.

    https://doi.org/10.1016/j.celrep.2015.02.053

    • Google Scholar
26. 1. Lin S
    2. Garcia BA

    (2012) Examining histone posttranslational modification patterns by high-resolution mass spectrometry

    Methods in Enzymology 512:3–28.

    https://doi.org/10.1016/B978-0-12-391940-3.00001-9

    • Google Scholar
27. 1. Liu X
    2. Gao Q
    3. Li P
    4. Zhao Q
    5. Zhang J
    6. Li J
    7. Koseki H
    8. Wong J

    (2013) UHRF1 targets DNMT1 for DNA methylation through cooperative binding of hemi-methylated DNA and methylated H3K9

    Nature Communications 4:1563.

    https://doi.org/10.1038/ncomms2562

    • Google Scholar
28. 1. MacLean B
    2. Tomazela DM
    3. Shulman N
    4. Chambers M
    5. Finney GL
    6. Frewen B
    7. Kern R
    8. Tabb DL
    9. Liebler DC
    10. MacCoss MJ

    (2010) Skyline: an open source document editor for creating and analyzing targeted proteomics experiments

    Bioinformatics 26:966–968.

    https://doi.org/10.1093/bioinformatics/btq054

    • Google Scholar
29. 1. Mahto SK
    2. Howard CJ
    3. Shimko JC
    4. Ottesen JJ

    (2011) A reversible protection strategy to improve Fmoc-SPPS of peptide thioesters by the N-Acylurea approach

    ChemBioChem 12:2488–2494.

    https://doi.org/10.1002/cbic.201100472

    • Google Scholar
30. 1. McIlwain S
    2. Tamura K
    3. Kertesz-Farkas A
    4. Grant CE
    5. Diament B
    6. Frewen B
    7. Howbert JJ
    8. Hoopmann MR
    9. Käll L
    10. Eng JK
    11. MacCoss MJ
    12. Noble WS

    (2014) Crux: rapid open source protein tandem mass spectrometry analysis

    Journal of Proteome Research 13:4488–4491.

    https://doi.org/10.1021/pr500741y

    • Google Scholar
31. 1. Mungamuri SK
    2. Qiao RF
    3. Yao S
    4. Manfredi JJ
    5. Gu W
    6. Aaronson SA

    (2016) USP7 enforces heterochromatinization of p53 target promoters by protecting SUV39H1 from MDM2-mediated degradation

    Cell Reports 14:2528–2537.

    https://doi.org/10.1016/j.celrep.2016.02.049

    • Google Scholar
32. 1. Musselman CA
    2. Lalonde ME
    3. Côté J
    4. Kutateladze TG

    (2012) Perceiving the epigenetic landscape through histone readers

    Nature Structural & Molecular Biology 19:1218–1227.

    https://doi.org/10.1038/nsmb.2436

    • Google Scholar
33. 1. Nishiyama A
    2. Yamaguchi L
    3. Sharif J
    4. Johmura Y
    5. Kawamura T
    6. Nakanishi K
    7. Shimamura S
    8. Arita K
    9. Kodama T
    10. Ishikawa F
    11. Koseki H
    12. Nakanishi M

    (2013) Uhrf1-dependent H3K23 ubiquitylation couples maintenance DNA methylation and replication

    Nature 502:249–253.

    https://doi.org/10.1038/nature12488

    • Google Scholar
34. 1. Noh KM
    2. Allis CD
    3. Li H

    (2016) Reading between the lines: "ADD"-ing histone and DNA methylation marks toward a new epigenetic "Sum"

    ACS Chemical Biology 11:554–563.

    https://doi.org/10.1021/acschembio.5b00830

    • Google Scholar
35. 1. Pruneda JN
    2. Stoll KE
    3. Bolton LJ
    4. Brzovic PS
    5. Klevit RE

    (2011) Ubiquitin in motion: structural studies of the ubiquitin-conjugating enzyme∼ubiquitin conjugate

    Biochemistry 50:1624–1633.

    https://doi.org/10.1021/bi101913m

    • Google Scholar
36. 1. Pruneda JN
    2. Stoll KE
    3. Bolton LJ
    4. Brzovic PS
    5. Klevit RE

    (2011b) Ubiquitin in motion: structural studies of the ubiquitin-conjugating enzyme∼ubiquitin conjugate

    Biochemistry 50:1624–1633.

    https://doi.org/10.1021/bi101913m

    • Google Scholar
37. 1. Pruneda JN
    2. Littlefield PJ
    3. Soss SE
    4. Nordquist KA
    5. Chazin WJ
    6. Brzovic PS
    7. Klevit RE
    8. Benirschke RC
    9. Thompson JR
    10. Nominé Y
    11. Wasielewski E
    12. Juranić N
    13. Macura S
    14. Hatakeyama S
    15. Nakayama KI
    16. Botuyan MV
    17. Mer G
    18. Bentley ML
    19. Corn JE
    20. Dong KC
    21. Phung Q
    22. Cheung TK
    23. Cochran AG

    (2012) Structure of an E3:E2~Ub complex reveals an allosteric mechanism shared among RING/U-box ligases

    Molecular Cell 47:933–942.

    https://doi.org/10.1016/j.molcel.2012.07.001

    • Google Scholar
38. 1. Qin W
    2. Wolf P
    3. Liu N
    4. Link S
    5. Smets M
    6. La Mastra F
    7. Forné I
    8. Pichler G
    9. Hörl D
    10. Fellinger K
    11. Spada F
    12. Bonapace IM
    13. Imhof A
    14. Harz H
    15. Leonhardt H

    (2015) DNA methylation requires a DNMT1 ubiquitin interacting motif (UIM) and histone ubiquitination

    Cell Research 25:911–929.

    https://doi.org/10.1038/cr.2015.72

    • Google Scholar
39. 1. Rajakumara E
    2. Wang Z
    3. Ma H
    4. Hu L
    5. Chen H
    6. Lin Y
    7. Guo R
    8. Wu F
    9. Li H
    10. Lan F
    11. Shi YG
    12. Xu Y
    13. Patel DJ
    14. Shi Y

    (2011) PHD finger recognition of unmodified histone H3R2 links UHRF1 to regulation of euchromatic gene expression

    Molecular Cell 43:275–284.

    https://doi.org/10.1016/j.molcel.2011.07.006

    • Google Scholar
40. 1. Ratcliff GC
    2. Erie DA

    (2001) A novel single-molecule study to determine protein--protein association constants

    Journal of the American Chemical Society 123:5632–5635.

    https://doi.org/10.1021/ja005750n

    • Google Scholar
41. 1. Rothbart SB
    2. Krajewski K
    3. Nady N
    4. Tempel W
    5. Xue S
    6. Badeaux AI
    7. Barsyte-Lovejoy D
    8. Martinez JY
    9. Bedford MT
    10. Fuchs SM
    11. Arrowsmith CH
    12. Strahl BD

    (2012) Association of UHRF1 with methylated H3K9 directs the maintenance of DNA methylation

    Nature Structural & Molecular Biology 19:1155–1160.

    https://doi.org/10.1038/nsmb.2391

    • Google Scholar
42. 1. Rothbart SB
    2. Dickson BM
    3. Ong MS
    4. Krajewski K
    5. Houliston S
    6. Kireev DB
    7. Arrowsmith CH
    8. Strahl BD

    (2013) Multivalent histone engagement by the linked tandem Tudor and PHD domains of UHRF1 is required for the epigenetic inheritance of DNA methylation

    Genes & Development 27:1288–1298.

    https://doi.org/10.1101/gad.220467.113

    • Google Scholar
43. 1. Rothbart SB
    2. Strahl BD

    (2014) Interpreting the language of histone and DNA modifications

    Biochimica Et Biophysica Acta 1839:627–643.

    https://doi.org/10.1016/j.bbagrm.2014.03.001

    • Google Scholar
44. 1. Ruthenburg AJ
    2. Li H
    3. Patel DJ
    4. Allis CD

    (2007) Multivalent engagement of chromatin modifications by linked binding modules

    Nature Reviews Molecular Cell Biology 8:983–994.

    https://doi.org/10.1038/nrm2298

    • Google Scholar
45. 1. Ruthenburg AJ
    2. Li H
    3. Milne TA
    4. Dewell S
    5. McGinty RK
    6. Yuen M
    7. Ueberheide B
    8. Dou Y
    9. Muir TW
    10. Patel DJ
    11. Allis CD

    (2011) Recognition of a mononucleosomal histone modification pattern by BPTF via multivalent interactions

    Cell 145:692–706.

    https://doi.org/10.1016/j.cell.2011.03.053

    • Google Scholar
46. 1. Saha A
    2. Deshaies RJ

    (2008) Multimodal activation of the ubiquitin ligase SCF by Nedd8 conjugation

    Molecular Cell 32:21–31.

    https://doi.org/10.1016/j.molcel.2008.08.021

    • Google Scholar
47. 1. Schulman BA

    (2011) Twists and turns in ubiquitin-like protein conjugation cascades

    Protein Science 20:1941–1954.

    https://doi.org/10.1002/pro.750

    • Google Scholar
48. 1. Sharif J
    2. Muto M
    3. Takebayashi S
    4. Suetake I
    5. Iwamatsu A
    6. Endo TA
    7. Shinga J
    8. Mizutani-Koseki Y
    9. Toyoda T
    10. Okamura K
    11. Tajima S
    12. Mitsuya K
    13. Okano M
    14. Koseki H

    (2007) The SRA protein Np95 mediates epigenetic inheritance by recruiting Dnmt1 to methylated DNA

    Nature 450:908–912.

    https://doi.org/10.1038/nature06397

    • Google Scholar
49. 1. Shogren-Knaak MA
    2. Fry CJ
    3. Peterson CL

    (2003) A native peptide ligation strategy for deciphering nucleosomal histone modifications

    Journal of Biological Chemistry 278:15744–15748.

    https://doi.org/10.1074/jbc.M301445200

    • Google Scholar
50. 1. Su Z
    2. Denu JM

    (2016) Reading the combinatorial histone language

    ACS Chemical Biology 11:564–574.

    https://doi.org/10.1021/acschembio.5b00864

    • Google Scholar
51. 1. Tian Y
    2. Paramasivam M
    3. Ghosal G
    4. Chen D
    5. Shen X
    6. Huang Y
    7. Akhter S
    8. Legerski R
    9. Chen J
    10. Seidman MM
    11. Qin J
    12. Li L

    (2015) UHRF1 contributes to DNA damage repair as a lesion recognition factor and nuclease scaffold

    Cell Reports 10:1957–1966.

    https://doi.org/10.1016/j.celrep.2015.03.038

    • Google Scholar
52. 1. Tropea JE
    2. Cherry S
    3. Waugh DS

    (2009) Expression and purification of soluble His(6)-tagged TEV protease

    Methods in Molecular Biology 498:297–307.

    https://doi.org/10.1007/978-1-59745-196-3_19

    • Google Scholar
53. 1. Vittal V
    2. Stewart MD
    3. Brzovic PS
    4. Klevit RE

    (2015) Regulating the regulators: recent revelations in the control of E3 ubiquitin ligases

    Journal of Biological Chemistry 290:21244–21251.

    https://doi.org/10.1074/jbc.R115.675165

    • Google Scholar
54. 1. Wenzel DM
    2. Stoll KE
    3. Klevit RE

    (2011) E2s: structurally economical and functionally replete

    Biochemical Journal 433:31–42.

    https://doi.org/10.1042/BJ20100985

    • Google Scholar
55. 1. Young NL
    2. DiMaggio PA
    3. Plazas-Mayorca MD
    4. Baliban RC
    5. Floudas CA
    6. Garcia BA

    (2009) High throughput characterization of combinatorial histone codes

    Molecular & Cellular Proteomics 8:2266–2284.

    https://doi.org/10.1074/mcp.M900238-MCP200

    • Google Scholar
56. 1. Yumerefendi H
    2. Lerner AM
    3. Zimmerman SP
    4. Hahn K
    5. Bear JE
    6. Strahl BD
    7. Kuhlman B

    (2016) Light-induced nuclear export reveals rapid dynamics of epigenetic modifications

    Nature Chemical Biology 12:399–401.

    https://doi.org/10.1038/nchembio.2068

    • Google Scholar
57. 1. Zhang ZM
    2. Rothbart SB
    3. Allison DF
    4. Cai Q
    5. Harrison JS
    6. Li L
    7. Wang Y
    8. Strahl BD
    9. Wang GG
    10. Song J

    (2015) An allosteric interaction links USP7 to deubiquitination and chromatin targeting of UHRF1

    Cell Reports 12:1400–1406.

    https://doi.org/10.1016/j.celrep.2015.07.046

    • Google Scholar
58. 1. Zhang H
    2. Liu H
    3. Chen Y
    4. Yang X
    5. Wang P
    6. Liu T
    7. Deng M
    8. Qin B
    9. Correia C
    10. Lee S
    11. Kim J
    12. Sparks M
    13. Nair AA
    14. Evans DL
    15. Kalari KR
    16. Zhang P
    17. Wang L
    18. You Z
    19. Kaufmann SH
    20. Lou Z
    21. Pei H

    (2016) A cell cycle-dependent BRCA1–UHRF1 cascade regulates DNA double-strand break repair pathway choice

    Nature Communications 7:10201.

    https://doi.org/10.1038/ncomms10201

    • Google Scholar
59. 1. Zhou L
    2. Holt MT
    3. Ohashi N
    4. Zhao A
    5. Müller MM
    6. Wang B
    7. Muir TW

    (2016) Evidence that ubiquitylated H2B corrals hDot1L on the nucleosomal surface to induce H3K79 methylation

    Nature Communications 7:10589.

    https://doi.org/10.1038/ncomms10589

    • Google Scholar

Author details

1. Joseph S Harrison

   1. Department of Biochemistry and Biophysics, University of North Carolina at Chapel Hill, Chapel Hill, United States
   2. Lineberger Comprehensive Cancer Center, University of North Carolina at Chapel Hill, Chapel Hill, United States

   Contribution

   JSH, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article

   Competing interests

   The authors declare that no competing interests exist.

2. Evan M Cornett

   Center for Epigenetics, Van Andel Research Institute, Grand Rapids, United States

   Contribution

   EMC, Conception and design, Acquisition of data, Analysis and interpretation of data

   Contributed equally with

   Dennis Goldfarb and Paul A DaRosa

   Competing interests

   The authors declare that no competing interests exist.

3. Dennis Goldfarb

   Department of Computer Science, University of North Carolina at Chapel Hill, Chapel Hill, United States

   Contribution

   DG, Conception and design, Acquisition of data, Analysis and interpretation of data

   Contributed equally with

   Evan M Cornett and Paul A DaRosa

   Competing interests

   The authors declare that no competing interests exist.

4. Paul A DaRosa

   Department of Biochemistry, University of Washington, Seattle, United States

   Contribution

   PAD, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article

   Contributed equally with

   Evan M Cornett and Dennis Goldfarb

   Competing interests

   The authors declare that no competing interests exist.

5. Zimeng M Li

   Department of Physics and Astronomy, University of North Carolina at Chapel Hill, Chapel Hill, United States

   Contribution

   ZML, Acquisition of data, Analysis and interpretation of data

   Competing interests

   The authors declare that no competing interests exist.

6. Feng Yan

   Department of Cell Biology and Physiology, University of North Carolina at Chapel Hill, Chapel Hill, United States

   Contribution

   FY, Conception and design, Acquisition of data

   Competing interests

   The authors declare that no competing interests exist.

7. Bradley M Dickson

   Center for Epigenetics, Van Andel Research Institute, Grand Rapids, United States

   Contribution

   BMD, Conception and design, Analysis and interpretation of data

   Competing interests

   The authors declare that no competing interests exist.

8. Angela H Guo

   Department of Biochemistry and Biophysics, University of North Carolina at Chapel Hill, Chapel Hill, United States

   Contribution

   AHG, Acquisition of data

   Competing interests

   The authors declare that no competing interests exist.

9. Daniel V Cantu

   Department of Biochemistry and Biophysics, University of North Carolina at Chapel Hill, Chapel Hill, United States

   Contribution

   DVC, Acquisition of data

   Competing interests

   The authors declare that no competing interests exist.

10. Lilia Kaustov

    Structural Genomics Consortium, University of Toronto, Toronto, Canada

    Contribution

    LK, Acquisition of data

    Competing interests

    The authors declare that no competing interests exist.

11. Peter J Brown

    Structural Genomics Consortium, University of Toronto, Toronto, Canada

    Contribution

    PJB, Acquisition of data, Analysis and interpretation of data

    Competing interests

    The authors declare that no competing interests exist.

12. Cheryl H Arrowsmith

    Structural Genomics Consortium, University of Toronto, Toronto, Canada

    Contribution

    CHA, Conception and design, Analysis and interpretation of data

    Competing interests

    The authors declare that no competing interests exist.

13. Dorothy A Erie

    Department of Chemistry, University of North Carolina at Chapel Hill, Chapel Hill, United States

    Contribution

    DAE, Analysis and interpretation of data

    Competing interests

    The authors declare that no competing interests exist.

14. Michael B Major

    1. Department of Computer Science, University of North Carolina at Chapel Hill, Chapel Hill, United States
    2. Department of Cell Biology and Physiology, University of North Carolina at Chapel Hill, Chapel Hill, United States

    Contribution

    MBM, Conception and design, Analysis and interpretation of data

    Competing interests

    The authors declare that no competing interests exist.

15. Rachel E Klevit

    Department of Biochemistry, University of Washington, Seattle, United States

    Contribution

    REK, Conception and design, Drafting or revising the article

    Competing interests

    The authors declare that no competing interests exist.

16. Krzysztof Krajewski

    Department of Biochemistry and Biophysics, University of North Carolina at Chapel Hill, Chapel Hill, United States

    Contribution

    KK, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article

    Competing interests

    The authors declare that no competing interests exist.

    "This ORCID iD identifies the author of this article:" 0000-0001-7159-617X

17. Brian Kuhlman

    1. Department of Biochemistry and Biophysics, University of North Carolina at Chapel Hill, Chapel Hill, United States
    2. Lineberger Comprehensive Cancer Center, University of North Carolina at Chapel Hill, Chapel Hill, United States

    Contribution

    BK, Conception and design, Analysis and interpretation of data, Drafting or revising the article

    Competing interests

    The authors declare that no competing interests exist.

18. Brian D Strahl

    1. Department of Biochemistry and Biophysics, University of North Carolina at Chapel Hill, Chapel Hill, United States
    2. Lineberger Comprehensive Cancer Center, University of North Carolina at Chapel Hill, Chapel Hill, United States

    Contribution

    BDS, Conception and design, Analysis and interpretation of data, Drafting or revising the article

    Competing interests

    The authors declare that no competing interests exist.

19. Scott B Rothbart

    Center for Epigenetics, Van Andel Research Institute, Grand Rapids, United States

    Contribution

    SBR, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article

    For correspondence

    scott.rothbart@vai.org

    Competing interests

    The authors declare that no competing interests exist.

    "This ORCID iD identifies the author of this article:" 0000-0003-1631-5392

• 6,450

  views

• 1,099

  downloads

• 119

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.