- Open Access
Citrullination of HP1γ chromodomain affects association with chromatin
© The Author(s) 2019
- Received: 31 January 2019
- Accepted: 19 March 2019
- Published: 2 April 2019
Stem cell differentiation involves major chromatin reorganisation, heterochromatin formation and genomic relocalisation of structural proteins, including heterochromatin protein 1 gamma (HP1γ). As the principal reader of the repressive histone marks H3K9me2/3, HP1 plays a key role in numerous processes including heterochromatin formation and maintenance.
We find that HP1γ is citrullinated in mouse embryonic stem cells (mESCs) and this diminishes when cells differentiate, indicating that it is a dynamically regulated post-translational modification during stem cell differentiation. Peptidylarginine deiminase 4, a known regulator of pluripotency, citrullinates HP1γ in vitro. This requires R38 and R39 within the HP1γ chromodomain, and the catalytic activity is enhanced by trimethylated H3K9 (H3K9me3) peptides. Mutation of R38 and R39, designed to mimic citrullination, affects HP1γ binding to H3K9me3-containing peptides. Using live-cell single-particle tracking, we demonstrate that R38 and R39 are important for HP1γ binding to chromatin in vivo. Furthermore, their mutation reduces the residence time of HP1γ on chromatin in differentiating mESCs.
Citrullination is a novel post-translational modification of the structural heterochromatin protein HP1γ in mESCs that is dynamically regulated during mESC differentiation. The citrullinated residues lie within the HP1γ chromodomain and are important for H3K9me3 binding in vitro and chromatin association in vivo.
- Stem cell differentiation
Embryonic stem cells (ESCs) are pluripotent cells with the unique ability to self-renew and differentiate into nearly every cell type of the body. This plasticity is maintained by a distinct nuclear architecture with particular epigenetic signatures including enrichment of active chromatin marks and dynamic binding of structural chromatin proteins [1, 2].
HP1 was originally described as a dominant suppressor of position-effect variegation in Drosophila melanogaster . The mammalian HP1 protein family consists of three members: HP1α, β and γ. As the main reader of repressive histone marks H3K9me2/3, HP1 is a key factor in heterochromatin formation and maintenance [4, 5]. However, whilst HP1α and β are mainly associated with constitutive heterochromatic regions, HP1γ is predominantly found in euchromatin [6, 7], within the transcribed regions of active genes. Here, it regulates transcriptional elongation [8–11] and co-transcriptional mRNA processing [7, 12].
HP1γ plays important roles in developmental processes and cell fate decisions . For instance, depletion of HP1γ in mESCs results in endodermal and neuronal differentiation defects [14, 15]. Interestingly, HP1γ is differentially localised during mESC differentiation. In mESCs, it predominantly occupies gene bodies, whereas in neuronal precursor cells (NPCs), it is significantly enriched at the promoters of active genes [14, 16]. This indicates a direct role for HP1γ regulating transcription during differentiation. However, little is known concerning how the recruitment of HP1γ and its association with specific genomic loci is regulated.
Peptidyl citrullination (deimination) is the post-translational conversion of an arginine (R) residue within a protein to the non-encoded amino acid citrulline. This leads to loss of a positive charge and a reduction in hydrogen bonding ability. It is mediated by a vertebrate-specific family of enzymes called peptidylarginine deiminases (PADIs) which are associated with the development of different pathological conditions such as autoimmunity, cancer and neurodegenerative diseases . The deiminases are widely expressed in mammalian tissues, and PADIs 1, 2 and 4 have been reported to localise to the nucleus [18–20]. Histones are the best-characterised nuclear targets of PADI enzymes, and their citrullination is directly linked to transcriptional regulation [18–22].
We previously identified a role for PADI4 during reprogramming and ground state pluripotent stem cells , where the deiminase regulates key pluripotency genes via a mechanism involving citrullination of H1.2. This results in its eviction from chromatin and global chromatin decompaction. Here, we report that the chromodomain of HP1γ is also citrullinated in mESCs, at R38 and R39. We show that binding of HP1γ to peptides bearing H3K9me3 enhances its citrullination by PADI4 in vitro. Furthermore, mutations in HP1γ designed to mimic the loss of charge caused by citrullination inhibit binding of HP1γ to peptides harbouring H3K9me3 and reduce its residence time on chromatin in differentiating mESCs. We also observe a reduction in the level of HP1γ citrullination upon mESC differentiation. These results support a role for citrullination as a mechanism to regulate HP1γ chromatin association during stem cell differentiation.
HP1γ chromodomain is citrullinated in vivo
Efficient binding of HP1γ to H3K9me3 requires R38 and R39
We confirmed these results by assessing real-time binding kinetics of GST-HP1γ proteins to immobilised H3 or H3K9me3 peptides using bio-layer interferometry (BLI). As expected, HP1γWT binds exclusively to H3K9me3 peptides (Fig. 2b, Additional file 2: Figure S2B and C, Additional file 3: Table S1), which agrees with previously published analyses . We repeated these analyses at a range of different GST protein concentrations (Additional file 2: Figure S2C). Again, the HP1γR38/9A mutant protein displayed a distinctly different kinetic behaviour compared to the WT protein (Fig. 2b, Additional file 2: Figure S2B and C, Additional file 3: Table S1). Binding of HP1γR38/9A appears to begin to plateau at significantly lower amounts compared to the WT protein. These results confirm the data obtained from pulldown assays (Fig. 2a, Additional file 2: Figure S2A). However, a more thorough biophysical analysis is required to fully assess the kinetic behaviour of HP1γR38/9A molecules binding to H3K9me3 sites.
Interestingly, mutation of both R38 and R39 to lysine (R38/9K), which retains the positive charge at these positions, maintained binding to H3K9me3 peptides in pulldown assays (Additional file 2: Figure S2A), as well as in BLI measurements (Additional file 2: Figure S2B and C, Additional file 3: Table S1).
PADI4 citrullinates HP1γ in vitro
Interestingly, the addition of methylated H3K9me3 peptides to the in vitro citrullination assay increased HP1γ citrullination by PADI4 by approximately twofold (Fig. 3b, Additional file 4: Figure S3G). The enhanced citrullination appears specific because it was abolished when the HP1γR38/9K mutant protein was used as a substrate (Fig. 3b, Additional file 4: Figure S3G), even though this protein binds H3K9me3 peptides with similar affinity to the WT protein (Additional file 2: Figure S2). We conclude that PADI4 preferentially citrullinates HP1γ at R38 and R39 when the protein is bound to H3K9me3 peptides.
HP1γ citrullination is reduced in differentiating mESCs
Mutations of HP1γ R38 and R39 do not affect soluble diffusion rate
Tagged HP1γ molecules were labelled using low concentrations of HaloTag–JF549 ligand , and dynamics were measured in mESCs at 13.5 ms time resolution. Jump distance analysis as illustrated in Fig. 5a, b  showed that both WT and CSD mutant molecules exhibit two major diffusion coefficients (bound and diffusing) (Additional file 6: Figure S5B and C). This corresponds to published results using fluorescence correlation spectroscopy (FCS) of GFP-tagged HP1 proteins [28, 29]. Whilst diffusion coefficients of bound molecules remained unchanged for HP1γI165K mutant proteins, diffusing mutant molecules moved significantly more quickly than wildtype molecules (Additional file 6: Figure S5D to F, Additional file 7: Table S2). Overall, these results confirm previously published experiments employing fluorescence recovery after photobleaching (FRAP) of HP1 CSD mutants in both heterochromatic and euchromatic regions in vivo  and validate SPT as a suitable technique to detect differences in HP1 mobility.
Given the success of the above proof of principle experiments, we next compared mESCs (+/− LIF) expressing HP1γWT protein with cells expressing HP1γR38/9A, mimicking citrullinated HP1γ. Importantly, exogenous expression of HP1γ proteins had no discernible effect on mESC differentiation as (i) ground state mRNA levels of the pluripotency markers (Additional file 5: Figure S4B); (ii) down-regulation of mRNA levels of the pluripotency markers after LIF withdrawal (Additional file 5: Figure S4C); and (iii) morphological changes after LIF withdrawal (Additional file 5: Figure S4D) in mESCs stably expressing either HP1γWT or HP1γR38/9A mutant proteins did not differ from those of parental mESCs.
We observed increased HP1γ foci formation in differentiating cells by fluorescence imaging of mEos3.2 (Fig. 5c), confirming a previous study using the same differentiation protocol . However, no apparent difference was detected between HP1γWT- and HP1γR38/9A-expressing cells. Similarly, SPT analyses of either HP1γWT or HP1γR38/9A mutant proteins showed that LIF withdrawal had no significant effect on their diffusion rate in mESCs (Fig. 5d–f, Additional file 7: Table S2). These results indicate that the diffusion of soluble HP1γ molecules is not affected by LIF withdrawal and it does not depend upon R38 and R39.
HP1γ R38 and R39 determine chromatin residence time in differentiating mESCs
Whilst LIF withdrawal had no effect on the unspecific residence time of either HP1γWT or HP1γR38/9A molecules, the specific residence time of HP1γWT molecules was significantly increased (20 s longer upon differentiation, Fig. 6b, d). In stark contrast, the specific residence time of R38/9A mutant HP1γ molecules was considerably shorter (Fig. 6b, d). Consequently, the average time that HP1γR38/9A molecules remain stably bound to chromatin in differentiating mESCs is significantly lower than for HP1γWT molecules (29.5 ± 4.1 vs. 49.2 ± 12.3 s (Fig. 6b, d)).
On the basis of these results, we conclude that HP1γ residues R38 and R39 are not critical for HP1 recruitment to chromatin during stem cell differentiation, because diffusion of soluble HP1γ molecules remained unchanged upon LIF withdrawal. However, the two residues influence the overall stability of HP1γ molecules bound to chromatin in differentiating mESCs.
Our finding that PADI4-mediated citrullination of HP1γ is significantly increased when HP1γ is bound to H3K9me3 peptides supports the notion that HP1γ is targeted by the deiminase when bound to chromatin. An X-ray crystal structure of HP1γ’s chromodomain bound to H3K9me3 (Fig. 1a) indicates that R38 and R39 are facing away from the H3K9me3 binding pocket, suggesting the residues are accessible to be targeted by an enzyme. However, to fully understand how PADI4 targeting is achieved structurally, biophysical analyses such as X-Ray crystallography or NMR of the enzyme in complex with H3K9me3 bound HP1γ will be required.
Given our in vitro citrullination data, we hypothesise that HP1γ citrullination may represent a relatively rare modification occurring at specific genomic loci to which PADI4 is recruited. Thus, HP1γ’s residence time would be affected by citrullination at specific genomic loci. In fact, our MS analysis assessed the stoichiometry of HP1γ citrullination in mESCs to be in the very low percentage range, which was supported by our immunoblotting experiments which detected relatively low levels of HP1γ citrullination in mESCs (Fig. 4d). Given this, we believe that the citrullination of HP1γ acts to regulate its chromatin binding at specific genomic sites, possibly at genes controlling cellular plasticity.
Citrullination of histone H3R8 by PADI4 at gene promoters reduces the binding affinity of HP1 to adjacent methylated H3K9 which results in gene activation . It is possible that PADI4 targets both H3R8 and HP1γ bound to the same H3N-terminal tail. The Nanog gene may represent an important example of such a locus, since chromatin immunoprecipitation experiments have shown that both HP1γ and PADI4 bind to its promoter and regulate gene expression in stem cells [16, 22].
Additional support of our model stems from a proteomics study demonstrating HP1γ is predominantly nucleoplasmic in mESCs, with only a small fraction stably bound to chromatin . In contrast, in cells with less pluripotent potential, HP1γ is more chromatin bound and predominantly interacts with H1(39), which itself is more chromatin bound  (Fig. 7).
HP1 proteins possess domains that allow interaction with chromatin in a multivalent fashion (reviewed in ). Here, we have shown that single-particle tracking can be used to reliably measure HP1 dynamics in live cells. Given the significant increase in diffusion of fast moving HP1 molecules after mutating the CSD, it is highly likely that HP1 molecules move and associate with chromatin as dimers and/or as part of larger protein complexes. The recently discovered ChAHP complex, comprised of transcription factor ADNP, chromatin remodeler CHD4 and HP1, plays a crucial role in transcriptionally regulating cell fate decisions . Hence, it will be interesting to investigate how this protein complex affects HP1 dynamics in stem cells.
Finally, it is currently unclear whether citrullination of HP1γ is involved in the reprogramming of cells to a more plastic state. However, this seems plausible since loss of HP1γ increases the reprogramming efficiency of mouse embryonic fibroblasts .
Our work identifies citrullination of HP1γ within its chromodomain as a novel post-translational modification in mESCs that is dynamically regulated during mESC differentiation. The arginines that are citrullinated within the chromodomain directly affect H3K9me3 binding in vitro and influence the overall stability of HP1γ molecules bound to chromatin in differentiating mESCs. Overall, our data highlight citrullination of HP1γ as a mechanism facilitating mESC differentiation.
Mass spectrometry and identification of citrullinated peptides
Previously published tandem mass spectrometry data of mESCs  were re-analysed using an improved search algorithm. Mass spectrometry data analysis was performed with the MaxQuant software suite (version 18.104.22.168) as described  supported by Andromeda (www.maxquant.org) as the database search engine for peptide identifications . We followed the step-by-step protocol of the MaxQuant software suite  to generate MS/MS peak lists that were filtered to contain at most six peaks per 100-Da interval and searched by Andromeda against a concatenated target/decoy  (forward and reversed) version of the Uniprot human database version (70.101 forward protein entries). Protein sequences of common contaminants such as human keratins and proteases used were added to the database. The initial mass tolerance in MS mode was set to 7 ppm, and MS/MS mass tolerance was set to 20 ppm. Cysteine carbamidomethylation was searched as a fixed modification, whereas protein N-acetylation, oxidised methionine, deamidation of asparagine and glutamine, and citrullination of arginines were searched as variable modifications. A maximum of two miscleavages was allowed whilst we required strict LysC specificity. Peptide assignments were statistically evaluated in a Bayesian model on the basis of sequence length and Andromeda score. We only accepted peptides and proteins with a false discovery rate of < 1%, estimated on the basis of the number of accepted reverse hits.
Mammalian and bacterial expression constructs
The vector pSB7 was generated from the pEF/myc/ER backbone (Invitrogen) by replacing myc/ER regulatory elements with the sequence encoding mEos3.2–HaloTag. Mouse HP1γ cDNA was cloned into Xho1/Xba1 restriction sites of the mammalian expression vector pSB7 and into BamH1/EcoR1 restriction sites of the bacterial expression vector pGEX-2T. Point mutants were generated using Quick change II site-directed mutagenesis kit (Stratagene). Bacterial expression vector pGEX6p-GST-hPADI4 was described previously .
Immunoblotting and antibodies
For immunoblotting, cells were lysed in 2 × Laemmli buffer and sonicated in a water bath (Fisherbrand) for 3 min to shear genomic DNA, spun down at full speed in a microfuge for 15 min and incubated for 5 min at 95 °C. Proteins were resolved by SDS-PAGE, transferred to nitrocellulose membrane (Millipore) using wet transfer (transfer buffer: 192 mM glycine, 25 mM Tris.HCl pH8.8) and incubated in blocking solution (5% BSA in TBS containing 0.5% Tween-20 for citrulline sensitive antibodies and 5% non-fat milk powder in TBS containing 0.1% Tween-20 for all other antibodies) for 1 h at room temperature (RT). Membranes were incubated with primary antibody at 4 °C o/n and appropriate HRP-conjugated secondary antibody for 1 h at RT. Membranes were then incubated for enhanced chemiluminescence (ECL, Promega), and proteins were detected by exposure to X-ray film or using the ChemiDoc™ MP imaging system (Bio-Rad). Images were quantified using ImageJ analysis software. Primary antibodies diluted in blocking solution were used to detect citrullinated histone H3R2 (H3R2-Cit, Abcam ab176843 at 1:1000 dilution), HP1γ (Millipore 05-690 MAB clone 42s2  at 1:1000 dilution), β-Tubulin (Abcam ab6046 at 1:1000 dilution) and peptidyl-citrulline (Millipore MABN328, clone F95  at 1:1000 dilution). HRP-conjugated secondary antibodies diluted in blocking solution were used against mouse IgG (Dako, P0447 at 1:5000 dilution), rabbit IgG (Abcam, ab6721 at 1:20,000 dilution) and sheep IgG (Abcam, ab6747 at 1:5000 dilution).
Antibody purification from crude sera
Site-specific citrulline antibodies against HP1γ R38/9-citrulline were raised by Orygen Antibodies Ltd. Sheep were immunised with HP1γ peptides (aa 34–44) citrullinated at residues 38 and 39 (KVLD(Cit)(Cit)VVNGKC) that was C-terminally coupled to KLH. After the initial immunisation, 3 boost injections every 4 weeks were performed. Crude serum of the final bleed was purified against the HP1γ R38/9-citrulline peptide (same as immunogen) columns. In brief, peptides were first reduced using Bond-Breaker™ TCEP Solution (Thermofisher Scientific) and then coupled to SulfoLink™ Coupling Resin (Thermofisher Scientific) according to the manufacturer’s instructions. Non-specific binding of the column was blocked using quenching reagent (50 mM l-cysteine in coupling buffer), followed by 3 washes in wash buffer (1 M NaCl in H2O) and twice in 1xTBS. Crude serum was centrifuged at maximum speed for 10 min; supernatant was added to the resin and incubated at RT for 1 h on a rotating wheel. Resin was then washed 4x in 1xTBS. To elute the antibodies, elution buffer (0.2 M glycine HCl pH 2.5) was added and tube was quickly flicked and directly centrifuged at 800 rpm for 1 min. Supernatant was moved to a fresh tube and placed on ice. Immediately ice-cold neutralisation buffer (1 M Tris HCl pH 8.8) was added and mixed thoroughly to neutralise the pH to 7.
Dot blot assay
Dot blot assays were performed by spotting synthetic peptides (HP1γ (34–44) unmodified: KVLDRRVVNGKC; HP1γR38Cit: KVLD(Cit)RVVNGKC; HP1γR39Cit: KVLDR(Cit)VVNGKC; HP1γR38/9Cit KVLD(Cit)(Cit)VVNGKC; HP1γ (104–111) R108Cit: SKKK(Cit)DAADKC; HP1α (103–112) R107Cit SKKK(Cit)EQSNDC) onto pre-wet PVDF membranes. Membranes were blocked in 5% BSA blocking buffer for 1 h at RT. Blocked membranes were then probed with the affinity purified HP1γR38/9-Cit antibody (at 1:500 dilution) at 4 °C o/n. Membranes were washed 3 times for 30 s in TBS with 0.1% Tween-20 followed by another 10 min wash in the same buffer to remove residual antibodies. Membranes were then incubated in appropriate secondary antibodies and developed as described in “Immunoblotting and antibodies” section.
mESCs were either cultured under serum/LIF conditions or without LIF for 72 h. 2 × 107 cells were harvested, washed twice in ice-cold 1xPBS and lysed in 1 ml of nuclear lysis buffer (NLB: 10 mM Hepes KOH pH 7.9, 1.5 mM MgCl2, 10 mM KCl, 0.5 mM DTT, 0.1% NP-40 supplemented with cOmplete™ proteinase inhibitor cocktail (Roche)). Lysate was incubated for 10 min on ice, and nuclei were pelleted at 2000 rpm for 10 min. Nuclear pellet was washed in NLB and lysed in 0.5 ml 1 × RIPA buffer (10 mM Tris pH 8.0, 1 mM EDTA, 1% Triton X-100, 0.1% sodium deoxycholate, 0.1% SDS, 140 mM NaCl) Supplementaled with cOmplete™ proteinase inhibitor cocktail (Roche)). The lysate was incubated on ice for 20 min with vortexing every 5 min and centrifuged at full speed for 15 min to remove cellular debris. 2 μg of HP1γ antibody (Millipore 05-690 MAB clone 42s2) or an isotype control (Abcam, HA.C5, ab18181) was added and incubated at 4 °C overnight on a rotating wheel. 30 μl of Protein G Dynabeads® (Thermo Fisher Scientific) was added and incubated for 2 h at 4 °C on a rotating wheel. Beads were washed 3 × in ice-cold 1× RIPA, lysed in 2 × Laemmli buffer, resolved by SDS-PAGE and subjected to immunoblotting.
Purification of recombinant GST-hPADI4 and in vitro deimination assay
Recombinant human GST-PADI4 was expressed from pGEX6p constructs  in LBAmp media, induced with 0.1 mM IPTG at 30 °C for 4 h, purified using glutathione-Sepharose resin, eluted using a 25 mM glutathione solution and stored in 20% glycerol at − 20 °C. In vitro deimination of GST-HP1 variants was carried out as described previously . In vitro deimination assays were performed in the presence and absence of 100 ng H3(1–16) or H3K9me3 peptides as appropriate. Samples were resuspended in 1 × Laemmli buffer for immunoblot analysis.
Bio-layer interferometry (BLI)
Recombinant mouse full-length HP1 wildtype and mutant proteins were expressed in (as GST fusion proteins), and purified from, Escherichia coli as described . Proteins were concentrated using Amicon Ultra centrifugation filters with a 30 kDa cutoff (Millipore), washed into binding buffer (20 mM Hepes KOH pH 7.2, 70 mM KCl, 1 mM DTT, 0.1% Tween-20, 5% glycerol). Protein concentrations were determined by Bradford protein assay against a BSA standard, and purity was verified by SDS-PAGE followed by Coomassie-blue staining. Real-time binding assays of GST-HP1γ were performed using the BLItz® system (ForteBio) in binding buffer (+ 0.1% BSA). Biotinylated H3(1–16) peptides (NH2-ARTKQTARK(me3)STGGKAP-Biotin) were immobilised to Dip and Read High Precision Streptavidin (SAX) Biosensors (ForteBio), and kinetics for a range of protein concentrations were measured under advanced kinetics settings (1. initial baseline measuring buffer only: 30 s, 2. loading of peptides onto sensors: 120 s, 3. effective baseline measurement: 30 s, 4. association of GST-HP1γ proteins: 120 s, 5. dissociation of GST-HP1γ proteins: 120 s). The data were normalised to the effective baseline signal by subtracting the average baseline value (150–180 s) from each data point (150–420 s). Values of 1. and 2. measurements were not included in analysis and are not displayed in figures. Raw data are summarised in Additional file 3: Table S1.
Peptide pulldown assay
H3K9me3 or unmethylated H3(1–16) peptides, linked to Dynabeads MyOne streptavidin T1 (Thermo Fisher Scientific), were used in pulldown assays with eluted GST-HP1 in HBS-EP (300 mM NaCl, 10 mM HEPES KOH pH 7.4, 5 mM EDTA and 0.2% NP-40). Pulldown assays were performed in 300 mM NaCl HBS-EP as previously described . Samples were resuspended in 2 × Laemmli buffer and resolved by SDS-PAGE followed by Coomassie-blue staining. Images were quantified using ImageJ software.
Parental E14 mouse embryonic stem cell (mESC) line was profiled by transcriptome analysis, qPCR and potency assays  and was routinely tested for mycoplasma contamination and tested negative. Cells were cultured in DMEM (Gibco) supplemented with 15% standard FCS (Gibco), 0.1 mM non-essential amino acids (Gibco), 1% penicillin/streptomycin, 2 mM l-glutamine (Gibco), 1 mM sodium pyruvate (Gibco), 0.1 mM β-mercaptoethanol (Life Technologies) and 106 units/l mLIF (ESGRO, Millipore). E14 cells were stored in incubators at 37 °C in 7.5% CO2 (100% humidity) and maintained in 0.1% gelatin coated 10-cm culture dishes with 1:6 passaging every 2 days. Media was changed on a daily basis. E14 ESCs expressing mouse mEos3.2–HaloTag-HP1 fusion proteins were generated by transfecting appropriate pSB7 plasmids into E14 cells using Lipofectamine 2000 transfection reagent (Invitrogen), followed by selection in 600 µg/ml geneticin (Life Technologies). After 2 weeks of geneticin selection, cells were sorted using a Sony SH800S cell sorter for the expression of mEos3.2. Differentiation of ESCs was induced by LIF withdrawal. ESCs cultured in complete ES cell media (+ LIF) were trypsinised and washed twice in 1x PBS to remove residual LIF, seeded into ES cell media without LIF and incubated for 24, 48 or 72 h.
cDNA synthesis and RT-PCR
Cells were lysed using QIAzol® lysis reagent (Qiagen) and total RNA was isolated using RNeasy® Mini Kit with in column DNase treatment (Qiagen). 1 μg of total RNA was reverse-transcribed using Superscript III Reverse Transcriptase kit (Invitrogen). Samples were diluted 1:10 in nuclease-free water prior to RT-PCR analysis. RT-PCR reactions were performed in duplicates for each sample. Each RT-PCR reaction was performed in a final volume of 10 μl. Fast SYBR Green Master Mix (Applied Biosystems) was used, according to the manufacturer’s instructions. A melting curve was obtained for each PCR product after each run, in order to confirm that the SYBR Green signal corresponded to a unique and specific amplicon. Relative expression was calculated with the 2−ΔΔCt method, normalising to the housekeeping gene β-actin. Data were analysed using GraphPad Prism 7 software. Primers used in this study are summarised in Additional file 13: Table S3.
An IX73 Olympus inverted microscope was used with circularly polarised laser beams aligned and focused at the back aperture of an Olympus 1.40 NA 100 × oil objective (Universal Plan Super Apochromat, 100 × , NA 1.40, UPLSAPO100XO/1.4). Continuous wavelength diode laser light sources used include a 561-nm (Cobolt, Jive 200, 200 mW) and a 405-nm laser (Stradus, Toptica, 405–100, 100 mW). Oblique angle illumination was achieved by aligning the laser off axis such that the emergent beam at the sample interface was near-collimated and incident at an angle slightly less than the critical angle θc ~ 67° for a glass/water interface. This generated a ~ 50 μm diameter excitation footprint. The power of the collimated beams at the back aperture of the microscope was 10 kW/cm2 and 10–100 W/cm2 for the 561-nm and 405-nm laser beams, respectively. The lasers were reflected by dichroic mirrors (Semrock, Di01-R405/488/561/635). The fluorescence emission was collected through the same objective and then further filtered using a combination of long-pass and band-pass filters (BLP01-561R and FF01-587/35 for 561 nm excitation). The emission signal was projected onto an EMCCD (Photometrics, Evolve 512 Delta) with an electron multiplication gain of 250 ADU/photon operating in a frame transfer mode. The instrument was automated using the open-source software micromanager , and the data were displayed using the ImageJ software.
Mammalian live-cell single-molecule imaging and analysis
mESCs expressing mEos3.2–HaloTag-tagged HP1γ were passaged 2 days before imaging onto 35-mm glass bottom dishes No 1.0 (MatTek Corporation P35G-1.0-14-C Case) in phenol red-free serum and LIF conditions, if not stated otherwise. Just before single-particle tracking (SPT) imaging experiments, cells were labelled with 0.5 nM HaloTag–JF549 ligand for at least 15 min, followed by two washes in PBS and a 30-min incubation at 37 °C in media, before imaging the cells in fresh phenol red-free serum and LIF conditions. Cells were under-labelled to prevent overlap of fluorophores during SPT. Using 561 nm excitation, fluorescence images were collected as movies of 10,000 frames at 13.5 ms or 1000 to 5000 frames at 500 ms exposure. For live-cell tracking, data were collected for three biological replicates on separate dishes and the average values compared. Experiments carried out on two separate days showed similar trends.
Live-cell 13.5-ms single-molecule traces were analysed using software that detects single-molecule trajectories from SPT movies . Briefly, the code detected puncta positions in each image frame using a brightness-weighted centroid after applying a band-pass filter to remove low-frequency background and high-frequency noise from the image data. To track the HP1 single molecules, only fluorescent puncta smaller than 3 pixels and with a signal to noise greater than 8 were analysed. Fluorescent puncta were considered to be the same molecule if they were within 6 pixels (936 nm) between frames. We collected more than 5000 single-molecule trajectories from single-molecule movies of HP1γWT, HP1γI165K and HP1γR38/9A molecules detected within an ROI of an average of 3–4 cells. Given we observed an increase in diffusion coefficient of the fast sub-population of HP1WT molecules compared to the CSD mutants (63% of molecules for HP1γWT), this is an appropriate sample size to demonstrate the change we observe. 13.5-ms SPT experiment (i) was recorded in 1 day, and repeat experiments (n = 2–9, number of replicates varies between cell lines) were recorded on a different day over the course of 5 days. 500-ms data repeat experiments (n = 3) were recorded on three different days. Independent replicates were compared using two-way ANOVA multiple corrections with Sidak post hoc test. This is justified because the variances calculated were similar between samples.
Live-cell 500-ms single-molecule traces were analysed using Rapidstorm software that determines single-molecule localisations from PALM movies , after using ImageJ’s rolling back background correction with a radius of 5 pixels. Only fluorescent puncta less than 3 pixels wide with a fixed global threshold above 25,000 were analysed. We collected more than 1000 single-molecule trajectories from single-molecule movies of HP1γWT and HP1γR38/9A molecules before and after mESC differentiation. Given we observed an increase in the residence time of the stable binding HP1γWT molecules (10–13% of molecules), this is an appropriate sample size to demonstrate the change we observe.
MW and AJB designed experiments and interpreted results. MW carried out all in vitro experiments, generated cell lines and prepared samples for imaging analysis. SB performed SPT experiments and data analysis with the help of WB and KW. SPT experiments were carried out using instrumentations in DK’s laboratory. Mass spectrometry data were generated and analysed by MAC and MLN. TK, SB and EDL supervised the project. MW and AJB wrote the manuscript with contributions from all authors. All authors read and approved the final manuscript.
We thank Dr. Claudia Keller Valsecchi for help setting up the BLI experiments and Dr. Asifa Akhtar for providing access to equipment in her laboratory. We thank Dr. Valentina Migliori for proof reading the manuscript and helpful comments.
TK is a co-founder of Abcam plc (Cambridge, UK).
Availability of data and material
The datasets generated and analysed during the current study as well as materials are available from the authors on request. The software used for live-cell diffusion analysis and residence time analysis can be found at https://github.com/TheLaueLab/weimann-tracking and https://github.com/TheLaueLab/trajectory-analysis, respectively. Other softwares used include the open-source software micromanager (https://www.micro-manager.org), ImageJ [53, 54], Rapidstorm  and PeakFit .
The Kouzarides laboratory is supported by a grant from Cancer Research UK (Grant Reference RG17001) in addition to benefiting from core support from the Wellcome Trust (Core Grant Reference WT203144) and Cancer Research UK (Grant Reference C6946/A24843). The Laue laboratory is supported by grants from the Wellcome Trust (206291/Z/17/Z) and the Medical Research Council (MR/P019471/1 and MR/M010082/1). M.L.N. was partly supported by the Novo Nordisk Foundation Center for Protein Research, the Lundbeck Foundation, and by and the European Commission’s 7th Framework Programme HEALTH-F7-2010-242129/SYBOSS. David Klenerman is supported by a Royal Society Professorship (RP150066), and the microscope used for single-particle tracking experiments was funded by the MRC (MR/K015850/1).
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Meshorer E, Misteli T. Chromatin in pluripotent embryonic stem cells and differentiation. Nat Rev Mol Cell Biol. 2006;7(7):540–6.PubMedView ArticleGoogle Scholar
- Meshorer E, Yellajoshula D, George E, Scambler PJ, Brown DT, Misteli T. Hyperdynamic plasticity of chromatin proteins in pluripotent embryonic stem cells. Dev Cell. 2006;10(1):105–16.PubMedPubMed CentralView ArticleGoogle Scholar
- Eissenberg JC, Morris GD, Reuter G, Hartnett T. The heterochromatin-associated protein HP-1 is an essential protein in Drosophila with dosage-dependent effects on position-effect variegation. Genetics. 1992;131(2):345–52.PubMedPubMed CentralGoogle Scholar
- Bannister AJ, Zegerman P, Partridge JF, Miska EA, Thomas JO, Allshire RC, et al. Selective recognition of methylated lysine 9 on histone H3 by the HP1 chromo domain. Nature. 2001;410(6824):120–4.PubMedView ArticleGoogle Scholar
- Nielsen PR, Nietlispach D, Mott HR, Callaghan J, Bannister A, Kouzarides T, et al. Structure of the HP1 chromodomain bound to histone H3 methylated at lysine 9. Nature. 2002;416(6876):103–7.PubMedView ArticleGoogle Scholar
- Maison C, Almouzni G. HP1 and the dynamics of heterochromatin maintenance. Nat Rev Mol Cell Biol. 2004;5(4):296–304.PubMedView ArticleGoogle Scholar
- Smallwood A, Hon GC, Jin F, Henry RE, Espinosa JM, Ren B. CBX3 regulates efficient RNA processing genome-wide. Genome Res. 2012;22(8):1426–36.PubMedPubMed CentralView ArticleGoogle Scholar
- Vakoc CR, Mandat SA, Olenchock BA, Blobel GA. Histone H3 lysine 9 methylation and HP1gamma are associated with transcription elongation through mammalian chromatin. Mol Cell. 2005;19(3):381–91.PubMedView ArticleGoogle Scholar
- Lomberk G, Bensi D, Fernandez-Zapico ME, Urrutia R. Evidence for the existence of an HP1-mediated subcode within the histone code. Nat Cell Biol. 2006;8(4):407–15.PubMedView ArticleGoogle Scholar
- Harouz H, Rachez C, Meijer BM, Marteyn B, Donnadieu F, Cammas F, et al. Shigella flexneri targets the HP1gamma subcode through the phosphothreonine lyase OspF. EMBO J. 2014;33(22):2606–22.PubMedPubMed CentralView ArticleGoogle Scholar
- Thorne JL, Ouboussad L, Lefevre PF. Heterochromatin protein 1 gamma and IkappaB kinase alpha interdependence during tumour necrosis factor gene transcription elongation in activated macrophages. Nucleic Acids Res. 2012;40(16):7676–89.PubMedPubMed CentralView ArticleGoogle Scholar
- Saint-Andre V, Batsche E, Rachez C, Muchardt C. Histone H3 lysine 9 trimethylation and HP1gamma favor inclusion of alternative exons. Nat Struct Mol Biol. 2011;18(3):337–44.PubMedView ArticleGoogle Scholar
- Morikawa K, Ikeda N, Hisatome I, Shirayoshi Y. Heterochromatin protein 1gamma overexpression in P19 embryonal carcinoma cells elicits spontaneous differentiation into the three germ layers. Biochem Biophys Res Commun. 2013;431(2):225–31.PubMedView ArticleGoogle Scholar
- Huang C, Su T, Xue Y, Cheng C, Lay FD, McKee RA, et al. Cbx3 maintains lineage specificity during neural differentiation. Genes Dev. 2017;31(3):241–6.PubMedPubMed CentralView ArticleGoogle Scholar
- Caillier M, Thenot S, Tribollet V, Birot AM, Samarut J, Mey A. Role of the epigenetic regulator HP1gamma in the control of embryonic stem cell properties. PLoS ONE. 2010;5(11):e15507.PubMedPubMed CentralView ArticleGoogle Scholar
- Sridharan R, Gonzales-Cope M, Chronis C, Bonora G, McKee R, Huang C, et al. Proteomic and genomic approaches reveal critical functions of H3K9 methylation and heterochromatin protein-1gamma in reprogramming to pluripotency. Nat Cell Biol. 2013;15(7):872–82.PubMedPubMed CentralView ArticleGoogle Scholar
- Wang S, Wang Y. Peptidylarginine deiminases in citrullination, gene regulation, health and pathogenesis. Biochim Biophys Acta. 2013;1829(10):1126–35.PubMedPubMed CentralView ArticleGoogle Scholar
- Wang Y, Wysocka J, Sayegh J, Lee YH, Perlin JR, Leonelli L, et al. Human PAD4 regulates histone arginine methylation levels via demethylimination. Science. 2004;306(5694):279–83.PubMedView ArticleGoogle Scholar
- Zhang X, Bolt M, Guertin MJ, Chen W, Zhang S, Cherrington BD, et al. Peptidylarginine deiminase 2-catalyzed histone H3 arginine 26 citrullination facilitates estrogen receptor alpha target gene activation. Proc Natl Acad Sci USA. 2012;109(33):13331–6.PubMedView ArticleGoogle Scholar
- Zhang X, Liu X, Zhang M, Li T, Muth A, Thompson PR, et al. Peptidylarginine deiminase 1-catalyzed histone citrullination is essential for early embryo development. Sci Rep. 2016;6:38727.PubMedPubMed CentralView ArticleGoogle Scholar
- Cuthbert GL, Daujat S, Snowden AW, Erdjument-Bromage H, Hagiwara T, Yamada M, et al. Histone deimination antagonizes arginine methylation. Cell. 2004;118(5):545–53.PubMedView ArticleGoogle Scholar
- Christophorou MA, Castelo-Branco G, Halley-Stott RP, Oliveira CS, Loos R, Radzisheuskaya A, et al. Citrullination regulates pluripotency and histone H1 binding to chromatin. Nature. 2014;507(7490):104–8.PubMedPubMed CentralView ArticleGoogle Scholar
- Bryan LC, Weilandt DR, Bachmann AL, Kilic S, Lechner CC, Odermatt PD, et al. Single-molecule kinetic analysis of HP1-chromatin binding reveals a dynamic network of histone modification and DNA interactions. Nucl Acids Res. 2017;45(18):10504–17.PubMedView ArticleGoogle Scholar
- Deplus R, Denis H, Putmans P, Calonne E, Fourrez M, Yamamoto K, et al. Citrullination of DNMT3A by PADI4 regulates its stability and controls DNA methylation. Nucl Acids Res. 2014;42(13):8285–96.PubMedView ArticleGoogle Scholar
- Tanikawa C, Ueda K, Suzuki A, Iida A, Nakamura R, Atsuta N, et al. Citrullination of RGG motifs in FET proteins by PAD4 regulates protein aggregation and ALS susceptibility. Cell Rep. 2018;22(6):1473–83.PubMedView ArticleGoogle Scholar
- Cheutin T, McNairn AJ, Jenuwein T, Gilbert DM, Singh PB, Misteli T. Maintenance of stable heterochromatin domains by dynamic HP1 binding. Science. 2003;299(5607):721–5.PubMedView ArticleGoogle Scholar
- Festenstein R, Pagakis SN, Hiragami K, Lyon D, Verreault A, Sekkali B, et al. Modulation of heterochromatin protein 1 dynamics in primary Mammalian cells. Science. 2003;299(5607):719–21.PubMedView ArticleGoogle Scholar
- Schmiedeberg L, Weisshart K, Diekmann S, Meyer Zu Hoerste G, Hemmerich P. High- and low-mobility populations of HP1 in heterochromatin of mammalian cells. Mol Biol Cell. 2004;15(6):2819–33.PubMedPubMed CentralView ArticleGoogle Scholar
- Muller KP, Erdel F, Caudron-Herger M, Marth C, Fodor BD, Richter M, et al. Multiscale analysis of dynamics and interactions of heterochromatin protein 1 by fluorescence fluctuation microscopy. Biophys J. 2009;97(11):2876–85.PubMedPubMed CentralView ArticleGoogle Scholar
- Mishima Y, Jayasinghe CD, Lu K, Otani J, Shirakawa M, Kawakami T, et al. Nucleosome compaction facilitates HP1gamma binding to methylated H3K9. Nucl Acids Res. 2015;43(21):10200–12.PubMedGoogle Scholar
- Zhang M, Chang H, Zhang Y, Yu J, Wu L, Ji W, et al. Rational design of true monomeric and bright photoactivatable fluorescent proteins. Nat Methods. 2012;9(7):727–9.PubMedView ArticleGoogle Scholar
- Chen J, Zhang Z, Li L, Chen BC, Revyakin A, Hajj B, et al. Single-molecule dynamics of enhanceosome assembly in embryonic stem cells. Cell. 2014;156(6):1274–85.PubMedPubMed CentralView ArticleGoogle Scholar
- Grimm JB, English BP, Chen J, Slaughter JP, Zhang Z, Revyakin A, et al. A general method to improve fluorophores for live-cell and single-molecule microscopy. Nat Methods. 2015;12(3):244–50 (3 p following 50).PubMedPubMed CentralView ArticleGoogle Scholar
- Weimann L, Ganzinger KA, McColl J, Irvine KL, Davis SJ, Gay NJ, et al. A quantitative comparison of single-dye tracking analysis tools using Monte Carlo simulations. PLoS ONE. 2013;8(5):e64287.PubMedPubMed CentralView ArticleGoogle Scholar
- Dialynas GK, Terjung S, Brown JP, Aucott RL, Baron-Luhr B, Singh PB, et al. Plasticity of HP1 proteins in mammalian cells. J Cell Sci. 2007;120(Pt 19):3415–24.PubMedView ArticleGoogle Scholar
- Etheridge TJ, Boulineau RL, Herbert A, Watson AT, Daigaku Y, Tucker J, et al. Quantification of DNA-associated proteins inside eukaryotic cells using single-molecule localization microscopy. Nucl Acids Res. 2014;42(19):e146.PubMedView ArticleGoogle Scholar
- Gebhardt JC, Suter DM, Roy R, Zhao ZW, Chapman AR, Basu S, et al. Single-molecule imaging of transcription factor binding to DNA in live mammalian cells. Nat Methods. 2013;10(5):421–6.PubMedPubMed CentralView ArticleGoogle Scholar
- Sharma P, Azebi S, England P, Christensen T, Moller-Larsen A, Petersen T, et al. Citrullination of histone H3 interferes with HP1-mediated transcriptional repression. PLoS Genet. 2012;8(9):e1002934.PubMedPubMed CentralView ArticleGoogle Scholar
- Zaidan NZ, Walker KJ, Brown JE, Schaffer LV, Scalf M, Shortreed MR, et al. Compartmentalization of HP1 Proteins in Pluripotency Acquisition and Maintenance. Stem Cell Rep. 2018;10(2):627–41.View ArticleGoogle Scholar
- Canzio D, Larson A, Narlikar GJ. Mechanisms of functional promiscuity by HP1 proteins. Trends Cell Biol. 2014;24(6):377–86.PubMedPubMed CentralView ArticleGoogle Scholar
- Ostapcuk V, Mohn F, Carl SH, Basters A, Hess D, Iesmantavicius V, et al. Activity-dependent neuroprotective protein recruits HP1 and CHD4 to control lineage-specifying genes. Nature. 2018;557(7707):739–43.PubMedView ArticleGoogle Scholar
- Lee YH, Coonrod SA, Kraus WL, Jelinek MA, Stallcup MR. Regulation of coactivator complex assembly and function by protein arginine methylation and demethylimination. Proc Natl Acad Sci USA. 2005;102(10):3611–6.PubMedView ArticleGoogle Scholar
- Cox J, Mann M. MaxQuant enables high peptide identification rates, individualized p.p.b.-range mass accuracies and proteome-wide protein quantification. Nat Biotechnol. 2008;26(12):1367–72.PubMedView ArticleGoogle Scholar
- Cox J, Neuhauser N, Michalski A, Scheltema RA, Olsen JV, Mann M. Andromeda: a peptide search engine integrated into the MaxQuant environment. J Proteome Res. 2011;10(4):1794–805.PubMedView ArticleGoogle Scholar
- Cox J, Matic I, Hilger M, Nagaraj N, Selbach M, Olsen JV, et al. A practical guide to the MaxQuant computational platform for SILAC-based quantitative proteomics. Nat Protoc. 2009;4(5):698–705.PubMedView ArticleGoogle Scholar
- Elias JE, Gygi SP. Target-decoy search strategy for increased confidence in large-scale protein identifications by mass spectrometry. Nat Methods. 2007;4(3):207–14.PubMedView ArticleGoogle Scholar
- Olsen JV, Mann M. Improved peptide identification in proteomics by two consecutive stages of mass spectrometric fragmentation. Proc Natl Acad Sci USA. 2004;101(37):13417–22.PubMedView ArticleGoogle Scholar
- Nicholas AP, Whitaker JN. Preparation of a monoclonal antibody to citrullinated epitopes: its characterization and some applications to immunohistochemistry in human brain. Glia. 2002;37(4):328–36.PubMedView ArticleGoogle Scholar
- Edelstein AD, Tsuchida MA, Amodaj N, Pinkard H, Vale RD, Stuurman N. Advanced methods of microscope control using muManager software. J Biol Methods. 2014;1(2):e10.PubMedPubMed CentralView ArticleGoogle Scholar
- Wolter S, Loschberger A, Holm T, Aufmkolk S, Dabauvalle MC, van de Linde S, et al. rapidSTORM: accurate, fast open-source software for localization microscopy. Nat Methods. 2012;9(11):1040–1.PubMedView ArticleGoogle Scholar
- Schwarz G. Estimating the dimension of a model. Ann Stat. 1978;6(2):461–4.View ArticleGoogle Scholar
- Lagerholm BC, Andrade DM, Clausen MP, Eggeling C. Convergence of lateral dynamic measurements in the plasma membrane of live cells from single particle tracking and STED-FCS. J Phys D Appl Phys. 2017;50(6):063001.PubMedPubMed CentralView ArticleGoogle Scholar
- Schneider CA, Rasband WS, Eliceiri KW. NIH Image to ImageJ: 25 years of image analysis. Nat Methods. 2012;9(7):671–5.PubMedPubMed CentralView ArticleGoogle Scholar
- Edelstein A, Amodaj N, Hoover K, Vale R, Stuurman N. Computer control of microscopes using μManager. Curr Protoc Mol Biol. 2010;14(20):1–17.Google Scholar
- Herbert A. Peak fit, a single-molecule plugins. 2014. http://www.sussex.ac.uk/gdsc/intranet/microscopy/imagej/smlm_plugins. Accessed 21 Jan 2019.