The Highest Density of Phosphorylated Histone H1 Appeared in Prophase and Prometaphase in Parallel with Reduced H3K9me3, and HDAC1 Depletion Increased H1.2/H1.3 and H1.4 Serine 38 Phosphorylation

Background: Variants of linker histone H1 are tissue-specific and are responsible for chromatin compaction accompanying cell differentiation, mitotic chromosome condensation, and apoptosis. Heterochromatinization, as the main feature of these processes, is also associated with pronounced trimethylation of histones H3 at the lysine 9 position (H3K9me3). Methods: By confocal microscopy, we analyzed cell cycle-dependent levels and distribution of phosphorylated histone H1 (H1ph) and H3K9me3. By mass spectrometry, we studied post-translational modifications of linker histones. Results: Phosphorylated histone H1, similarly to H3K9me3, has a comparable level in the G1, S, and G2 phases of the cell cycle. A high density of phosphorylated H1 was inside nucleoli of mouse embryonic stem cells (ESCs). H1ph was also abundant in prophase and prometaphase, while H1ph was absent in anaphase and telophase. H3K9me3 surrounded chromosomal DNA in telophase. This histone modification was barely detectable in the early phases of mitosis. Mass spectrometry revealed several ESC-specific phosphorylation sites of H1. HDAC1 depletion did not change H1 acetylation but potentiated phosphorylation of H1.2/H1.3 and H1.4 at serine 38 positions. Conclusions: Differences in the level and distribution of H1ph and H3K9me3 were revealed during mitotic phases. ESC-specific phosphorylation sites were identified in a linker histone.


Introduction
Chromatin consists of DNA wrapped around an octamer of core histone proteins H2A, H2B, H3, and H4. Genomic regions between nucleosomes are protected by linker histone H1, responsible for chromatin condensation and access of other regulatory factors to linker DNA [1]. It is well known that histones, H1, are small proteins consisting of a ∼75 residue globular domain that is flanked by the N-terminal tail (∼20-35 aa) and the C-terminal region (∼100 aa) [2]. In human cells, we know eleven H1 variants with seven somatic subtypes (H1. 1 to H1. 5, H1.0, and H1X), three testis-specific variants (H1t, H1T2, and HILS1), and one oocyte-specific H1 variant, called H1oo [3,4]. Among others, Millán-Ariño et al. showed that the H1.2 variant is less abundant in transcription start sites of transcriptionally inactive loci and is enriched in guanine-cytosine (GC)-poor genomic regions located in close proximity to lamina-associated domains [5]. Conversely, H1.0 and H1X variants are abundant in CpG islands and gene-rich genomic regions [5]. Importantly, H1.0 accumulates in terminally differentiated cells [4]. Li  differentiated fibroblasts [6]. Importantly, H1.5 was found in transcriptionally silent loci encoding membrane-associated proteins in differentiated cells, but it was not the case with pluripotent human ESCs.
For regulation of chromatin structure, DNA replication, and transcription, posttranslational modifications (PTMs) of core histones are functionally significant, and a linker histone H1 is not an exception. The following H1 modifications were described: methylation, acetylation, ubiquitination, formylation, poly-ADP ribosylation, and phosphorylation [7]. H1 phosphorylation (H1ph) is responsible for chromatin condensation and cell cycle regulation [8]. In mammalian cells, the lowest phosphorylation level of H1 is in the G1 phase of the cell cycle, but H1ph increases in S/G2 phases, with the maximum peak in metaphase [7,9]. Interestingly, dephosphorylation of H1 variants precedes DNA oligonucleosomal fragmentation, representing a late-stage hallmark of apoptosis [10]. So, it is evident that H1 phosphorylation status is crucial in many cellular processes, including cell cycle regulation, cell differentiation, and apoptosis. Importantly, H1.2 phosphorylation also affects the p53-mediated DNA damage response [11,12], which should also be connected with the p53-dependent apoptotic pathway.
From the view of PTMs of histone H1, it is essential to mention that specific methylation sites of H1 were also detected. Ohe et al. identified Ezh2-dependent methylation of H1.4K26 in mammalian cells [13]. Due to the catalytic activity of Ezh2 (a component of the Polycomb group PRC2 complex), it is expected that this H1 modification has a silencing effect [14]. In addition to this observation, G9a histone methyltransferase (HMT), generally mediating dimethylation of histone H3 at the lysine 9 (H3K9me2), can also methylate human histone H1.4 at the position of lysine K26 in vivo. This process is essential for chromatin condensation, namely, heterochromatin formation. In addition, Terme et al. showed a functional link between H1.4K26 methylation and H1S27 phosphorylation and extended this to H1.4K26 acetylation [15]. Among others, H1.4K34 acetylation regulates the recruitment of H1 to chromatin and contributes to efficient transcription processes [16].
Based on the observation mentioned above, we tested a hypothesis of whether there is a specific nuclear arrangement of phosphorylated histone H1 in distinct cell cycle phases. Additionally, we analyzed a functional link between seemingly unrelated factors such as phosphorylated histone H1 and a marker of heterochromatin, a core histone H3 trimethylated at lysine 9. We studied if depletion of histone deacetylase 1 (HDAC1) affects the nuclear arrangement of linker histone H1 and changes the epigenetic profile of H1 variants, including H1.0, H1.1, H1.2, H1.3, H1.4, and H1.5.
We acquired images with Leica TCS SP8X SMD confocal microscope (Leica Microsystem, Wetzlar, Germany), equipped with HC PL APO 63×/1.4 oil CS2 objective. Image acquisition was performed using a white light laser (WLL) and 405 nm laser with the following excitation wavelengths (λ ex. ) for fluorophores mentioned above: Cy3 (λ ex. = 554 nm); Cy5 (λ ex. = 649 nm) and DAPI counterstain (λ ex. = 358 nm). DAPI was visualized by a 405 nm laser connected to Leica TCS SP8X SMD confocal microscope. The axial and lateral resolution values of HC PL APO 63×/ 1.4 oil CS2 objective are provided in Table 1 (below). Histones were isolated as described by [22]. Briefly, cells were washed twice with ice-cold phosphate-buffered saline (PBS), re-suspended in lysis buffer (80 mmol/L NaCl, 20 mmol/L EDTA, 1% Triton X-100, 45 mmol/L sodium butyrate, and 100 mmol/L phenylmethylsulfonyl fluoride), incubated for 20 min on ice, and centrifuged at 2000× g for 8 min. Pellets were re-suspended in 900 µL ice-cold 0.2 mol/L H 2 SO 4 and incubated for 2 h, shaking at 4 • C. After centrifugation at 16,000× g, proteins were precipitated from the supernatant with trichloroacetic acid to a final concentration of 25% and incubated for 30 min on ice. After centrifugation at 5000× g for 30 min at 4 • C, the pellet was washed with 50 mmol/L HCl in acetone, then with 100% acetone, and subsequently dissolved in water. The protein concentration was determined by the Bradford assay (Bio-Rad, Hercules, CA, USA). A 150 µg portion of the histone sample was diluted with acidified acetonitrile solution (80% acetonitrile, 2% formic acid). MS Phospho-mix standards (MSP1L, MSP2L, and MSP3L; Merck, Czech Republic) were added to the samples. Phosphopeptides were enriched using the Pierce Magnetic Titanium Dioxide Phosphopeptide Enrichment kit (#88811, Thermo Fisher Scientific, USA) according to manufacturer protocol. Eluates were concentrated under vacuum to 3 µL, diluted with 90 µL of 0.1% formic acid, and purified using a Hypersep SpinTip C-18 column (Thermo Fisher Scientific, USA).

LC-MS/MS Analysis, Database Search, and Data Evaluation
Phospho-enriched histone peptides represented by 3-4 replicates of each condition were analyzed by LC-MS/MS as described by [22]. The LC-MS/MS equipment consisted of an RSLCnano system, equipped with an Acclaim Pepmap100 C18 analytical column (3 µm particles, 75 µm × 500 mm; Thermo Fisher Scientific), coupled to an Orbitrap Elite hybrid spectrometer (Thermo Fisher Scientific) equipped with a Digital PicoView 550 ion source (New Objective) using PicoTip SilicaTip emitter (FS360-20-15-N-20-C12), and Active Background Ion Reduction Device. The mobile phase consisted of 0.1% formic acid in water (A) and 0.1% formic acid in 80% acetonitrile (B), with the following proportions of B: 1% B for 16 min at 600 nL/min to concentrate peptides, then (with a switch to 300 nL/min) 1-13% B over 20 min, 13-33% B over 25 min, 33-56% B over 20 min and 56-80% B over 5 min followed by isocratic washing at 80% B for 5 min. The analytical column outlet was directly connected to the ion source of the MS. MS data were acquired using a data-dependent strategy selecting up to the top 10 precursors based on precursor abundance in a survey scan (350-2000 m/z). The resolution of the survey scan was 60,000 (400 m/z) with a target value of 1 × 10 6 , one microscan, and a maximum injection time of 1000 ms. HCD MS/MS spectra were acquired with a target value of 50,000 and resolution of 15,000 (400 m/z). The maximum injection time for MS/MS was 500 ms. Dynamic exclusion was enabled for 45 s after one MS/MS spectrum acquisition, and early expiration was disabled. The isolation window for MS/MS fragmentation was set to 2 m/z.
The RAW mass spectrometric data files were analyzed using Proteome Discoverer software (version 1.4; Thermo Fisher Scientific, USA) with the in-house Mascot search engine to compare acquired spectra with entries in the in-house histone Mus musculus database. Settings for all searches included trypsin enzyme specificity and up to five missed cleavages. The following variable modifications were set for searches: methyl (R, K), di-methyl (K), tri-methyl (K), acetyl (K, protein N-term), and phosphorylation (S, T, Y). Mass tolerances of peptides and MS/MS fragments for MS/MS ion searches were 7 ppm and 0.03 Da, respectively. Manual peak labeling and calculation of the peak area corresponding to each precursor ion from the extracted ion chromatograms (XICs) were done via Skyline 21.1 software (Seattle, WA, USA). A spectral library was created using the Proteome discoverer platform (version 1.4; Thermo Fisher Scientific, USA). Only peptides with statistically significant peptide scores (p < 0.01) were included. Rank 1 peptides with Mascot expectation value <0.01 and at least six amino acids were considered. Peptide identifications were manually verified, and quantitative data were evaluated using Skyline 21.1 software. Precursor areas of phosphorylated histone peptides were normalized to signals of MS Phosphomix standards.
The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE [23] partner repository with the dataset identifier PXD033544.

Localization of Phosphorylated Histone H1 and H3K9me3 Is Cell Cycle Specific
By immunofluorescence, combined with confocal microscopy, we observed in HeLa Fucci cells that the distribution and density of phosphorylated histone H1, similarly to H3K9me3, were identical in the G1, S, and G2 phases of the cell cycle ( Figure 1A,B). In the interphase of wild-type mouse ESCs and HDAC1 double-knockout cells, H1ph occupied nucleoli and appeared in the nucleoplasm outside H3K9me3-dense chromocenters (Figure 2(Aa-Ac)). Interestingly, DAPI-dense DNA surrounding a compartment of nucleoli did not colocalize with phosphorylated H1 that occupied a very central region of nucleoli (Figure 2(Ac)). A low level of H1ph was on the nuclear periphery ( Figure 2(Ba,Bb)) that was characterized by a high density of heterochromatin marker H3K9me3 (Figure 2(Ba,Bc)). We observed that HDAC1 deficiency did not change the nuclear distribution of H1 phosphorylated form and H3K9me3 (Figure 2(Aa-Bc)).

Localization of Phosphorylated Histone H1 and H3K9me3 Is Cell Cycle Specific
By immunofluorescence, combined with confocal microscopy, we observed in HeLa Fucci cells that the distribution and density of phosphorylated histone H1, similarly to H3K9me3, were identical in the G1, S, and G2 phases of the cell cycle ( Figure 1A,B). In the interphase of wild-type mouse ESCs and HDAC1 double-knockout cells, H1ph occupied nucleoli and appeared in the nucleoplasm outside H3K9me3-dense chromocenters (Figure 2Aa-c). Interestingly, DAPI-dense DNA surrounding a compartment of nucleoli did not colocalize with phosphorylated H1 that occupied a very central region of nucleoli (Figure 2Ac). A low level of H1ph was on the nuclear periphery (Figure 2Ba,b) that was characterized by a high density of heterochromatin marker H3K9me3 (Figure 2Ba,c). We observed that HDAC1 deficiency did not change the nuclear distribution of H1 phosphorylated form and H3K9me3 (Figure 2Aa-Bc).  Next, we found the most intensive clustering of H1ph in specific regions of mitotic cells ( Figure 3B). Importantly, in the prophase of mitosis, H1ph was concentrated in significant clusters ( Figure 3A,B). In prometaphase, H1ph was highly dense in the whole cellular content ( Figure 3B, and File S1). In anaphase, tiny H1ph foci occupied space outside condensed DAPI-dense chromosomal DNA ( Figure 3B, and File S1). Additionally, in telophase, we observed a barely detectable level of H1ph, but condensed chromosomes and their DNA were surrounded by H3K9me3 ( Figure 3B, and File S1). Next, we found the most intensive clustering of H1ph in specific regions of mitotic cells ( Figure 3B). Importantly, in the prophase of mitosis, H1ph was concentrated in significant clusters ( Figure 3A,B). In prometaphase, H1ph was highly dense in the whole cellular content ( Figure 3B, and File S1). In anaphase, tiny H1ph foci occupied space outside condensed DAPI-dense chromosomal DNA ( Figure 3B, and File S1). Additionally, in telophase, we observed a barely detectable level of H1ph, but condensed chromosomes and their DNA were surrounded by H3K9me3 ( Figure 3B, and File S1).

Discussion
Histone H1.0 variant, specific for terminally differentiated cells, is enriched in the repetitive sequences of ribosomal genes (summarized by [24]). Moreover, Okuwaki et al. showed the enrichment of the H1.0 variant in the intergenic regions of ribosomal genes, while H1.0 is less abundant in their promoters and coding regions [25]. This observation fits well with our data showing high relative phosphorylation of histone H1 in a compartment of nucleoli studied in mESCs (Figure 2(Ab,Ac)). In general, a high density of phosphorylated histone H1 was found in genomic regions with a pronounced RNA synthesis [26]. In this case, nucleoli, as sites of ribosomal genes' transcription, can be considered the biggest transcription factories [27]. Therefore, an occurrence of phosphorylated H1 inside nucleoli fits well with the theory of Zheng et al., showing that site-specific H1 phosphorylation in interphase facilitates transcription that is mediated by both RNA polymerases I and II. These authors observed that especially H1.2/H1.4 (e.g., pS187-H1.4) variants occupy nucleoli [26].
In prophases and prometaphase of mESCs, we have found a high level of phosphorylated H1, while anaphase and telophase were characterized by an absence of H1 phosphorylation. In the G1, S, and G2 phases of the cell cycle, the distribution profile of H1ph was identical (Figures 1 and 3A,B and File S1). Additionally, Sarg et al. summarized that the individual H1 subtypes differ in their degree of phosphorylation during the cell cycle [28]. Zheng et al. in human cells showed specific sites in histones H1.2 and H1.4 that were phosphorylated only during mitosis or during both mitosis and interphase [26]. Green A et al. documented that interphase H1 phosphorylation is the most pronounced in the G1 or early S phase of the cell cycle [29]. Talasz et al. showed a low level of phosphorylated H1 in the G1 phase and a peak of H1 phosphorylation in the G2 phase and mitosis [9]. These authors additionally published that phosphorylation of histone H1.5 Ser17 appears early in the G1 phase, while the Ser172 phosphorylation is activated later in the G1 phase nuclei [30]. Importantly, histone H1.5 Thr10 phosphorylation exclusively occurred in mitotic cells. These data document significant distinctions in the abundance of histone H1 variants and their post-translational modifications in cell cycle phases, including mitosis.
Using mass spectrometry, in mouse ECSs, we found specific phosphorylation and acetylation sites of histone H1. In this case, we analyzed histones H1.0, H1.1, H1. Here, we observe in mESCs that the histone H1.0 variant is acetylated at threonine 2 on the N-terminal domain (Figure 4), but HDAC1 depletion did not change this acetylation profile. Interestingly, deficiency in HDAC1 affected phosphorylation of H1.2, H1.3, and H1.4 ( Figure 5). These data support the existence of cross-talk between acetylation and phosphorylation. For example, in bacteria, van Noort et al. showed that deletion of the two N-acetyltransferases affects protein phosphorylation [32]. Additionally, Uhart and Bustos documented a link between phosphorylation and lysine acetylation in human cells. In this case, it seems to be likely that functional cross-talk between phosphorylation and acetylation could apply not only to non-histone but also to histone proteins [33].
Several studies showed that H1 phosphorylation is a feature of decondensed chromatin rather than highly condensed heterochromatin [6,34]. This claim also supports our observation that phosphorylated histones H1 appear in transcriptionally active nucleoli and in the nuclear interior that is considered to contain more relaxed and transcriptionally active chromatin [35]. We found that the nuclear periphery, abundant in silencing epigenetic marker H3K9me3, is characterized by a low density of phosphorylated histone H1 (Figure 2(Ba-Bc), arrow in the direction of the nuclear periphery). Similarly, highly condensed chromocenters (clusters of centromeric heterochromatin) were absent of H1 phosphorylation (Figure 2(Ba-Bc), arrow in the direction to selected chromocenters). These data fit well with the theory that histone H1 phosphorylation affects chromatin condensation and appears in the transcriptional active genomic region to potentiate transcription efficacy [26].

Conclusions and Future Directions
Together, we showed that the nuclear distribution of phosphorylated histone H1 and H3K9me3 was identical in the G1, S, and G2 phases of the cell cycle. In mitosis, the highest level of H1 phosphorylation was in prophase and prometaphase with a low level of H3K9me3, while H1ph was barely detectable in anaphase and especially in telophase, characterized by an appearance of H3K9me3 around mitotic chromosomes. Significantly, the N-terminal domain of H1 was post-translationally modified, and HDAC1 depletion in mESCs affected phosphorylation of H1.2, H1.3, and H1.4 histone variants. However, HDAC1 deficiency did not change the acetylation status of linker histones.
Our data imply a future direction of research in which cross-talk between histone acetylation, phosphorylation, and methylation should be investigated. We should clarify if an absence of one epigenetic mark (e.g., specific histone acetylation) could be replaced by another but seemingly unrelated epigenetic trait (e.g., histone phosphorylation).
Author Contributions: S.L. was responsible for immunofluorescence, Western blot analysis, FLIM-FRET analysis, and deconvolution of confocal images. E.B. wrote this paper and performed confocal microscopy for Figures 2 and 3. G.L. is responsible for mass spectrometry data and statistical analysis of MS data. All authors have read and agreed to the published version of the manuscript.