Histone Modification Is Involved in Okadaic Acid (OA) Induced DNA Damage Response and G2-M Transition Arrest in Maize

Histone modifications are involved in regulation of chromatin structure. To investigate the relationship between chromatin modification and cell cycle regulation during plant cell proliferation, Okadaic acid (OA), a specific inhibitor of serine/threonine protein phosphatase, was applied in this study. The results showed that OA caused the cell cycle arrest at preprophase, leading to seedling growth inhibition. Western blotting assay revealed that the spatial distribution of phosphorylation of Ser10 histone H3 tails (H3S10ph) signals was altered under OA treatment. Reactive oxygen species (ROS) was found to be at higher levels and TdT-mediated dUTP nick end labeling (TUNEL) assay displayed DNA breaks happened at the chromatin after treatment with OA, companied with an increase in the acetylation of histone H4 at lysine 5 (H4K5ac) level. From these observations, we speculated that the alteration of the spatial distribution of H3S10ph and the level of H4K5ac was involved in the procedure that OA induced DNA breaks and G2-M arrested by the accumulation of ROS, and that the histone H3S10ph and H4K5ac might facilitate DNA repair by their association with the chromatin decondensation.


Introduction
Recent data shows that epigenetic marks including DNA and histone modification are involved in chromatin DNA replication and genetic transmission during the cell cycle [1]. Study of dynamics of epigenetic marks through cell cycle will be helpful for understanding mechanisms and means that ensure the correct information transmission in cell division. During last few years numerous research focusing on the role of histone modification have been reported [2,3]. N-terminus tails of histones could be catalyzed by various enzymes, leading to different modifications, such as acetylation, methylation and phosphorylation [4]. These histone modifications alter chromatin structure and accessibility of transcription complexe, and then regulate gene expression [5]. For instance, hyperacetylated histone is generally located on transcriptional active chromatin region and deacetylated histone always appears on repressive were distributed into two groups, which were treated with distilled water and 100 nM OA respectively. After 24 h, 48 h and 72 h, leaves from different conditions were collected and processed for different assays.

Growth analysis
Images of seedlings were captured to measure the length of maize leaves and roots. The length of each sample was quantified using the software ImageJ (NIH, Bethesda, MD, USA).

Flow Cytometry
Nuclei were isolated based on the described method [30]. The samples were chopped rapidly with nucleus isolation buffer [10 mM MgSO4, 50 mM KCl, 5 mM Hepes, 1 mg /ml dithiothreitol (Sigma, St. Louis, MO) and 0.2% Triton X-100] and filtered through a 33 μm nylon mesh. The nuclei were fixed in 4% paraformaldehyde for 30 min at 4°C, then were precipitated (200 g, 10 min, 4°C), resuspended in the isolation buffer and stained with Propidium Iodide (PI). The cell-cycle profile was determined with a FACSCalibur flow cytometer (Becton Dickinson, San Jose, CA, USA) equipped with an argon-ion laser, using the 488 nm laser line for excitation. The results of Flow Cytometry were quantified using the software Flowjo 7.6.

Mitosis phase counting
The edge of leaf samples was collected and fixed in Carnoy's Fluid (a solution of ethanol and gladual acetic acid in a 3:1 volume ratio) at 4°C for 30 min. Afterwards, leaf edges were washed several times with 70% ethanol, followed by enzymolysis in a mixture of 2% pectinase and 4% cellulase at 37°C for 2 h. The digested leaf edge was squashed in a drop of 1×PBS on a slide below a coverslip. After removal of the coverslip, all slides were stained with DAPI (Sigma, St. Louis, MO, USA), mounted with Vectashield (Vector labs, Burlingame, CA, USA) and examined under a fluorescence microscope (Olympus BX-60). Images were captured with a CCD monochrome camera Sensys 1401E. Microscope settings and camera detector exposure times of each channel were kept constant respectively. More than 1000 nuclei were measured for each treatment group.

Immunostaining
Nucleus isolation and immunostaining were performed according to the described method [31]. Nuclei were spread on a slide, incubated with 3% bovine serum albumin at 37°C for 1h, followed by overnight incubation at 4°C with primary antibody and 2h incubation at 37°C with secondary antibody. Secondary antibodies were conjugated with FITC. All slides were stained with DAPI, mounted with Vectashield and examined under an Olympus BX-60 fluorescence microscope with filter blocks for DAPI and fluorescein. Images captured with a CCD monochrome camera Sensys 1401E were pseudo-colored and merged using the software Meta-MorphH 4.6.3. Microscope settings and camera detector exposure time of each channel were kept constant respectively. More than 500 nuclei were detected for each treatment group. For both control and treated groups, three independent immunostaining experiments were performed with each antibody.

TUNEL assay
TUNEL staining was performed using the Deadend™ Fluorometric TUNEL System Kit (G3250). Nucleus isolation was performed according to the method described previously [31]. After centrifugation under 300 g for 10 min at 4°C, nuclei were resuspended in the isolation buffer and then fixed in 4% paraformaldehyde (pH 7.3) for 20 min. Next, the mixture was centrifuged under 300 g for 10 min at 4°C. Nuclei were resuspended in 80 μl equilibration buffer, followed by incubation for 5 min. After centrifugation under 300 g for 10 min, terminal deoxynucleotidyl transferase (TdT) reaction buffer (containing 45 μl incubation buffer, 5 μl of a fixed concentration buffer of dATP, dGTP, dCTP and fluorescein-12-dUTP, and 1μl TdT) were applied to serial sections for 1 h at 60°C, then the stop buffer (EDTA) was added. Nuclei were spread on a slide and all slides were stained with DAPI (Sigma, St. Louis, MO, USA), mounted with Vectashield (Vector labs, Burlingame, CA, USA) and examined under an Olympus BX-60 fluorescence microscope with filter blocks for DAPI and fluorescein. Images captured with a CCD monochrome camera Sensys 1401E were pseudo-colored and merged using the software MetaMorph 4.6.3 (Universal Imaging Corp., Downingtown, PA, USA).

Quantitative real-time PCR
The first-strand cDNA was synthesized from total RNA isolated from leaf samples with Trizol reagent (Invitrogen, Carlsbad, CA, USA) by using Revert Aid First Strand cDNA Synthesis Kit (Fermentas, Canada). Quantitative real-time PCR was carried out using SYBR1 Green Real-time PCR Master Mix (TOYOBO, Japan) in a StepOne Plus real-time PCR system (Applied Biosystems, Carlsbad, USA) with the following cycling conditions: 94°C for 2 min, followed by 40 amplification cycles at 94°C for 5 s, 56°C for 15 s, and 72°C for 20 s. All primers used for quantitative real-time PCR of topoisomerase genes are listed in Table 1. Fluorescence data were acquired at the 72°C step and during the melting-curve program. To ensure the amplification of a single PCR product for each gene, preliminary experiments were conducted. Relative expression levels were normalized with the maize β-actin gene [32]. Quantitative PCR primers were designed to amplify ca. 200 base pair (bp) fragments. RT-PCR was repeated three times for each sample from three independent experiments.

Western blotting assay
Proteins were extracted by grinding leaf samples in liquid nitrogen and resuspended in extraction baffer (100 mM Tris-HCl pH 7.4, 50 mM NaCl, 5 mM EDTA and 1 mM PMSF). Protein were loaded onto a 12% SDS-PAGE gel and separated by electrophoresis and then transferred to nitrocellulose membranes which were incubated with primary and secondary antibodies step by step [30]. Detection was performed using alkaline phosphatease (AP) conjugated antirabbit IgG antibody and chemiluminescence visualization. The mean gray value of the signals of H3S10ph and H4K5ac was measured with ImageJ 1.48 software. Abundance index was calculated as H3S10ph or H4K5Ac band intensity/H3 band intensity. Western blots were repeated three times. Histone H3 was used as equal loading control.

ROS and related enzyme activity detection assays
Leaves were differently ground in PBS buffer (0.04575 mol/L Na2HPO4, 0.00425 mol/L NaH2PO4, pH 7.8) mixed with quartz sands. Supernatant was collected after centrifugation (8000 g, 20 min) for detection. Supernatant was used for superoxide dismutase (SOD), peroxidase (POD) and catalase (CAT) activity detection. SOD assay was performed according to Wong et al [33]. One unit of SOD was equal to amount which caused a 50% decrease of NBT reduction. POD assay was performed according to Wu et al [34]. Enzymatic activity was determined by catalyzing decomposition of H 2 O 2 in the presence of guaiacol. One unit of POD was equal to that amount which decreased the absorbance at 470 nm by 0.1 per minute. SOD and POD activity were expressed as U g-1 FW. CAT was assayed according to the reported procedure [35]. The decomposition degree of H 2 O 2 was detected at 240 nm wavelength. The concentration of the atomic oxygen radical anion (O 2 -) was measured according to Li et al [36]. The supernatant was added to a mixture composed of 10 mM hydroxylamine hydrochloride and 0.05 M PBS buffer (pH 7.8). After incubation, 7 mM α-naphthylamine and 17 mM sulfanilamide were added to the mixture. The light absorbance was measured at 530 nm wavelength.

Statistical analysis
Each data and error bars were calculated from three independent experiments. Data in this manuscript was analysed for significant differences between the treated and control groups by means of t-test. It was considered statistically significant when P < 0.05.

Suppression of maize seedling growth and alteration of cell cycle by Okadaic acid
To investigate the relationship between histone phosphorylation and chromatin condensation during plant cell proliferation, Okadaic acid (OA), a specific inhibitor of serine/threonine protein phosphatase, was applied to treat maize seedlings. First, we valued the growth state of maize seedlings during OA treatment by measuring primary root length and leaf length. The results showed that primary root and leaf length of treated groups was shorter than that in the control group, revealing the suppression of seedling growth caused by Okadaic acid (Fig 1A-1C). Furthermore, cell cycles of both group seedlings were analyzed using Flow Cytometry and Flowjo software. The results showed that the rates of G1 phases were decreased and the rates of G2 phase significantly increased after treatment with OA compared with the control group (Fig 1D and 1E). The inhibition effect of Okadaic acid was increasing significant within the treated period because the G1/G2 ratio of the control group was gradually raised while the seedlings exposed to OA revealed a constant and lower level (Fig 1F). These data suggested that Okadaic acid could arrest of the cell cycle of maize seedlings leading to growth inhibition.
Effect of Okadaic acid on the phosphorylation of H3S10 in the overall level qPCR analysis of PP1 was performed showing that OA actually had an effect on the enzymes that it targets (S1 Fig). In addition, we performed western blotting assay with proteins from the control and treated group to investigate the state of histone phosphorylation during the arrest of cell cycle, using specific antibodies against H3S10ph and H3. The result showed that the total H3S10ph levels of the genome were unchanged in both control and treated groups (Fig  2A and 2B).

Okadaic acid affects spatial distribution of H3S10ph on chromosomes in mitosis
To further investigate the alteration of H3S10ph on chromosome structures, we detected the distribution of the H3S10ph signals throughout cell cycle following OA treatment by Immunostaining assay. The results showed that H3S10ph signals appeared at preprophase and disappeared in telophase in the control and treated group, consistent with the pervious reported results (Fig 3A). However, interestingly, OA treatment obviously changed the spatial distribution of H3S10ph on chromosomes, as the two regular H3S10ph signals located close to the centromeres turn to the dispersion signal, revealing the way that OA affect phosphorylation of H3S10 (Fig 3A and 3B).

The cell cycle is arrested at preprophase due to DNA damage
To determine the exact phase of the cell cycle arrested by Okadaic acid, mitotic phase was divided into two phases, phase I representing preprophase and phase II including from prophase to telophase (Fig 4A). Analysis of 1000 cells revealed that after treating with OA, cells in phase I were significantly increased and cells in phase II were decreased compared with the control group, proving that the cell cycle was exactly arrested at preprophase ( Table 2; Fig 4B  and 4C). Since the cell cycle can be delayed at G 2 -phase checkpoint due to DNA damage, TUNEL assays were performed. TUNEL assay is used for detecting DNA breaks. The treated group displayed the DNA damage signal in TUNEL compared with the control group after 3d of exposure, suggesting that genomic DNA did break in the cell cycle arrest (Fig 5). Combining these results with the chromosomes decondensation throughout M phase, we tested expression of topoisomerase genes, which plays a key role at the transition point from chromatin to chromosome, as well as the DNA repair. The results showed that expression of topoisomerase genes in treated groups was higher than that in the control group (Fig 6).

Histone H4 lysine 5 acetylation is elevated in the DNA damage response after OA treatment
It has been demonstrated that histone tail acetylation is required directly for DNA repair [16]. Therefore, western blotting assay was performed to investigate the state of various histone modifications during the arrest of cell cycle, using modification antibodies against H3ac, H3K9ac, H4K5ac, H3K4me2 and H3. Although most of histone modification levels kept stable,  H4K5ac was significantly increased under OA treatment compared with the control group ( Fig  7A and 7B), which agree with the reported result that H4 acetylation is interrelated with DNA repairs during cell cycle. Further immunostaining of nuclei also showed that H4K5ac signals were enhanced after treatment with OA ( Fig 7C).

ROS accumulation is involved in the arrest of cell cycles
Oxygen pressure is thought to be related with the cell cycle arrest via DNA damage. In order to verify whether ROS accumulation was involved in the preprophase arrest induced by OA, we measured the concentration of O 2 in seedling leaves. The results showed that the concentration of O 2 kept higher levels in the treated group than that in control group, suggesting the involvement of ROS in the preprophase arrest ( Fig 8A). Furthermore, activities of ROS-related enzymes, catalase (CAT), superoxide dismutase (SOD) and peroxidase (POD), were found to be raised in treated groups compared with the control group (Fig 8B and 8D). To further confirm the involvement of ROS in the cell cycle arrest, antioxidant thiourea was tested for reducing concentration of O 2 molecule in leaf cells of treated groups. OA exposure for 2 d caused an increase of O 2 in leaves, and OA plus thiourea treatment did not cause significant changes in O 2 concentration (Fig 9A), suggesting that thiourea could decrease OA-induced ROS accumulation. Cell cycle analysis showed that additional oxygen stress induced the cell cycle arrest could be partly released by the addition of antioxidant thiourea (Fig 9B), proving that O 2 molecule played an important role in the preprophase arrests induced by OA.

Discussion
OA was applied to investigate the relation between histone modification and the cell cycle regulation in maize seedlings in this study. H3S10 phosphorylation is a key histone modification in mitosis and starts in early prophase and sustains until telophase [10]. The dynamic process of H3S10ph in mitosis, associated with chromosome remodeling, is conserved during evolution in eukaryotes, which suggested this modification is very important for chromatin remolding.
In this study the maize seedling growth was inhibited and cell cycle was arrested, accompanied with the dispersed signal distributions of H3S10ph on chromosomes after OA treatment. The Histone Modification and OA Induced G 2 -M Cell Cycle Arrest results further provided evidence that the cell cycle was arrested at preprophase, the transition period from chromatin to chromosome. Treatment of tobacco cells with OA resulted in similar chromosome condensation abnormalities, and mitosis was arrested at G2-M transition [37]. Chromatin become condensed gradually in interphase and proprephase, and highly condensed chromosomes facilitate accurate segregation of the genetic material to daughter cells [38]. We speculated that OA treatment could affect chromatin structure by means of the change in H3S10 phosphorylation signal, consequently causing cell cycle arrest in maize seedlings. It has been reported that histone modification inhibitors could increase ROS production [39,40]. OA can both affect H3S10 phosphorylation and ROS/MAPK signaling pathways in vitro [41,42]. The ROS mediated cell cycle arrest via DNA damage responses, and the DNA damage response signal-transduction pathway regulates cell cycle progression [43]. In the research of cell cycle alterations induced by PM fraction (PM2.5; aerodynamic diameter 2.5 μm), ROS formation was reported to be involved in the cell cycle delay [44]. Similarly, our results revealed that OA induced ROS accumulation in the cell and the use of  deoxidizer reduced the content of O 2 and relieved the arrest of cell cycle induced by OA, confirming that the ROS participated in the preprophase arrest. Furthermore, it has been reported that eukaryotic DNA topoisomerase participate in the final stage of chromosome condensation [45,46]. In OA-treated maize seedlings, expression of DNA topoisomerase family all kept relatively higher levels than that in the control group, suggesting that the cells were undergoing a repairing process with the aim of releasing cell cycle from decondensation caused by DNA damage. However, the packaging of DNA into chromatin is likely to be a barrier to access of DNA repair related complex. The previous works indicated that histone hyperacetylation is concerned with responses to DNA damage induced by ROS accumulation [47]. Histone acetylation is related to DNA repair [16,48], because decondensation of the nucleosome structure mediated by histone acetylation renders chromatin more accessible to nuclear protein factors [49]. Histone acetylation can reduce the interactions between histones and DNA, thereby relaxing the DNA structure and easily recruiting DNA repair proteins for DNA repair [50]. So we speculated that the enrichment of H4K5ac after OA treatment is associated with the chromatin decondesation for efficient repair of DNA breaks through the genome, which consequently also cause preprophase arrest in maize.
This study demonstrates a mechanism of OA-induced alteration of cell cycle progression via augmented accumulation of ROS and the downstream DNA damage in the maize seedlings, and the changes in histone H3S10 phosphorylation and H4K5 acetylation signal are involved in this progression. The cell cycle is arrested at proprephase, the transition point from chromatin to chromosome, allowing more time for cells to detoxify ROS and to repair damaged DNA.