Histone H4 Methyltransferase Suv420h2 Maintains Fidelity of Osteoblast Differentiation

CELL CULTURE

MC3T3-E1, a clonal pre-osteoblastic cell line derived from newborn mouse calvaria and MLO-A5, a mature osteoblast to pre-osteocyte like cell line established from the long bones of 14-day-old mice (kindly donated by Lynda Bonewald, University of Missouri—Kansas City) were cultured in a humidified atmosphere with 5% CO₂ at 37°C and were sub-cultured twice per week using Trypsin (TrypLExpress, Gibco) before achieving confluence. MC3T3-E1 and MLO-A5 cells were cultured in α-minimum essential medium (α-MEM; Gibco) containing 100 U/ml penicillin, and 100 µg/ml streptomycin (Pen-strep, Gibco); for MC3T3-E1 culture media was supplemented with 10% heat inactivated fetal bovine serum (FBS; Atlanta Biologicals, Flowery Branch, GA) and MLO-A5 cells were cultured on rat-tail derived collagen Type I (Roche) coated dishes (final concentration of 0.15 mg/ml) and culture media was supplemented with 5% fetal calf serum (FCS, Hyclone) and 5% calf serum (CS, Hyclone).

Differentiation of MC3T3-E1 cells was induced with 50 µg/ml ascorbic acid (Sigma–Aldrich) and 5 mM β-glycerophosphate (Sigma–Aldrich) in medium containing 10% FCS. Cells were seeded in culture dishes at a density of 10,000 cells/cm². Before cell differentiation was induced, the culture medium was changed.

Suv420H2 SIRNA TRANSFECTIONS

For Suv420h2 depletion by siRNA, MC3T3-E1 cells were seeded at 10,000 cells/cm² in diverse multiwell plate formats for protein and RNA collection as well as for alkaline phosphatase activity measurements and for extra cellular matrix (ECM) deposition and mineralization assessment by picro sirius red staining and by alizarin red staining, respectively. Six hours after seeding, cells were transfected with scrambled or Suv420h2 specific siRNA (# L-052050-00, GE Dharmacon) using Lipofectamine RNAiMAX (Invitrogen) as described by the supplier. One day after transfection, medium change was performed and osteoblastic induction was induced as described above. Protein was isolated 3 days after transfection, RNA was isolated 3, 6, and 10 days after transfection, ECM deposition was analyzed 7 days after transfection, alkaline phosphatase activity was measured 10 days after siRNA transfection and ECM mineralization was analyzed 29 days after siRNA transfection.

PROTEIN ISOLATION AND IMMUNOBLOTTING

Whole cell protein extracts were prepared using SDS sample buffer (3% SDS and 125 mM Tris–HCl, pH = 6.8) and heated at 95°C for 5 min. For immunoblotting analysis, 5 µg of total protein extracts were separated on 11% SDS poly-acryl amide gels, transferred to polyvinylidene difluoride membranes (Millipore), blocked for 1 h with 5% non-fat dry milk in TBST buffer (137 mM NaCl, 2.7 mM KCl, 19 mM Tris, 0.1% Tween). The following primary antibodies were used: rabbit-anti-Gapdh (#5174S, Cell Signaling), mouse-anti-H4K20me3 (#39672, Active Motif), rabbit-anti-histone H4 (#61200, Active Motif), and rabbit-anti-Suv420h2 (#ab91224, Abcam). Washing was performed with TBS buffer containing 0.1% Tween. Subsequently, binding of the secondary antibody (anti-rabbit IgG/anti-mouse IgG horseradish peroxidase-coupled) (Santa Cruz) diluted 1:10,000 in 5% non-fat dry milk in TBST followed by detection with the Amersham ECL Prime Western Blot Detection Reagent (Ge Dharmacon) was carried out as described by the supplier. Chemo-luminescence was measured with an image acquisition system (Chemidoc Touch, Biorad). Measurements are given as means of three immuno-blots and representative blots are shown.

ISOLATION OF RNA AND REVERSE-TRANSCRIPTASE QUANTITATIVE POLYMERASE CHAIN REACTION (RT-qPCR)

Total RNA was extracted using the Direct-Zol RNA Mini Prep Kit (Zymo) following the supplier's instructions. cDNA was synthesized from ∼0.5 µg RNA using the SuperScript III first strand synthesis system (Invitrogen) as described by the supplier. The resulting cDNAs were subjected to quantitative PCR amplification with a real-time cycler using the QuantiTect SYBR-Green PCR Kit (Qiagen) for the genes Suv39h1, Suv39h2, Suv420h1, Suv420h2, Sp7, Bglap2, Alpl, Runx2, Ezh1, Ezh2, Ccnb2, Ibsp, Col1a1, and Gapdh (for primers, see Table I). SYBR-Green RT-qPCR was started with an initial denaturation step at 95°C for 15 min and then continued with 45 cycles consisting of 30 s denaturation at 95°C, 30 s annealing at 60°C, and 30 s extension at 72°C in a CFX384 real time system machine (Bio-Rad). All RT-qPCR assays were performed in triplicate and expression was evaluated using the comparative quantification method [Pfaffl, 2001].

GENE EXPRESSION PROFILING OF MOUSE CELLS DERIVED FROM FRACTURE CALLUS AND EZH2 CKO CALVARIA

We performed gene expression profiling by next generation RNA sequencing (RNA-seq) of calvaria bone or callus-derived cells in a mouse model for long bone healing derived from fracture callus. Calvarial bone was derived from either wild-type mice, or mice in which the Ezh2 gene was conditionally ablated using a Prrx1-Cre driver as previously described [Dudakovic et al., 2015]. Cells derived from the callus were isolated using a lineage-traced method by recombination of a floxed stop allele and activation of downstream red fluorescent protein (tomato) upon activation of a SMACreERT2 reporter [Matthews et al., 2014]. This genetic approach permits marking of all cells derived from SMACre expressing mesenchymal stem cells, which are rendered Tomato positive, while Sma-negative cells that include hematopoietic cells remain Tomato-negative. Callus cell preparations were sorted by flow cytometry to yield Tomato-positive and -negative that were subjected to RNA-seq. Raw read data from all samples were processed using a standard bioinformatic pipeline that has been described in detail previously to obtain values for mRNA levels that are expressed as reads per million mapped reads per kilobasepair.

RELATIVE QUANTIFICATION OF ECM DEPOSITION

Seven days after Suv420h2 siRNA transfection, MC3T3-E1 cells were removed using 0.5% sodium deoxycholate. The residual ECM was washed with PBS and stained with picro Sirius red as described before [Thaler et al., 2010a, 2013]. Fluorescence quantification was performed on a Tecan (Maennedorf, Switzerland) multi-well reader.

ALKALINE PHOSPHATASE ACTIVITY (ALP)

MC3T3-E1 cells were seeded in 12 multi well culture dishes at a density of 10,000/cm² and transfected with Suv420h2 specific or scrambled siRNA as described above. On the next day, the medium was changed and cell differentiation was induced. After 10 days, cell number and ALP-activity were analyzed. For determination of cell number (DNA amount employed as surrogate), cell layers were washed with PBS and frozen with 1 mM Tris–HCl buffer (pH 8.0) containing 0.1 mM EDTA. During thawing, Hoechst 33258 dye (Polysciences, Warrington, PA) was added (1 µg/ml) and, after an incubation of 15 min at room temperature, the fluorescence was measured (excitation 360, emission 465 nm). The amount of DNA was estimated using a standard curve prepared from calf thymus DNA (Roche). Thereafter, alkaline phosphatase (ALP) activity was measured with p-nitrophenylphosphate (Sigma) (2.5 mg/ml in 0.1 M diethanolamine buffer [pH 10.5], 150 mM NaCl, 2 mM MgCl₂) by incubation of the cell layers for 20 min at room temperature. Absorption was measured in a microplate reader at 405/490 nm. ALP activity per well was estimated using a standard curve prepared from calf intestinal ALP (Roche) and referred to the amount of DNA previously measured.

ALIZARIN RED STAINING

For assessment of mineralization, MC3T3-E1 cells were fixed in 4% paraformaldehyde at 29 days after siRNA transfections and 28 days after induction of osteoblastic differentiation. Fixed cells were stained with 40 mM Alizarin Red S (Sigma–Aldrich), pH 4.2, by incubating the cells for 20 min at room temperature. Subsequently, Alizarin Red S dye was extracted using 10% cetylpyridinium chloride in 10 mM sodium phosphate, pH 7.0, for 45 min at room temperature. Alizarin Red S absorbance was measured at 562 nm in a multiplate reader (Tecan) and normalized to total DNA amount which previously measured by Hoechst stain with a multiplate reader (Tecan).

STATISTICAL ANALYSIS

Statistical analyses were performed using ANOVA or Student's t-test in Prism 4.03 (GraphPad Software). Values of P ≤ 0.05 were considered significant. Each experiment consisted of at least three biological replicates. For RT-qPCR data, results from technical triplicates were averaged and the mean value was treated as a single, biological statistical unit. Results are presented as means ± SD.

TRANSCRIPTOME ANALYSIS BY NEXT GENERATION SEQUENCING REVEALS ELEVATED IN VIVO EXPRESSION OF SUV420H2 IN OSTEOBLASTS

We performed gene expression profiling using RNA-seq of mouse cells derived from fracture callus and calvaria. RNA-seq values were compared to assess which epigenetic regulators are most specific for mesenchymal cells and bone tissues. Heat-map analysis revealed that there are many epigenetic genes (n = 113 out of 345 total) that are selectively expressed at higher levels in mesenchymal cells or tissues with an osteo-chondrogenic biomarker signature (Fig. 1A and B). Among these genes, we noted that the six methyltransferases Ezh1, Ezh2, Suv39h1, Suv39h2, Suv420h1, and Suv420h2 are prominently expressed in callus cells and calvarial tissue (Fig. 1C).

Direct comparison of the six methyltransferases in our study reveals that Suv420h2 is expressed at the highest mRNA levels compared to Ezh1, Ezh2, Suv39h1, Suv39h2, and Suv420h1 (Fig. 2A) in calvarial bone from mice with a conditional Ezh2 null mutation that accelerates osteoblast differentiation and elevates expression of the bone markers Alpl and Ibsp (Fig. 2B). Furthermore, Suv420h2 represents the only gene of this set of six that is selectively enriched in mesenchymal cells in the fracture callus at day 6 and that are marked by RFP expression by lineage-tracing (Fig. 2C). These same RFP-positive cells also express very high levels of Alpl and Ibsp compared to RFP-negative cells in the callus (Fig. 2D). Thus, the RNA-seq data indicate that Suv420h2 is selectively expressed at higher levels in tissues and cells that robustly express osteoblastic markers. This finding suggests that H4K20 methyl transferase activity may be linked to progression of osteoblast differentiation.

SUV METHYLTRANSFERASES ARE DOWN-REGULATED DURING ONSET OF QUIESCENCE AND DIFFERENTIATION IN MC3T3-E1 OSTEOBLASTS AND MLO-A5 PRE-OSTEOCYTES

Having established that Suv420h2 is strongly expressed in osteogenic cells by RNA-seq, we validated its expression by RT-qPCR in cell culture models for osteoblast and early osteocyte differentiation (Figs. 3 and 4). As expected, the proliferation markers histone H4 (Hist2h4) and Cyclin B2 (Ccnb2) are down regulated after day 3 in both MC3T3-E1 and MLO-A5 cells (Figs. 3A and 4A). Similarly, Ezh2, Suv39h1, and Suv39h2 also have peak expression during the early culturing period in MC3T3-E1 cells (Fig. 3B and C). However, Ezh1, Suv420h1, and Suv420h2 appear to be expressed at maximal levels at day 7 and their expression gradually fades during progression of osteoblast differentiation over 28 days of MC3T3-E1 cell culture (Fig. 3B–D). For comparison, expression patterns in MLO-A5 osteocytes tend to be generally similar, albeit more variable and expression changes are less pronounced (Fig. 4B and D). We note that Suv420h2 is expressed at lower levels and shows less modulation in its expression compared to Suv420h1 in MLO-A5 osteocytes (Fig. 4D). Taken together, these data corroborate the in vivo expression data presented in Figure 1 and indicate that Suv420h2 is actively expressed during osteoblast maturation but is expressed at lower levels after the osteoblast-osteocyte transition.

SUV420H2 CONTROLS H4K20 TRI-METHYLATION IN PRE-OSTEOBLASTIC CELLS

H4K20 tri-methylation is a hallmark of transcriptional repression and is regulated by Suv420h1 and Suv420h2. Because Suv420h2 is robustly expressed in osteoblastic cells (see Figs. 1-4), we analyzed the effects of Suv420h2 knock down by siRNA in MC3T3-E1 pre-osteoblasts to understand its functional role in osteoblastogenesis. We observe a clear two- to threefold reduction in Suv420h2 expression at both RNA (Fig. 5A) and protein levels (Fig. 5B and C) at 3 days after siRNA transfection. More importantly, immunoblot analysis shows that depletion of Suv420h2 reduces H4K20me3 levels (Fig. 5D) but not H4K20me1 levels (Fig. 5E). These studies support the conclusion that Suv420h2 is rate-limiting for global H4K20me3 levels in MC3T3-E1 cells and thus a major regulator of H4K20me3-directed transcriptional repression during osteoblastogenesis.

PRIMARY OSTEOGENIC FUNCTIONS ARE DISRUPTED AFTER REDUCTION OF SUV420H2 EXPRESSION

Extra cellular matrix (ECM) deposition and its mineralization are among the main activities of differentiating osteoblasts. To elucidate the role of Suv420h2 on ECM maturation during osteoblastogenesis, we analyzed the effect of Suv420h2 depletion by siRNA on ECM deposition around MC3T3-E1 osteoblasts. Collagen fibers were visualized using Picro Sirius Red and staining intensity was determined at day 7 after initiation of osteoblastic differentiation. We observed a modest albeit not significant reduction in ECM deposition in MC3T3-E1 cells treated with Suv420h2 siRNAs (Fig. 6A). However, alkaline phosphatase activity, an enzyme marker that is deposited in the collagen-rich ECM during osteoblast maturation and ECM mineralization, was significantly decreased at day 10 in cells in which Suv420h2 levels were reduced by siRNA treatment (Fig. 6B). Furthermore, ECM mineralization was significantly reduced by about 25% in MC3T3-E1 cells at day 28 of differentiation (Fig. 6C), even though Suv420h2 expression was only actively suppressed by siRNA during the initial stages of the culture period. The lasting effects of a temporary knock down of Suv420h2 by siRNA indicates that Suv420h2 may affect the molecular memory of MC3T3-E1 cells through epigenetic mechanisms operative at early stages of osteoblastogenesis.

SIRNA INHIBITION OF Suv420H2 NEGATIVELY AFFECTS THE EXPRESSION OF HISTONE METHYLTRANSFERASES AND OF PRIMARY OSTEOBLASTIC MARKERS IN DIFFERENTIATING MC3T3-E1 CELLS

Because the long-term effects of Suv420h2 inhibition on osteoblastic differentiation of MC3T3-E1 cells are instigated during the first several days of the culturing period, we analyzed the acute regulatory functions of Suv420h2 by analyzing gene expression of multiple markers and select transcription factors genes in cells in which Suv420h2 expression was reduced by siRNA (see Fig. 5). Our analysis focused on day 3 of MC3T3-E1 differentiation (Fig. 7) as a key time-point within a broader time-course up to day 10 (Figs. 7-9). Depletion of Suv420h2 does have small (less than 1.5-fold), although partially significant effects on the expression of other methyltransferases (Fig. 7A). However, cells exhibit two- to eightfold a down-regulation of several ECM-related osteoblastic genes, including Sparc, Col1a1, Alpl, and Ibsp (Fig. 7B). We did not detect significant changes in the levels of Bglap2, presumably because cells at day 3 in culture are immature and do not have a robust basal level of expression. More importantly, we observed a strong reduction in the expression of the osteoblastic transcription factor Sp7 (Osx) at day 3 after Suv420h2 siRNA transfection, while Runx2 levels remain relatively constant (Fig. 7C). These results suggest that Suv420h2 alters key osteoblastic markers at early stages of osteoblastic differentiation in cell culture.

LONG-TERM EFFECTS OF TRANSIENT Suv420H2 INHIBITION ON THE EXPRESSION OF PRIMARY OSTEOBLASTIC MARKERS IN DIFFERENTIATING MC3T3-E1 CELLS

Transient depletion of Suv420h2 expression by siRNA for the first 3 days at early stages of osteoblast differentiation affects late osteoblastic events like ECM mineralization (see Fig. 6C). Therefore, we assessed whether these early reductions in Suv420h2 levels have sustained effects on osteoblastic maturation (Figs. 8 and 9). Interestingly, although MC3T3-E1 cells were treated with Suv420h2 siRNA for only 3 days, Suv420h2 expression remains low throughout the remainder of the time course until day 10 (Fig. 8F). Even though this reduced level of Suv420h2 has only modest effects on the expression of other epigenetic regulators at day 3 (Fig. 8A–F; see also Fig. 7A), greater changes in expression of Ezh2, Suv39h1, and Suv39h2 are observed at day 6 (Fig. 8B–D). Thus, loss of Suv420h2 expression may have broad ranging secondary epigenetic effects on gene expression during osteoblast differentiation (Fig. 8F). Furthermore, transient Suv420h2 knock down (Fig. 8F) has primarily transient effects on most osteoblastic markers like Col1a1, Ibsp, Alpl (Fig. 9B–D) or osteoblastic transcription factors like Sp7 (Fig. 9E and F). However, Suv420h2 siRNA treatment has a delayed effect on the late-stage osteoblast marker Bglap2 at day 10 (Fig. 9A). We conclude that transient knock down of Suv420h2 has delayed effects on ECM mineralization by altering lysine-methylation related epigenetic mechanisms, and these alterations reduce expression of early osteoblastic markers and delay expression of late osteoblast markers that together control ECM maturation.