Loss of ZC4H2, an Arthrogryposis Multiplex Congenita Associated Gene, Promotes Osteoclastogenesis in Mice

Abstract

ZC4H2 encodes a C4H2-type zinc finger protein, mutations of which lead to a spectrum of diseases known as ZC4H2 associated rare disorders (ZARD). In addition to neurological phenotypes, the most typical symptoms of ZARD are multiple joint contractures of varying degrees, accompanied by abnormal development of muscles and bones, and osteoporosis in some cases. The pathogenic mechanisms of such bone related phenotypes, however, remain unclear. Here, we showed that ZC4H2 is expressed in the developing bones in mice. ZC4H2 knockout mice were neonatal-lethal and smaller in size, with reduced calcification of long bones. Upon induced loss of ZC4H2 postnatally, the femoral bones developed an osteoporosis-like phenotype, with reduced bone mineral density, bone-volume fraction, and trabecular bone number. Knockdown of ZC4H2 showed no clear effect on the expression of osteogenic differentiation genes in in vitro models using mesenchymal stem cells. Interestingly, ZC4H2 knockdown significantly enhanced osteoclast differentiation and bone resorption in induced bone marrow-derived macrophages. We further confirmed that the number of osteoclasts in the long bone of ZC4H2 knockout mice was increased, as well as the expression of the serum bone resorption/osteoporosis marker CTX-1. Our study unveils a new role of ZC4H2 in osteoclast differentiation and bone development, providing new clues on the pathology of ZARD.

1. Introduction

Zinc finger C4H2-type containing (ZC4H2) is a zinc finger protein, and the ZC4H2 gene is located on the X chromosome in both mice and humans. In humans, ZC4H2 has been identified as a causal gene of Wieacker–Wolff syndrome. Affected patients with ZC4H2 mutations show various clinical phenotypes, including arthrogryposis multiplex congenita (AMC), intellectual disability, and epilepsy, which are now collectively referred to as ZC4H2-associated rare disorders (ZARD) [1,2,3,4,5,6]. ZC4H2 deficiency leads to severe neurodevelopmental impairments in humans and has been shown to be widely involved in neural development in mouse and zebrafish models [2,6,7,8,9,10]. Mechanistically, ZC4H2 has been shown mostly to work as a stabilizing co-factor of the ubiquitin E3 ligase RNF220 [9,11,12]. RNF220 and ZC4H2 cooperate to pattern the embryonic ventral neural tube through regulation of Shh/Gli signaling as well as the stability of key transcription factors (Dbx1/2 and Nkx2.2) [11]. They are both required for noradrenergic neuron development in the locus coeruleus (LC), where ZC4H2 works as an adaptor to bridge RNF220 to its targets, Phox2a/b, for mono-ubiquitination [7]. In addition, ZC4H2/RNF220 are also involved in cerebellum development and Shh-group medulloblastoma progression [13].

In addition to the neurological phenotypes, the most typical symptoms of ZARD are multiple joint contractures of varying degrees, accompanied by the abnormal development of muscles and bones, and osteoporosis in some cases [2,4,5,14,15]. Whether ZC4H2 is directly involved in bone development, however, remains unclear. Bone development and homeostasis are regulated by the balance of osteoblast-mediated bone formation and osteoclast-mediated bone resorption [16,17,18,19]. In this study, ZC4H2 mutant mice were used to study the roles of ZC4H2 in bone development. In vivo and in vitro experiments were conducted to investigate the effects of ZC4H2 deletion on the differentiation and proliferation of osteoblast and osteoclast lineage cells.

2. Materials and Methods

2.1. Animal Experimentation

All animal procedures conducted in this study were approved by the Institutional Animal Care and Use Committee (IACUC) of Kunming Institute of Zoology, Chinese Academy of Sciences. Mice containing conditional alleles of ZC4H2 were previously described [11]. They were mated to Vasa-Cre or Rosa26-CreERT2 transgenic mice. Vasa-Cre, which is expressed in the oocyte, would induce complete loss of ZC4H2 in the ZC4H2^fl/Y embryos. ZC4H2^fl/Y; Vasa-Cre-positive mice are referred to as ZC4H2 knockout (KO); ZC4H2^+/Y; Vasa-Cre-positive mice were used as controls (WT). ZC4H2^fl/Y; Rosa26-CreERT2-positive mice induced with tamoxifen are referred to as ZC4H2 TAM-knockout (TAM-KO); ZC4H2^fl/Y; Rosa26-CreERT2-positive mice without tamoxifen injection were used as controls (WT). For tamoxifen-inducible knockout mice, mice at one-week old were administered an intraperitoneal injection of tamoxifen (75 mg/kg) once daily for five consecutive days. Note that as ZC4H2 is located on the X chromosome, in this study, male ZC4H2^fl/Y and female ZC4H2^fl/+; Vasa-Cre mice were used as crossing pairs, and only male ZC4H2^fl/Y; Vasa-Cre mice were used for following phenotypic and cellular analysis, to keep consistent with the ZC4H2^+/Y littermate controls. The presence of ZC4H2 floxed and wild-type alleles were detected by PCR using genomic DNA prepared from distal tails. Sequences of the primers used for genotyping are referenced from previously reported [11].

2.2. Quantitative Real-Time Polymerase Chain Reaction (qRT-PCR)

Total RNA was extracted from cells by Trizol reagent (TianGen, Beijing, China) and was reverse-transcribed into complementary DNA (cDNA) using a PrimeScript RT reagent kit (TaKaRa, Tokyo, Japan) according to the manufacturer’s instructions. qRT-PCR was performed using SYBR Green PCR Master Mix (TaKaRa, Tokyo, Japan) and on the CFX96Touch Real Time PCR system (Bio-Rad, Hercules, CA, USA). Primer sequences used are given in

Table 1. Primer sequences used for qRT-PCR.

Gene	Forward Sequence	Reverse Sequence
ACTB	GGCTGTATTCCCCTCCATCG	CCAGTTGGTAACAATGCCATGT
ZC4H2	AAAGATCAAGGCCCGTTTG	TTGTATTCTTTCAGGTGCCTCTC
OSX	ACTCATCCCTATGGCTCGTG	GGTAGGGAGCTGGGTTAAGG
OCN	ACTCCGGCGCTACCTTGGAGCC	GCAGGGTTAAGCTCACACTG
Runx2	CCTAGTTAGAGTGGTAGCAGA	ACAGACAACGAAGAAAGTTCC
Col1a1	GCTCCTCTTAGGGGCCACT	CCACGTCTCACCATTGGGG
Ctsk	CTCGGCGTTTAATTTGGGAGA	TCGAGAGGGAGGTATTCTGAGT
Mmp9	CTGGACAGCCAGACACTAAAG	CTCGCGGCAAGTCTTCAGAG
Acp5	CACTCCCACCCTGAGATTTGT	CATCGTCTGCACGGTTCTG
Dcstamp	GGGGACTTATGTGTTTCCACG	ACAAAGCAACAGACTCCCAAAT
Ocstamp	CTGTAACGAACTACTGACCCAG	CCCAGGCTTAGGAAGACGAAG
Atp6v0d2	CAGAGCTGTACTTCAATGTGGA	AGGTCTCACACTGCACTAGGT
Oscar	CCTAGCCTCATACCCCCAG	CGTTGATCCCAGGAGTCACAA

.

2.3. MSCs, BMDM Isolation and In Vitro Differentiation

Mesenchymal stem cells (MSCs) and bone marrow derived macrophages (BMDM) were isolated from 6–8 week-old ZC4H2^fl/Y; Rosa26-CreERT2 mice as previously reported [20]. 4-OH-tamoxifen (4-OHT, 1.5 μM) was used to induce ZC4H2 knockout in MSCs and BMDM, and ethanol was used as a control. The MSCs and BMDM with or without 4-OHT treatment for 24 h were cultured and induced to osteoblast and osteoclast differentiation, respectively, as previously reported [21,22]. In brief, for in vitro osteoblast differentiation, 2 × 10⁵ MSCs were seeded in 12-well plates and cultured in osteogenic medium with α-MEM medium containing 10% FBS, 1% Penicillin/Streptomycin, 10 mM β-glycerol phosphate, and 50 μg/mL ascorbic acid, and the medium was changed every 2 days. For in vitro osteoclast differentiation, BMDM were seeded in 24-well plates at a density of 1 × 10⁵ cells per well and cultured in osteoclastogenic medium with α-MEM containing 30 ng/mL M-CSF and 100 ng/mL RANKL (PeproTech, Cranbury, NJ, USA); the medium was changed every 2 days.

2.4. RNA-Seq Analysis

For RNA-Seq analysis of the differentiated BMDM, cells were collected one week after osteoclast differentiation and RNAs were extracted for transcriptome sequencing. The sequencing library was constructed and sequenced by BGI-Genomics on the Illumina platform. First, the sequencing data were filtered with SOAPnuke (v1.5.2) [23]. Subsequently, the clean reads were mapped to the reference genome GRCm38 (https://www.ncbi.nlm.nih.gov/grc/mouse (accessed on 15 August 2023)) using HISAT2 (v2.0.4) [24]. Next, Bowtie2 (v2.2.5) [25] was used to align the clean reads to the reference coding gene set, followed by gene expression level calculation using RSEM (v1.2.12) [26]. Differential expression analysis was performed using DESeq2 (v1.4.5) [27] with a Q value ≤ 0.05 and an absolute fold change ≥ 1.5. Kyoto Encyclopedia of Genes and Genomes (KEGG, https://www.kegg.jp/ (accessed on 15 August 2023)) enrichment analysis of annotated differentially expressed genes was conducted using Phyper (https://en.wikipedia.org/wiki/Hypergeometric_distribution (accessed on 15 August 2023)) based on the Hypergeometric test. The significance levels of terms and pathways were corrected by Q value using a rigorous threshold (Q value ≤ 0.05) according to Bonferroni [28,29].

2.5. Von Kossa Staining

Femur sections were deparaffinized and rehydrated, stained with von Kossa silver solution under strong light for 15–60 min. The sections were then washed, fixed with sodium thiosulfate solution for 2 min, counterstained with hematoxylin and eosin (HE), mounted, and scanned using a digital slide scanning system (Olympus Corporation, Tokyo, Japan).

2.6. Skeletal Staining

Alcian blue and Alizarin red S (ARS) staining of mouse skeletons was performed according to previous reports [21]. The samples were fixed in 95% ethanol for 24 h, stained with Alcian blue staining solution for 12–18 h, and washed 5 times with 95% ethanol. For ARS staining, the samples are treated with 2% KOH for 3–4 h, then stained with ARS staining solution for 3–4 h. Subsequently, they were washed 5 times with 1% KOH/20% glycerol solution. Images are quantitatively analyzed using Nikon software (NIS-Elements D (v5.10.00), Nikon, Tokyo, Japan).

2.7. Micro-Computed Tomography (Micro-CT)

Mouse femurs of control and TAM-ZC4H2 KO groups at 3 months old were collected and scanned using a SkyScan1276 (Bruker, Kartuizersweg, Belgium) micro-CT at 8 μm resolution for quantitative analysis. The turn on voltage was 85 kV and current density was 150 μA. Analyses of bone mineral density (BMD), bone volume-to-total volume ratio (BV/TV), trabecular number (Tb.N), and trabecular separation (Tb.Sp) are conducted in the region 0–100 layers below the growth plate. Analyses of cortical thickness (Ct.Th) and cortical bone mineral density (Ct.BMD) are conducted in the region 300–400 layers below the growth plate. Volumetric reconstructions and analyses were performed using built-in software CTvox (v3.3.0r1383) and CTAn (v1.17.7.2), respectively.

2.8. Alkaline Phosphatase (ALP) and Alizarin Red S (ARS) Staining

For analysis of the osteoblast differentiated MSCs, after 8 days of induction, ALP staining and quantification were performed using an ALP staining kit (Solarbio, Beijing, China) and an ALP quantification kit (Beyotime, Shanghai, China), respectively. After about three weeks of induction, ARS staining was performed using an ARS staining kit (Solarbio, Beijing, China) and quantified using 10% cetylpyridinium chloride (CPC). The quantified ALP and ARS activities were normalized to protein content.

2.9. Tartrate Resistant Acid Phosphatase (TRAP) Staining Assay

For analysis of the osteoclast differentiated BMDM, after 7 days of induction, TRAP staining was performed using a TRAP staining kit (WAKO, Osaka, Japan). Osteoclasts with more than three nuclei were defined as mature osteoclasts, and the number of mature osteoclasts was analyzed using ImageJ software (v1.54i) [30].

TRAP staining of the femur sections of control and ZC4H2 KO mice were carried out according to the instructions provided with the TRAP staining kit (WAKO, Osaka, Japan). The femoral sections were deparaffinized, rehydrated and stained with a freshly prepared TRAP staining solution at room temperature for 30 min. The sections were then counterstained with methyl green, mounted, and scanned using a digital slide scanning system (Olympus Corporation, Tokyo, Japan). The number of TRAP-positive osteoclasts was analyzed using ImageJ software [30].

2.10. F-Actin Ring Formation Assay

BMDM were induced to differentiate into osteoclasts for 7 days, followed by cell fixation, permeabilization, and blocking. The cells were then incubated overnight at 4 °C with Alexa Fluor™ 647-phalloidin (Thermo Fisher Scientific, Waltham, MA, USA). After incubation, the cells were washed 4 times with phosphate-buffered saline (PBS). Images were captured using an IX73 fluorescence microscope (Olympus Corporation, Tokyo, Japan), and the number of F-actin rings was quantified using ImageJ software.

2.11. Scanning Electron Microscopy

BMDM cells were seeded on bovine bone slices and induced to differentiate into osteoclasts for approximately one week. After removing the culture medium, pure water was added and left at room temperature for 10 min to swell and detach the cells from the bone surface. The bone slices were then gently wiped with absorbent cotton, followed by ultrasonic treatment for 4 min. The slices were dehydrated with a gradient series of alcohols to absolute ethanol and air-dried at room temperature. The samples were attached to the sample holder using conductive adhesive and thoroughly dried under vacuum. Gold sputtering was performed for approximately 45 s, and the samples were scanned and imaged using a scanning electron microscope (Sigma 500, Zeiss, Oberkochen, Germany) with an accelerating voltage of 10 kV.

2.12. ELISA Assay of CTX-1

The detection of CTX-1 content in mouse serum was performed using an ELISA kit (CUSABIO, Wuhan, China) according to the product instructions. The optical density (OD) was measured at 450 nm using a microplate reader. The results were analyzed using CurveExpert 1.4 (Hyams Development, Chattanooga, TN, USA).

2.13. Cell Counting Kit-8 (CCK-8) Assay

Cells were seeded at a density of 1000 cells/well in a 96-well plate, with 100 μL of complete medium added to each well. Then, 10 μL of CCK-8 was added to each well at 0, 2, 4, and 6 days, mixed, and incubated for 2 h. The absorbance at 450 nm was measured using a microplate reader. The OD values of blank wells (without cells, only medium and CCK-8) were used for calibration, followed by statistical analysis.

2.14. BrdU Incorporation Assay

Cells were seeded in a 24-well plate and cultured for 1–2 days. Then, BrdU (10 μM, Sigma, B5002) was added to the medium and incubated for 45 min. The medium was removed, and the cells were washed three times with PBS. Cells were then permeabilized with 0.3% Triton X-100, followed by the addition of 2N HCl and incubation at room temperature for 30 min. The cells were washed 4 times with PBS, then blocked with 10% goat serum for 1 h. Subsequently, cells were incubated with BrdU primary antibody (Sigma, B8434-25UL) and corresponding secondary antibody. After thorough washing, cells were stained with DAPI for nuclear staining, and finally observed and photographed using an IX73 fluorescence microscope (Olympus Corporation, Tokyo, Japan).

2.15. Statistical Analysis

All experiments were repeated at least three times, and the data were analyzed using GraphPad Prism 7.0 (GraphPad Software, San Diego, CA, USA) and presented as means ± standard deviation (SD). Unpaired two-tailed Student’s t-test or two-way analysis of variance (ANOVA) was used to determine the statistical significance. p < 0.05 was identified as statistically significant. n.s. represents no significant difference (p > 0.05).

3. Results

3.1. Loss of ZC4H2 Reduces Long Bone Calcification in Mice

We first analyzed the expression of ZC4H2 in different tissues of newborn mice (P0) using qRT-PCR. The results showed that ZC4H2 is widely expressed in many tissues, strongest in the brain, and moderate in the skeletal muscle and bone (both cranium and femur, Figure 1A), which were lost in the ZC4H2 knockout (KO) mice. As previously reported [11], the ZC4H2 knockout mice were neonatal lethal (Figure 1B). We noticed that the neonatal ZC4H2 embryos were slightly smaller, with lower body weight compared with wild-type littermates (Figure 1C). Histological analysis revealed no clear alterations in the morphology of the long bone of ZC4H2-deficient mice. We analyzed the calcification status of the long bones of the limbs in the newborn ZC4H2 knockout and control embryos by Alcian Blue (which stains the cartilage) and Alizarin Red S (which stains the calcified bone) staining. Although the total length of the long bones (humerus, radius, ulna, femur, and tibia) of the ZC4H2 knockout embryos were similar to those of the wild-type ones, the length of their calcified parts was generally shorter than those in the WT (Figure 1D,E).

3.2. Loss of ZC4H2 Reduces Femoral Bone Mass

We further analyzed the effect of loss of ZC4H2 on calcium deposition in the femur using von Kossa staining. The results revealed lower bone mass in the ZC4H2 KO femur than in the control (Figure 2A). To study the potential effect of ZC4H2 in postnatal bone development, we crossed the ZC4H2^fl/Y mice with a tamoxifen inducible Rosa26-CreERT2 line, and the ZC4H2^fl/Y; Rosa26-CreERT2 mice were injected with tamoxifen at 1 week old to induce ZC4H2 deletion. The bone mass of the femur of the TAM-ZC4H2 KO and control mice were analyzed by micro-CT at 3 months old. Compared to the control group, the femoral bone in the KO group exhibited an osteoporosis-like phenotype, with reduced BMD, BV/TV, and Tb.N, increased Tb.Sp, and a slight reduction in Ct.Th and Ct.BMD (Figure 2B,C).

3.3. Loss of ZC4H2 Does Not Affect Osteoblast Differentiation

The above data support that loss of ZC4H2 leads to reduced bone mass, which could be due to inhibited osteogenesis or enhanced bone resorption. We therefore tested the effects of loss of ZC4H2 on the differentiation of osteoblasts and osteoclasts using in vitro models.

Bone marrow mesenchymal stem cells (MSCs) were isolated from ZC4H2^fl/Y; Rosa26-CreERT2 mice at 6–8 weeks of age and induced into osteoblasts with or without tamoxifen treatment to induce deletion of ZC4H2. Addition of 4-OH-tamoxifen efficiently reduced the expression of ZC4H2 in MSCs (Figure 3A). Alkaline phosphatase (ALP) and Alizarin red S (ARS) staining were conducted to analyze the osteogenic differentiation states. The results showed that loss of ZC4H2 had no significant effect on the osteogenic differentiation of MSCs (Figure 3B–E). In the histochemical assays, the staining of ALP and ARS may seem more reduced in the KO group than in the control cells (Figure 3B,D). However, this is largely due to the reduced cell numbers because of the inhibitory effect of ZC4H2 knockout on MSCs proliferation (see below). Indeed, in the quantified assays normalized to total protein levels (Figure 3C,E), no clear difference was observed between these two groups. We examined the expression of a group of osteogenic differentiation associated genes (OCN, OSX, Runx2, and Col1a1) by qRT-PCR. No significant change in their expression was observed in the ZC4H2 KO group (Figure 3F). These results suggest that loss of ZC4H2 does not affect osteogenesis in general.

3.4. Loss of ZC4H2 Promotes Osteoclastogenesis and Bone Resorption

To examine the effects of ZC4H2 knockout on osteoclast differentiation, BMDM were isolated from ZC4H2^fl/Y; Rosa26-CreERT2 mice, and osteoclast differentiation assays were carried out in the presence or absence of 4-OH-tamoxifen. Addition of 4-OH-tamoxifen efficiently reduced the expression of ZC4H2 in BMDM. Tartrate resistant acid phosphatase (TRAP) staining was used to visualize the multi-nucleated osteoclasts. The number of osteoclasts increased clearly in the ZC4H2 KO group compared with the control cells (Figure 4A). This is further supported by the increase in the F-actin rings which represent active osteoclasts (Figure 4B). An RNA-Seq analysis was carried out on the control and ZC4H2 KO BMDM upon osteoclast differentiation. Annotation of the differentially expressed genes showed clear enrichment in the osteoclast differentiation pathway as well as the PI3K-Akt signaling pathway (Figure 4C). In qRT-PCR analysis, the expression of the osteoclast differentiation associated genes (Ctsk, Oscar, Atp6v0d2, Acp5, Ocstamp, Dcstamp, and Mmp9) increased collectively, especially that of Dcstamp and Mmp9 (Figure 4D). In bone resorption assays on bovine bone slices, the osteoclast group from ZC4H2 KO BMDM produced more resorption pits than would be derived from control BMDM (Figure 4E), supporting increased bone resorption activity of the ZC4H2 KO group.

We tested whether this is also the case in vivo. TRAP staining of the P0 femur slice revealed more TRAP⁺ osteoclasts in the ZC4H2 KO mice than in controls (Figure 4F). We also checked the serum levels of CTX-1 (a marker of osteoclastic activity) in P0 ZC4H2 KO mice. The results showed clearly increased serum CTX-1 levels in the ZC4H2 KO mice (Figure 4G), suggesting increased osteoclastic activity. These data support that loss of ZC4H2 promotes osteoclastogenesis and bone resorption in vitro and in vivo.

3.5. Loss of ZC4H2 Inhibits Proliferation of MSCs and BMDM

Using MSCs and BMDM derived from the ZC4H2^fl/Y; Rosa26-CreERT2 mice, we tested the effects of 4-OH-tamoxifen induced deletion of ZC4H2 on their proliferation using Cell Counting Kit-8 (CCK-8) and BrdU incorporation assays. CCK-8 assays showed significantly reduced cell proliferation of both the MSCs and BMDM on day 6 when ZC4H2 was knocked down (Figure 5A,C), which was consistent with the lower BrdU incorporation rate in both cells (Figure 5B,D).

4. Discussion

In this report, we analyzed the potential roles of ZC4H2 in bone development in mice. We showed that ZC4H2 is expressed in different bone tissues, and loss of ZC4H2 reduces bone calcification and leads to an osteoporosis-like phenotype. In an in vitro assay, knockdown ZC4H2 had showed no clear effect on the osteogenic differentiation of mesenchymal stem cells, but the proliferation of the cells was inhibited. We cannot rule out the contribution of ZC4H2 on osteogenesis in vivo through its effects on cell proliferation. Interestingly, we showed that ZC4H2 knockout enhances osteoclast differentiation in mice. We utilized mouse bone marrow-derived macrophages to establish in vitro models for osteoclast differentiation. Morphological and transcriptomic analysis showed that ZC4H2 knockdown significantly enhanced osteoclast differentiation and the expression of related genes, leading to increased bone resorption. We further confirmed that the number of osteoclasts in the long bone of ZC4H2 knockout mice was increased, as well as the expression of the serum bone resorption/osteoporosis marker CTX-1. Thus, our work unveils a specific role of ZC4H2 in regulating osteoclast differentiation and bone resorption activity, providing new clues for the pathology for ZARD. It would be of great interest to test whether bone resorption activities might be increased in ZARD patients, which could potentially contribute to the development of joint contracture, although this is not obvious in the ZC4H2 KO mice.

How ZC4H2 is involved in osteoclast differentiation remains to be elucidated. The first question is whether ZC4H2 regulates osteoclastogenesis through RNF220. Osteoclasts mainly originate from bone marrow monocyte–macrophage cell lines. We previously showed that RNF220 is expressed in bone marrow macrophages, which is responsive to pathogenic infection and IFN signaling. RNF220 mediated STAT1 ubiquitination contributed significantly to STAT1 activation and innate immune responses [31]. STAT1 signaling has been shown to be a negative regulator of osteoclast differentiation [32]. RNF220 is also involved in Shh, BMP, and Wnt signaling in different contexts [10,11,12], which are all involved in osteoclastogenesis. It is important to test whether osteoclast differentiation is also affected in RNF220 knockout mice. We cannot rule out the possibility that ZC4H2 might work independently of RNF220 in osteoclastogenesis, which remains to be investigated in the future.

In summary, this study revealed that ZC4H2 regulates bone resorption activity by inhibiting osteoclast differentiation in mice, providing new clues regarding the pathology of bone related phenotypes of ZARD.