Contact Co-culture of Osteoblasts and Sympathetic Neuronal Cells Promotes Differentiation of Both Cell Types

Abstract

Background/Aim: Bone and nerve reconstruction is crucial for treating various diseases of the oral and maxillofacial region. However, the relationship between bone and nervous system has not yet been fully elucidated. Therefore, we aimed to examine the interaction between osteoblasts and neuronal cells in contact co-culture.

Material and Methods: Osteoblasts and sympathetic neuronal cells were grown in contact co-culture. Microscopic observation, a mineralization assay, immunofluorescence staining, and DNA microarray analysis were performed.

Results: Microscopic observation revealed morphological changes in the osteoblasts that were co-cultured with sympathetic neuronal cells. Contact co-culture enhanced osteoblast calcification and upregulated a neuronal marker. Not only osteoblast differentiation signals, but also neuronal signals were increased in murine osteoblasts that were co-cultured with rat sympathetic neuronal cells. We also found that not only rat neuron differentiation signals, but also osteoblast differentiation signals were increased in rat sympathetic neuronal cells that were co-cultured with murine osteoblasts.

Conclusion: In the contact co-culture with osteoblasts and sympathetic neuronal cells, the sympathetic neuronal cells promoted osteoblast differentiation, and the osteoblasts promoted neuron differentiation.

Reconstruction of bone defects in the oral and maxillofacial region that are caused by trauma, tumor resection, infection, or genetic disorders is an important part of the medical profession. Autologous bone grafts collected from the ramus, ilium, or several other skeletal sites are considered to be the first strategy to regenerate bone; however, the availability of autologous bone is limited, and its collection is a highly invasive process. Therefore, osteoinductive factors are being investigated as alternatives to improve bone repair with minimal use of autologous grafts (1).

Bone homeostasis is maintained by the balance between osteoblastic bone formation and osteoclastic bone absorption. Recent studies revealed the interaction between the nervous system and bone metabolism (2,3). The central nervous system controls bone through several osteoblast receptors (4). The efferent sympathetic pathway regulates osteoblasts via noradrenaline, which binds to the β-2-adrenergic receptor, and neuropeptide Y, which binds to the Y1 receptor (5,6). On the other hand, sensory nerves regulate osteoblasts via semaphorin 3A, which binds to the plexin A4 receptor (7).

Electron microscopy has revealed the presence of nerve axons close to osteoblasts in bone tissue (8,9). Additionally, sympathetic nerve fibers innervate the periosteum, and their interaction with osteoblasts affects the metabolism of the periosteum (10). Contact co-culture of osteoblasts and neuronal cells has been used in several studies (11-13). However, the comprehensive gene expression patterns of contact co-cultured osteoblasts and sympathetic neuronal cells have not yet been examined. Herein, we performed a DNA microarray analysis to investigate the relationship between osteoblasts and sympathetic neuronal cells in contact co-culture.

Materials and Methods

Cell cultures. The mouse pre-osteoblastic cell line MC3T3-E1 was maintained in α-modified minimum essential medium (αMEM; Wako, Osaka, Japan) supplemented with 1% penicillin/streptomycin and 10% fetal bovine serum (FBS) (14). The sympathetic neuronal cell line PC12, derived from rat pheochromocytoma, was maintained in high glucose Dulbecco’s modified eagle medium (DMEM; Wako) supplemented with 10% horse serum, 10% FBS, and 1% penicillin/streptomycin (15). The medium was changed every 3 days. All cultures were maintained at 37˚C in humidified air containing 5% CO₂. For the contact co-culture of MC3T3-E1 cells and PC12 cells, DMEM was used.

Microscopic observation. Cells were observed, and images were taken using an inverted microscope CKX53 (Olympus, Tokyo, Japan).

Mineralization assay. For the contact co-culture, PC12 and MC3T3-E1 cells were combined at 6×10⁴ cells each in 12-well plates. Single cultures were seeded at 6×10⁴ cells in 12-well plates. The mineralization assay was performed as previously described (16). After reaching confluence, cells were incubated with ascorbic acid (50 μg/ml) and β-glycerophosphate (10 mM) for 21 days. Calcium mineralization of the cells was determined by alizarin red staining after the cells were fixed with ice-cold 70% ethanol. Each reaction was performed in triplicate on three individual samples.

Immunofluorescence staining. Cells were seeded in Falcon™ 8-well chambered cell culture slides (Thermo Fisher Scientific, Wilmington, DE, USA) for 48 h. Then, they were fixed with ice-cold 4% paraformaldehyde for 15 min, permeabilized with 1.0% Triton X-100 in phosphate-buffered saline for 30 min at 37˚C, blocked with 10% normal rabbit immunoglobulin (Ig) G and 10% normal mouse IgG in tris-buffered saline for 1 h, and incubated with anti-connexin 43 rabbit polyclonal antibody (#3512, Cell Signaling, Danvers, MA, USA) and anti-neurofilament-L (DA2) mouse monoclonal antibody (#2835, Cell Signaling). Cells were then incubated with both Alexa Fluor 488 goat anti-mouse IgG and Alexa Fluor 568 goat anti-rabbit IgG diluted at 1/200. The cell nuclei were visualized using Vectashield Mounting Medium with DAPI (H-1500, Vector Laboratories, Burlingame, CA, USA). Each reaction was performed in triplicate on three individual samples.

DNA microarray analysis. The microarray analysis was performed using total RNA samples from the MC3T3-E1 cells alone, PC12 cells alone, and the cells in the contact co-culture that was incubated for 72 h. The total RNA was extracted from cells using the RNeasy mini kit (Qiagen, Düsseldorf, Germany), and DNA microarray analyses were performed using the 3D-Gene Mouse Oligo chip 24k and Rat Oligo chip 20k (Toray Industries Inc., Tokyo, Japan) according to the manufacturers’ instructions. For efficient hybridization, these microarrays have a columnar structure to stabilize the spot morphology and enable microbead agitation. The total RNA was labeled with Cy5 using the Amino Allyl MessageAMP II aRNA Amplification Kit (Thermo Fisher Scientific). The Cy5-labeled aRNA pools were mixed with hybridization buffer and hybridized for 16 h according to the manufacturer’s protocols (www.3d-gene.com). The hybridization signals were obtained using a 3D-Gene Scanner (Toray Industries Inc.) and were processed by the 3D-Gene Extraction software (Toray Industries Inc.). Detected signals for each gene were normalized by a global normalization method (the median of the detected signal intensity was adjusted to 25). Gene ontology (GO) analysis using GeneCodis was performed for the significant genes. Each reaction was performed in triplicate on three individual samples.

Results

Contact co-culture with sympathetic neuronal cells enhanced the calcification of osteoblasts. We first observed the cells for morphological changes by inverted microscopy. The shape of MC3T3-E1 cells had become thinner (Figure 1A, B and C). We then examined whether the calcification of osteoblasts was promoted by the presence of the sympathetic neuronal cells in the contact co-culture. The calcification of osteoblasts was enhanced by the contact co-culture with sympathetic neuronal cells (Figure 2), suggesting that neuronal cells enhance osteoblast calcification under contact incubation conditions.

Figure 1. Cellular changes in the contact co-culture with PC12 and MC3T3-E1 cells. Microscopic observation by inverted microscopy at an actual magnification of 100×. (A) MC3T3-E1 cells alone. (B) PC12 cells alone. (C) Contact co-culture of MC3T3-E1 and PC12 cells incubated for 72 h.

Figure 2. Contact with sympathetic neuronal cells promotes calcification in osteoblasts. Effect of contact co-culture with PC12 cells on the calcium deposits of osteoblasts at 21 days, as determined by Alizarin red staining.

DNA microarray analysis of co-cultured cells by using Mouse Oligo Chip. To compare the gene expression patterns between the murine osteoblasts alone and the murine osteoblasts that were co-cultured with rat sympathetic neuronal cells, DNA microarray analysis of each sample was performed using the Mouse Oligo chip. As a result of GO enrichment analysis (biological process category) with the condition that the hypergeometric p-value is more than 0.05, we found that the following terms related to osteoblast differentiation were enriched: “Notch signaling pathway” (–1 × log10(p) =2.23), “positive regulation of odontogenesis” (–1 × log10(p)=1.84), “cartilage development” (–1 × log10(p)=1.50), “Wnt receptor signaling pathway” (–1 × log10(p)=1.45), “canonical Wnt receptor signaling pathway” (–1 × log10(p)=1.41), “regulation of osteoblast differentiation” (–1 × log10(p)=1.34), and “skeletal system development” (–1 × log10(p)=1.32) (Figure 3). Surprisingly, we found that the following terms related to neuron differentiation were also enriched: “nervous system development” (–1 × log10(p)=6.62), “neuron differentiation” (–1 × log10(p)=4.31), “neurotransmitter secretion” (–1 × log10(p)=2.79), and “axon guidance” (–1 × log10(p)=1.44).

Figure 3. Both osteoblastic and neuronal gene expression levels were up-regulated in the contact co-culture in the microarray analysis using the Mouse Oligo chip. GO enrichment analysis with the condition that the hypergeometric p-value is more than 0.05.

We filtered for genes with a Cy3/Cy5 ratio more than 2.0-fold in the biological process category by the search term “bone mineralization, osteoblast differentiation, or ossification”. As shown in Table I, we found nine genes, i.e., tuftelin 1 (Tuft1), osteopetrosis-associated transmembrane protein 1 (Ostm1), bone gamma-carboxyglutamate protein, related sequence 1 (Bglap-rs1), bone morphogenetic protein 5 (Bmp5), bone morphogenetic protein 8a (Bmp8a), GLI-Kruppel family member GLI2 (Gli2), dentin sialophosphoprotein (Dspp), parathyroid hormone receptor 1 (Pthr1), and trans-acting transcription factor 7 (Sp7). We also filtered for genes with a Cy3/Cy5 ratio more than 2.0-fold in the biological process category by the search term “neuron or differentiation”. As shown in Table II, we found seven genes, i.e., brain expressed gene 1 (Bex1), homeo box C8 (Hoxc8), neurofilament light polypeptide (Nefl), BarH-like 2 (Barhl2), achaete-scute complex homolog-like 1 (Ascl1), GATA binding protein 2 (Gata2), and phosphatidylinositol glycan anchor biosynthesis class T (Pigt). These results suggested that not only osteoblast differentiation signals, but also neuronal signals are increased in murine osteoblasts that are co-cultured with rat sympathetic neuronal cells.

Table I. Osteoblast differentiation-related genes expressed by osteoblasts using the Mouse Oligo chip.

Tuft1, Tuftelin 1; Ostm1, osteopetrosis associated transmembrane protein 1; Bglap-rs1, bone gamma-carboxyglutamate protein, related sequence 1; Bmp5, bone morphogenetic protein 5; Bmp8a, bone morphogenetic protein 8a; Gli2, GLI-Kruppel family member GLI2; Dspp, dentin sialophosphoprotein; Pthr1, parathyroid hormone receptor 1; Sp7, trans-acting transcription factor 7.

Table II. Neuronal differentiation-related genes expressed by osteoblasts using Mouse Oligo chip.

Bex1, Brain expressed gene 1; Hoxc8, homeo box C8; Nefl, neurofilament, light polypeptide; Barhl2, BarH-like 2; Ascl1, achaetescute complex homolog-like 1; Gata2, GATA binding protein 2; Pigt, phosphatidylinositol glycan anchor biosynthesis class T.

Contact co-culture with osteoblasts increased the levels of a neuronal marker in sympathetic neuronal cells. We next examined the effect of osteoblast differentiation on sympathetic neuronal cells in the contact co-culture. We found that the level of a marker of neuron differentiation, neurofilament L, was increased by the contact co-culture with osteoblasts (Figure 4), suggesting that osteoblasts promote neuron differentiation under contact incubation conditions.

Figure 4. Contact with osteoblasts increased the level of a neuronal marker in sympathetic neuronal cells. Effect of contact co-culture with MC3T3-E1 cells on PC12 cells incubated for 72 h, as determined by immunofluorescence staining.

DNA microarray analysis of co-cultured cells by using Rat Oligo Chip. To compare the gene expression patterns between the rat sympathetic neuronal cells alone and the rat sympathetic neuronal cells that were co-cultured with murine osteoblasts, DNA microarray analysis of each sample was performed using the Rat Oligo chip. As a result of GO enrichment analysis (biological process category) with the condition that the hypergeometric p-value is more than 0.05, we found that the following terms related to neuron differentiation were enriched: “axon regeneration” (–1 × log10(p)=3.86), “neuron projection development” (–1 × log10(p)=3.18), “axonogenesis” (–1 × log10(p)=2.89), “neuron projection morphogenesis” (–1 × log10(p)=2.58), “peripheral nervous system axon regeneration” (–1 × log10(p)=2.41), “forebrain neuron differentiation” (–1 × log10(p)=2.41), “nervous system development” (–1 × log10(p)=2.38), and “neuron development” (–1 × log10(p)=2.32) (Figure 5). Surprisingly, we also found that the following terms related to osteoblast differentiation were also enriched: “skeletal system development” (–1 × log10(p)=6.28), “ossification” (–1 × log10(p)=3.44), “endochondral ossification” (–1 × log10(p)=3.39), “positive regulation of canonical Wnt receptor signaling pathway” (–1 × log10(p)=3.19), “intramembranous ossification” (–1 × log10(p)=2.95), “cartilage development involved in endochondral bone morphogenesis” (–1 × log10(p)=2.78), “embryonic skeletal system development” (–1 × log10(p)=2.44), and “skeletal system morphogenesis” (–1 × log10(p)=2.36).

Figure 5. Both osteoblastic and neuronal gene expression levels were up-regulated in the contact co-culture in the microarray analysis using the Rat Oligo chip. GO enrichment analysis with the condition that the hypergeometric p-value is more than 0.05.

We filtered for genes with a Cy3/Cy5 ratio more than 2.0-fold in the biological process category by the search term “neuron differentiation or myelination”. As shown in Table III, we found five genes, i.e., TG interacting factor (Tgif), metalloproteinase inhibitor 2 precursor (TIMP-2), pro-neuregulin-1, membrane-bound isoform precursor (Nrg1), LIM/homeobox protein Lhx4 (Lhx4), and nerve growth factor (NGF)-inducible anti-proliferative protein PC3 (Btg2). We also filtered for genes with a Cy3/Cy5 ratio more than 2.0-fold in the biological process category by the search term “osteoblast, ossification or differentiation”. As shown in Table IV, we found five genes, i.e., fibronectin precursor (Fn1), macrophage colony-stimulating factor 1 precursor (Csf1), collagen alpha-1(I) chain precursor (Col1a1), homeobox protein Distal-Less Homeobox 5 (Dlx5), and matrix metalloproteinase 2 (MMP-2). These results suggest that not only rat neuron differentiation signals, but also osteoblast differentiation signals are increased in rat sympathetic neuronal cells that are co-cultured with murine osteoblasts.

Table III. Neuronal differentiation-related genes expressed by sympathetic neuronal cells using the Rat Oligo chip.

Tgif, TG interacting factor; Timp2, metalloproteinase inhibitor 2 precursor; Nrg1, pro-neuregulin-1, membrane-bound isoform precursor; Lhx4, LIM/homeobox protein Lhx4; Btg2, NGF-inducible antiproliferative protein PC3.

Table IV. Osteoblast differentiation-related genes expressed by sympathetic neuronal cells using the Rat Oligo chip.

Fn1, Fibronectin precursor; Csf1, macrophage colony-stimulating factor 1 precursor; Col1a1, collagen alpha-1(I) chain precursor; Dlx5, homeobox protein DLX-5; MMP-2, matrix metalloproteinase 2.

Discussion

Nerves accompanying the vasculature are distributed throughout skeletal tissues (17,18). In humans, innervation was significantly higher in the cortical pores than in the periosteum and bone marrow (19). In the periosteum, and during the healing of bone, osteoblasts encounter neuronal cells (20). Electron microscopy showed that osteoblasts are located near or are in contact with neuronal cells (8,9). Thus, contact co-culture of osteoblasts and neuronal cells is meaningful as it replicates, in part, the in vivo conditions.

Several studies have shown that the intracellular Ca²⁺ levels in osteoblasts are increased when grown in contact co-culture with neuronal cells. Obata et al. demonstrated that scorpion venom elicited Ca²⁺ mobilization in osteoblasts cultured with sympathetic neuronal cells via the α1-adrenergic receptor (11). Asada et al. reported that mechanical stimulation induced an increase in Ca²⁺ levels in osteoblasts that were co-cultured with dorsal root ganglia sensory neurons via P2X receptors (12). Moreover, dorsal root ganglion neurons were shown to induce an increase in the Ca²⁺ levels in osteoblasts via α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptors and natural killer receptors (13).

Takeda et al. demonstrated that the sympathetic nervous system positively modulates bone resorption (9). Fu et al. also showed that sympathetic signaling promotes osteoblast proliferation via β-adrenergic receptor (21). Another study showed that fenoterol, an agonist of β-adrenergic receptor and adrenaline, did not promote osteoblast differentiation, while Bone morphogenetic protein (BMP)-2 with adrenaline enhanced osteoblast differentiation more than BMP-2 alone (22).

In this study, we found that osteoblasts and sympathetic neuronal cells interact with each other. Interestingly, genes related to neuron differentiation were expressed by osteoblasts, and genes related to osteoblast differentiation were expressed by neurons. In osteoblasts, Bex1, Hoxc8, Nefl, Barhl2, Ascl1, Gata2, and Pigt were up-regulated. Among these genes, both Hoxc8 and Gata2 are negative regulators of osteoblast differentiation, indicating that they did not play a role in the osteoblast differentiation in our co-culture system (23,24). Our speculation is that unidentified factors derived from sympathetic neuronal cells may inhibit the negative function of Hoxc8 and Gata2 in osteoblast differentiation. In contrast, the levels of Fn1, Dlx5, and MMP-2, which are involved in the positive regulation of osteoblastogenesis, were dramatically increased in the sympathetic neuronal cells (25,26). As Biswas et al. reported that MMP-2 contributes to protection against neuronal cell death (27), it is reasonable to assume that MMP-2 was up-regulated in our co-culture system. However, Dlx5 is also associated with the promotion of gamma-Aminobutyric acid (GABA)ergic neuron differentiation in embryonic forebrain (28).

This study has certain limitations. First, the design of the study does not reflect the actual in vivo situation, because various other cells are present near the neurons and osteoblasts in vivo. Second, our study could not clarify whether the up-regulation of the Dlx5 gene in the sympathetic neuronal cells is related to osteoblastic differentiation or neuron differentiation. Third, the cells used in this study were derived from mouse and rat, and we could not exclude the possibility of cross-reactivity of the mouse and rat probes. Nevertheless, with these limitations, our study results indicated that the nervous system supports osteoblast differentiation, and that bone metabolism supports neuron differentiation. Further experiments are needed to elucidate in detail the relationship between the nervous system and bone metabolism.

Conflicts of Interest

The Authors have no conflicts of interest to declare.

Authors’ Contributions

MO and KI conducted all the experiments, collected the data and drafted the manuscript. FY, YI and YM analyzed the data. SA and MU critically revised the manuscript. TS designed the study and drafted the manuscript. All Authors gave final approval of the manuscript and agreed to be accountable for all aspects of the work.

Acknowledgements

The Authors would like to thank Dr. Masahito Matsumoto (Juntendo University) for his insightful comments and Mariko Hayakawa for excellent technical assistance.

Funding

This work was supported by JSPS KAKENHI Grant Number 17H04409 to Tsuyoshi Sato.