The role and potential mechanism of p75NTR in mineralization via in vivo p75NTR knockout mice and in vitro ectomesenchymal stem cells

Abstract Objective The aim of this study is to investigate the role and potential mechanism of p75NTR in mineralization in vivo using p75NTR‐knockout mice and in vitro using ectomesenchymal stem cells (EMSCs). Materials and methods Femur bone mass and daily incisor mineralization speed were assessed in an in vivo p75NTR‐knockout mouse model. The molecular signatures alkaline phosphatase (ALP), collagen type 1 (Col1), melanoma‐associated antigen (Mage)‐D1, bone sialoprotein (BSP), osteocalcin (OCN), osteopontin (OPN), distal‐less homeobox 1 (Dlx1) and Msh homeobox 1 (Msx1) were examined in vitro in EMSCs isolated from p75NTR+/+ and p75NTRExIII−/− mice. Results p75NTR‐knockout mice were smaller in body size than heterozygous and wild‐type mice. Micro‐computed tomography and structural quantification showed that the osteogenic ability of p75NTRExIII‐knockout mice was significantly decreased compared with that of wild‐type mice (P < .05). Weaker ALP and alizarin red staining and reduced expression of ALP, Col1, Runx2, BSP, OCN and OPN were also observed in p75NTRExIII−/− EMSCs. Moreover, the distance between calcein fluorescence bands in p75NTRExIII‐knockout mice was significantly smaller than that in wild type and heterozygous mice (P < .05), indicating the lower daily mineralization speed of incisors in p75NTRExIII‐knockout mice. Further investigation revealed a positive correlation between p75NTR and Mage‐D1, Dlx1, and Msx1. Conclusion p75NTR not only promotes osteogenic differentiation and tissue mineralization, but also shows a possible relationship with the circadian rhythm of dental hard tissue formation.


| INTRODUC TI ON
The p75 neurotrophin receptor (p75NTR) is a 75-kDa transmembrane protein that is a member of the tumour necrosis factor receptor (TNFR) superfamily and is also known as nerve growth factor receptor, TNFR superfamily member 16 or CD271. 1 It has been reported to participate in multiple intracellular signalling pathways to regulate a wide range of biological functions, including stem cell differentiation, cell adhesion, tumour cell invasion, apoptosis, signal transduction and metastasis. [2][3][4][5] p75NTR has been widely used as a marker of isolated oesophageal epithelial stem cells, 6 adipose tissue-derived mesenchymal stem cells (MSCs), 7 and MSCs in the growth zones of regenerating fallow deer antlers and from the pedicle periosteum. 8 Recently, reports have increasingly shown that p75NTR is involved in tooth morphogenesis and development. 9,10 p75NTR has been used as a marker of isolated cranial neural crest-derived ectomesenchymal stem cells (EMSCs), 11,12 which present a useful stem cell model to investigate the mechanism of tooth morphogenesis and development. This is because this stem cell population gives rise to cells of the dental papilla and dental follicle, which subsequently form all tooth tissues except for enamel. In subsequent studies, p75NTR was reportedly positively related to the in vitro mineralization of EMSCs and promoted EMSCs differentiating into cells such as cementoblasts and odontoblasts. [13][14][15] We previously revealed that p75NTR was strongly expressed at the cap and bell stages during tooth development and showed similar expression patterns as those of the mineralization-related marker Runx2, 16 implying the important role of p75NTR in tooth morphogenesis, especially during dental hard tissue mineralization. Some researchers have speculated that the effect of p75NTR on tooth development might be related to the Wnt/β-catenin pathway and the factors sclerostin and melanoma-associated antigen (Mage)-D1. 13,17,18 However, regulation by these transcription factors alone does not sufficiently explain the exact molecular mechanism of p75NTR in tooth morphogenesis and development.
p75NTR-knockout mice are a good animal model used to investigate the specific role and mechanism of p75NTR in the morphogenesis of a variety of organs. 19 There are two types of p75NTR-knockout mice used in the literature, namely p75NTR ExIIIand p75NTR ExIV -knockout mice. 20,21 p75NTR ExIII -knockout mice are hypomorphic because they express a short variant of p75NTR as a consequence of alternative splicing. p75NTR ExIV -knockout mice were created by deleting exon IV, resulting in a loss of both the full-length and short isoform of p75NTR. p75NTR ExIII -knockout mice, which were reported by Lee et al 22 in 1992, were selected for this study. These mutant mice have a targeted deletion of exon III of p75NTR, and no functional p75NTR mRNA, protein or crosslinked products were detected in homozygous embryos.
Genetic rescue further confirmed that the mutant phenotype described above was caused by the targeted mutation of the gene encoding p75NTR. Previous reports have shown that homozygous mice were a good model for studies on the biological functions of p75NTR. 20-23 p75NTR ExIII -knockout mice, obtained from the Jackson Laboratory, were used to reveal the role and potential mechanism of p75NTR in mineralization during tooth morphogenesis and development.
Previous studies have indicated that p75NTR might play an important role in tooth morphogenesis and EMSC mineralization. However, the exact mechanism is unclear. The present study aimed to elucidate the role and potential mechanism of p75NTR in tooth morphogenesis and tissue mineralization via an in vivo study of p75NTR-knockout mice and an in vitro study of EMSCs isolated from both p75NTR ExIII−/− and p75NTR +/+ mice. The findings will contribute to the understanding of the molecular mechanism underlying tooth development and promote dental tissue engineering.

| Genotype identification of p75NTR-knockout mice
p75NTR-knockout mice were obtained from the Jackson Laboratory and housed under specific-pathogen-free conditions (22°C, 12/12-hour light/dark cycle, 50%-55% humidity) in the Chongqing Medical University Animal Laboratory. All animal experiments were performed in accordance with protocols approved by the Medical Ethics Committee of the Chongqing Medical University. p75NTR-knockout and wild-type littermates were generated by mating between heterozygous females and males ( Figure 1A and 1B). Genotyping of tail DNA was performed to distinguish p75NTR-knockout from wild-type and heterozygous progenies by polymerase chain reaction (PCR) as previously described 24 ( Figure 1B).

| Calcein fluorescence assay
Nine newborn male mice (three for each group: p75NTR-knockout, wild type and heterozygous) were routinely fed for 40 days.
Calcein was administered by intraperitoneal injection at 25 mg/ kg every fifth day for four times, and the mice were sacrificed at the second day after the final injection. The calcein fluorescence assay was performed as previously described. 25 Lower jaws dissected from p75NTR-knockout, wild-type and heterozygous mice were fixed in 2.5% glutaraldehyde and dehydrated with ethanol in a graded concentration series. The hard tissue specimens were embedded, sliced at 8 μm with the EXAKT precision cutting and grinding system (EXAKT Vertriebs GmbH) and observed using an upright fluorescence microscope (Olympus) with excitation and emission wavelengths of 485 nm and 510 nm, respectively.
The distance between the observed fluorescence bands was measured, and the mineral apposition rate (μm/d) was calculated using Image-Pro Plus 6.0.

| Identification of p75NTR +/+ and p75NTR ExIII−/− EMSCs
Flow cytometry was carried out to identify isolated p75NTR +/+ and F I G U R E 1 Generation and genotyping of p75NTR-knockout mice. Heterozygous female and male (A) were mated to generate the three types of littermates: p75NTR-knockout, wild type and heterozygous (B). The littermates with bands detected at both 280 bp and 345 bp were identified as heterozygous mice, and those with one band detected at 280 bp or 345 bp only were identified as p75NTR ExIII−/− -knockout or wildtype (p75NTR +/+ ) mice, respectively (C). Abbreviations: H, heterozygous; K, knockout; W, wild type 2.6 | Cell cycle assay p75NTR +/+ and p75NTR ExIII−/− EMSCs at passages 3 were collected for cell cycle analysis. The cells were trypsinized with 1% trypsin/1 mmol/L EDTA solution and centrifuged at 800 rpm for 5 minutes. The supernatant was removed, and the cells were washed twice with phosphate-buffered saline (PBS). Then, 2 mL of cold 70% dehydrated alcohol was added quickly to fix the cells at 4°C for 24 hours. The samples were washed with PBS and incubated with 100 μg/mL RNase A at 4°C for 30 minutes. Thereafter, the cells were filtered, stained with 2 mg/mL propidium iodide at 4°C for 30 minutes and analysed using FACS Calibur flow cytometry.

| Alkaline phosphatase and alizarin red staining
p75NTR +/+ and p75NTR ExIII−/− EMSCs were seeded in 6-well plates at a density of 1 × 10 5 cells/well and incubated in osteogenic medium (containing 50 mg/mL ascorbic acid, 10 mmol/L β-glycerol phosphate and 10 − 8 m dexamethasone), and the medium was changed every 3 days. On days 3 and 7, the cells were washed twice with PBS, fixed in 4% paraformaldehyde for 30 minutes and subjected to alkaline phosphatase (ALP) staining using a kit (Beyotime) according to the manufacturer's instructions. After 14 days of incubation in osteogenic medium, the cells were subjected to both ALP and alizarin red staining (Sangon). After staining, the cells were washed three times with distilled water and observed using a phase-contrast microscope.

| Real-time PCR
Total RNA was extracted from approximately 1 × 10 6 EMSCs using Trizol reagent (Invitrogen) according to the manufacturer's protocol.
RNA was quantified and reverse-transcribed into cDNA using the RevertAidTM First Strand cDNA Synthesis Kit (MBI Fermentas) according to the manufacturer's instructions. Real-time PCR (RT-PCR) was performed as previously described 12 to further confirm the findings on the role and mechanism of p75NTR in mineralization.
PCR amplification was performed for 30 cycles in a thermal cycler, with initial denaturation at 94°C for 30 seconds, subsequent annealing at 60°C for 60 seconds and extension at 72°C for 90 seconds.
The PCR products were visualized on a 1.5% agarose gel containing 5 mg/mL ethidium bromide. The primers used are listed in Table 1.

Primer sequences
GenBank ® Accession no.

| Statistical analysis
Data for the calcein fluorescence assay, bone analysis, growth curve and CCK-8 assays, greyscale analysis and RT-PCR were presented as the mean ± standard deviation. Statistical significance was assessed using Prism 5 (GraphPad Software). Comparisons were made using a t test or one-way analysis of variance (Tukey's test) for experiments involving more than three groups. All experiments were performed three times, and differences were considered significant at P < .05.

| Identification and visual observation of p75NTR-knockout mice
The genotyping results are shown in Figure 1C. The littermates with two bands detected at 280 bp and 345 bp were identified as heterozygous mice, and those with one band detected at either 280 bp or 345 bp only were identified as p75NTR ExIII−/− -knockout or wildtype (p75NTR +/+ ) mice, respectively. When the littermates grew to 8 weeks of age, an obvious difference in body size was observed ( Figure 1B). p75NTR-knockout mice (length 7.2 cm; weight 18.6 g) were smaller than the wild-type (length 8.7 cm; weight 21.1 g) and heterozygous mice (length 8.4 cm; weight 20.6 g).

| Daily incisor mineralization speed
Fluorescence microscopic observation showed that the daily mineralization speeds of the incisors were different between the three types of mice, as observed in the calcein fluorescence assay ( Figure 2A). The distance between the calcein fluorescence bands, representing the mineralization on every fifth day, was 20.84 μm in p75NTR-knockout mice, which was significantly lower than that in wild-type (28.72 μm) and heterozygous mice (31.60 μm) ( Figure 2B; P < .01). No significance was found between wild-type mice and heterozygous mice (P > .05). The data indicated that p75NTR might participate in the regulation of the daily mineralization speed of mouse incisors.

| Bone mass of p75NTR-knockout and wildtype mice
Micro
The proliferation assays showed that both cell populations began to grow exponentially from day 2 and the population doubling times were 29.17 hours for p75NTR ExIII−/− EMSCs and 28.92 hours for p75NTR +/+ EMSCs. There was no significant difference in cell proliferation ability between p75NTR +/+ and p75NTR ExIII−/− EMSCs in the cell cycle, growth curve and CCK-8 assays (P > .05) ( Figure 5B and 5C).  Figure 6B) and RT-PCR ( Figure 6C). All of these factors were significantly higher in p75NTR +/+ EMSCs at both mRNA and protein levels (P < .05) after 14 days of culture in osteogenic medium. These results confirmed that p75NTR ExIII−/− EMSCs possessed lower mineralization ability than p75NTR +/+ EMSCs.

| Potential mechanism of p75NTR in regulating mineralization
Based on a previous report, 16 this study continued to investigate the potential mechanism of p75NTR in regulating mineralization.
Immunohistochemical staining demonstrated that the expres- Mage-D1 siRNA transfection confirmed this speculation. Runx2, F I G U R E 4 Isolation and genotypic identification of mouse embryonic EMSCs. E13.5 heterozygous mice were selected and each embryo was separated (A, B). The embryonic maxillofacial processes were dissected and minced into fine pieces for the culture of EMSCs (C-F). EMSCs exhibited a fibroblastlike morphology (G, H). Genotypic identification was shown for each embryo used to isolate EMSCs (I). Scale bar represents 50 μm. Abbreviations: H, heterozygous; K, knockout; W, wild type F I G U R E 5 Characterization of p75NTR +/+ and p75NTR ExIII−/− EMSCs. The MSC markers CD29, CD90 and CD146 were highly expressed in both p75NTR ExIII−/− and p75NTR +/+ EMSCs, while the hematopoietic marker CD45 was hardly detected (A). Cell cycle assay showed no significant difference between p75NTR +/+ and p75NTR ExIII−/− EMSCs in proliferation ability (B). Growth curves and CCK-8 assay (C) showed that both cell populations began to grow exponentially on day 2 and the population doubling times were 29.17 h for p75NTR ExIII−/− EMSCs and 28.92 h for p75NTR +/+ EMSCs, calculated by the formula PDT = T × log2/ (logNt -logN0), T: day of culture, Nt: number of cells on day T, N0: number of cells on day 0. Abbreviations: K, knockout; W, wild type F I G U R E 6 Mineralization assay. After three days of osteogenic induction, ALP staining was hardly detected in both p75NTR +/+ and p75NTR ExIII−/− EMSCs (A), but on day 7, more abundant and deeper ALP staining was present in p75NTR +/+ EMSCs compared with that in p75NTR ExIII−/− EMSCs. On day 14, this difference became more prominent in not only ALP but also alizarin red staining. ALP staining was deeper and the mineralized nodules of alizarin red staining were larger in p75NTR +/+ EMSCs. Western blot on day 14 showed increased expression of the mineralization-related markers ALP, Col1, Runx2, BSP, OCN and OPN in p75NTR +/+ EMSCs (B). Similar results were obtained by RT-PCR on day 14 (C). The mRNA expression of ALP, Col-1, Runx2, BSP, OCN and OPN in p75NTR +/+ EMSCs was significantly higher than that in p75NTR ExIII−/− EMSCs. Scale bar represents 100 μm. Abbreviations: K, knockout (p75NTR ExIII−/− ); W, wild type (p75NTR +/+ ) a mineralization-related marker, was significantly down-regulated when Mage-D1 was suppressed by siRNA treatment (Figure 7B), demonstrating that Mage-D1 positively regulated Runx2 expression. To further examine the potential mechanism of p75NTR in tooth morphogenesis, the homeobox genes Dlx1 and Msx1 were detected. The results in Figure 7C indicate that both Dlx1 and Msx1 were weakly expressed in p75NTR ExIII−/− EMSCs but strongly expressed in p75NTR +/+ EMSCs, revealing their positive correlation with p75NTR.

| D ISCUSS I ON
The sequential and reciprocal interactions between the oral epithelium and cranial neural crest-derived mesenchyme trigger tooth morphogenesis. 26,27 In this process, EMSCs, which arise from the cranial neural crest, migrate to and populate the branchial arches, giving rise to cells of the dental papilla and dental follicle and subsequently forming dentin, pulp, cementum and periodontal ligaments. 28 p75NTR, which is abundantly expressed in cranial neural crests, is considered a typical marker for this stem cell population. 29 It has been used to successfully select cranial neural crest-derived EMSCs (p75 + EMSCs) in vitro, providing a useful stem cell model for research on tooth morphogenesis. 12 Moreover, p75NTR has been reported to participate in tooth morphogenesis and dental hard tissue mineralization. [13][14][15][16][17] However, most molecular signatures on the effect of p75NTR in tooth morphogenesis remain to be uncovered, and the elusive mechanism underlying tooth development has severely restricted dental tissue engineering thus far. In this study, p75N-TR ExIII -knockout mice and EMSCs isolated from p75NTR ExIII−/− and p75NTR +/+ male mice were investigated to reveal the exact mechanism of p75NTR in mineralization regulation, contributing to dental tissue engineering.
Knockout mice are a useful method of investigating the effects of a specific molecule and examining the potential mechanism within molecular signature networks. Recently, p75NTR-knockout mice have been widely used in studies on neurological diseases and neural regeneration. 19,30,31 This model greatly contributes to the understanding of disease pathogenesis and exploration of methods to reverse neurodegeneration or achieve nerve regeneration. In this study, p75NTR ExIII -knockout mice were used to reveal the mechanism of tooth morphogenesis and tissue mineralization. Interestingly, p75NTR-knockout mice were found to be smaller in body shape than wild-type and heterozygous mice.
Moreover, micro-CT and bone microstructural parameters analysis in p75NTR-knockout mice showed obvious bone loss in both the femur trabecular and cortical bone, implying that the osteogenic potential was remarkably decreased in the absence of p75NTR. These in vivo data confirmed the findings of previous in vitro studies on the effect of p75NTR in mineralization. 13,17,18 F I G U R E 7 Investigations revealing the potential mechanism of p75NTR in mineralization. Immunohistochemistry showed similar expression patterns for p75NTR and Mage-D1 (A). They were strongly expressed in the same areas at the cap stage (E13.5 d), but their expression became weak at the bell stage (E15.5 d) and at the beginning of dental hard tissue formation (E18.5 d). Western blot showed that Runx2 was significantly decreased when Mage-D1 was downregulated by siRNA transfection (B). The homeobox genes Dlx1 and Msx1 were both weakly expressed in p75NTR ExIII−/− EMSCs but strongly expressed in p75NTR +/+ EMSCs (C). Abbreviations: K, knockout; W, wild type Mineralization-related markers such as Runx2, ALP, Col1, BSP, OCN and OPN have been reported to display a positive correlation with p75NTR in rat EMSCs in vitro. These findings supported the speculation that p75NTR plays an up-regulatory role in the osteogenic differentiation of stem cells and hard tissue formation.
In previous studies on the effect of p75NTR in tooth development, EMSCs were mostly isolated from the facial processes of rat embryos. To confirm the abovementioned speculation, EMSCs were isolated from mice in this study. Both p75NTR +/+ and p75NTR ExIII−/− EMSCs showed high proliferation ability and were well characterized as MSCs. However, a significant difference in cell mineralization ability was observed between the two stem cell populations. ALP and alizarin red staining were weak in p75NTR ExIII−/− EMSCs, and the mineralization-related genes ALP, It is well known that the formation of dental hard tissues is periodic, and the growth lines are an evidence of this phenomenon.
Based on p75NTR-knockout mice, we found that p75NTR might play Dlx and Msx, which belong to the homeobox gene family, have been widely recognized as key factors in craniofacial and tooth development. 35,36 Evidence has shown that Msx1 was expressed in the mesenchyme subjacent to the dental lamina of mouse E11.5 d embryos, and its expression became the strongest in the dental papilla and dental follicle at the cap stage (E13.5 d). 37,38 Dlx1 was also reportedly strongly expressed in the areas near epithelial-mesenchymal interactions. 39 Interestingly, we showed that p75NTR and Mage-D1 were strongly expressed in the mesenchyme of the dental papilla and dental follicle and the inner enamel epithelium, similar to the expression of Dlx1 and Msx1 reported in previous studies.
Therefore, we speculated that there might be connections between p75NTR, Mage-D1 and Dlx/Msx. In this study, Dlx1 and Msx1 were expressed at low levels in p75NTR ExIII−/− EMSCs, implicating their positive correlation with p75NTR. Mage-D1 was also found to be positively correlated with p75NTR, and thus, we suggest that cer-

ACK N OWLED G EM ENTS
This study was supported by the National Natural Science

CO N FLI C T O F I NTE R E S T
All authors declare that they have no competing interests.

AUTH O R CO NTR I B UTI O N S
All authors contributed to the study concept and design. ZM and LG

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.