Nogo-A Regulates the Fate of Human Dental Pulp Stem Cells toward Osteogenic, Adipogenic, and Neurogenic Differentiation

Human teeth are highly innervated organs that contain a variety of mesenchymal stem cell populations that could be used for cell-based regenerative therapies. Specific molecules are often used in these treatments to favorably modulate the function and fate of stem cells. Nogo-A, a key regulator of neuronal growth and differentiation, is already used in clinical tissue regeneration trials. While the functions of Nogo-A in neuronal tissues are extensively explored, its role in teeth still remains unknown. In this work, we first immunohistochemically analyzed the distribution of Nogo-A protein in the dental pulp of human teeth. Nogo-A is localized in a variety of cellular and structural components of the dental pulp, including odontoblasts, fibroblasts, neurons and vessels. We also cross-examined Nogo expression in the various pulp cell clusters in a single cell RNA sequencing dataset of human dental pulp, which showed high levels of expression in all cell clusters, including that of stem cells. We then assessed the role of Nogo-A on the fate of human dental pulp stem cells and their differentiation capacity in vitro. Using immunostaining, Alizarin Red S, Nile Red and Oil Red O staining we showed that Nogo-A delayed the differentiation of cultured dental pulp stem cells toward the osteogenic, adipogenic and neurogenic lineages, while addition of the blocking anti-Nogo-A antibody had opposite effects. These results were further confirmed by qRT-PCR, which demonstrated overexpression of genes involved in osteogenic (RUNX2, ALP, SP7/OSX), adipogenic (PPAR-γ2, LPL) and neurogenic (DCX, TUBB3, NEFL) differentiation in the presence of the anti-Nogo-A antibody. Conversely, the osteogenic and adipogenic genes were downregulated by Nogo-A. Taken together, our results show that the functions of Nogo-A are not restricted to neuronal cells but are extended to other cell populations, including dental pulp stem cells. We show that Nogo-A regulates their fates toward osteogenic, adipogenic and neurogenic differentiation, thus indicating its potential use in clinics.

Surprisingly, there is yet no information on the function of Nogo-A in the highly innervated dental tissues, despite its reported expression in human teeth [73]. In this study we examined the distribution of the Nogo-A protein in the dental pulp of human teeth, and we assessed its role and effects on the plasticity of hDPSCs by analyzing their differentiation potential in vitro. These studies were complemented by gene expression analyses of specific stem cell, osteogenic, adipogenic and neuronal markers. Our results show that Nogo-A plays an important role in stem cell fate choices and differentiation events, and point to Nogo-A as a suitable candidate molecule for future regenerative therapies in the clinics.

Data and Code Availability
Data can be found at GEO; access number: GSE161267. All code is publicly available at: https://github.com/TheMoorLab/Tooth (accessed 7 April 2021). Single cell RNA sequencing data is described in detail in [11].

Collection of Human Tissues and Cells
Healthy human teeth were obtained from anonymous patients at the Center of Dental Medicine, University of Zurich. The Ethics Committee of the Canton of Zurich has approved the project (reference number 2012-0588) and the patients gave their written informed consent. The teeth were extracted by dentists at the Clinic of Cranio-Maxillofacial and Oral Surgery Department at the Center of Dental Medicine of the University of Zurich. All procedures were implemented following the current guidelines. Human dental pulp Cells 2022, 11, 3415 3 of 23 cells were isolated from the dental pulp of extracted wisdom teeth of healthy patients as previously described [11]. Briefly, the dental pulps were gently removed with forceps, minced with scalpels, rinsed with PBS and were finally enzymatically digested for 1 h at 37 • C in a solution of Collagenase (3 mg/mL; Life Technologies BV, Zug ZG, Switzerland) and Dispase (4 mg/mL; Sigma-Aldrich Chemie GmbH, Buchs SG, Switzerland).

Staining
Immunohistochemistry: Immunoperoxidase staining on paraffin sections was performed as previously described [74,75]. Briefly, 5 μm sections were deparaffinized, exposed to a 0.3% solution of hydrogen peroxide in methanol and then incubated overnight at 4 °C in a humid atmosphere either with the mouse monoclonal anti-Nogo-A antibody [76] diluted 1:100 in PBS/0.2% BSA. Peroxidase was revealed by incubation with 3-amino-9-ethylcarbazole (AEC) reaction solutions. After staining, the sections were mounted with Experimental plan for analyzing the effects of Nogo-A (Nogo-A ∆20 fragment) and anti-Nogo-A antibody on the osteogenic, adipogenic, and neurogenic differentiation of human dental pulp stem cells.

Staining
Immunohistochemistry: Immunoperoxidase staining on paraffin sections was performed as previously described [74,75]. Briefly, 5 µm sections were deparaffinized, exposed to a 0.3% solution of hydrogen peroxide in methanol and then incubated overnight at 4 • C in a humid atmosphere either with the mouse monoclonal anti-Nogo-A antibody [76] diluted 1:100 in PBS/0.2% BSA. Peroxidase was revealed by incubation with 3-amino-9ethylcarbazole (AEC) reaction solutions. After staining, the sections were mounted with Eukit. In control sections, the primary antibodies were omitted. Immunohistochemistry on cultured dental pulp cells was performed after 4% paraformaldehyde (PFA) in PBS fixation, followed by permeabilization of cells for 15 min with 0.5% Triton X-100 in PBS. Fixed and permeabilized samples were then incubated with the primary antibodies overnight at 4 • C, and the signal was detected by AEC. Upon staining, the slides were mounted with Aquamount (BDH, Gurr, UK). Controls were performed by incubations without primary antibodies.
Immunofluorescence Staining and Imaging: Polyclonal anti-Collagen type-I antibody (ab34710, Abcam, Cambridge, UK), rabbit anti-FABP4 monoclonal antibody (EPR3579, Abcam), rabbit anti-Neurofilament light chain (NFL; NB300-131, Novus biology, Littleton, Colorado, USA), rabbit anti-Tubulin Beta-III (TUBB3; ab18207, Abcam, Cambridge, UK), and alexa-488-conjugated anti-rabbit IgG (A21206, Abcam, Cambridge, UK) antibodies were used. The cells on IBIDI µ-Slide 8-well chamber slide were fixed with 4% PFA for 10 min, after which the fixative was removed and cells were washed twice with PBS. Next, the cells were permeabilized with PBST (0.1% Triton X-100 in PBS) for 10 min at room temperature (RT). Slides were then washed thrice with PBST (5 min each). Nonspecific binding was blocked by incubation with 5% BSA in PBST for 1 h at RT. The cells were incubated in primary antibodies for 1 h at RT. Slides were washed three times with PBST before incubation with Alexa Fluor ® 488 donkey anti-rabbit IgG secondary antibody (Invitrogen, Carlsbad, CA, USA) for 1 h at RT. Nuclei were stained with DAPI (1 ug/mL) for 10 min after which the chamber slides were mounted in ProLong™ Diamond Antifade Mountant (Invitrogen, P36961, Carlsbad, CA, USA). All images were acquired using Leica Stellaris confocal microscope.
Alizarin Red S Staining: The extend of mineralization was analyzed in hDPSCs cultured in osteogenic conditions, at different time points (7, 14, and 21 days). Cultures were washed twice with PBS and fixed with 4% PFA for 10 min at RT.   Table 1. All primers were synthesized by Microsynth (Microsynth AG, Balgach, Switzerland). All of the reactions were run in triplicate. After the reactions were realized, the Cq values were established by setting a fixed threshold. All measurements were normalized to GAPDH and the relative changes in target gene expression between the control and treatment groups were measured using the 2-∆∆ CT method [77]. Table 1. Gene sequences used for qRT-PCR for the analysis of the osteogenic, adipogenic and neurogenic potential of human dental pulp stem cells.

Target Genes
Sequence

Statistical Analysis
All qRT-PCR data were analyzed with Student's t-test. p values < 0.05 were considered statistically significant. Statistical analysis was accomplished with Microsoft Excel Software (Version 16.55, Microsoft Corporation, Redmond, WA, USA).

Localization of the Nogo-A Protein in Human Dental Pulp Cells
We analyzed the distribution of Nogo-A in the dental pulp of healthy permanent human teeth (Figure 2A-C) and found that this molecule was abundantly expressed within this tissue ( Figure 2D-H). Nogo-A immunoreactivity was observed in various dental pulp cell populations, including odontoblasts, fibroblasts, nerves and vessels, albeit at different intensities ( Figure 2D-H). Intense staining was detected in the nerves and blood vessels ( Figure 2B-H), while strong staining was also observed in odontoblasts ( Figure 2B,D,E). Immunohistochemical ( Figure 2I) and immunofluorescent ( Figure 2J) staining was also detected in dissociated human dental pulp cells that were cultured for 5 days in vitro. Immunostaining was not observed in control sections in which the Nogo-A antibody was omitted (data not shown).  We have performed recently a detailed single cell RNA sequencing analysis of the human dental pulp and periodontium [11]. We have interrogated this dataset to determine the number of cells expressing NOGO (RTN4) and the NOGO-related receptors (LINGO1, RTN4R, S1PR2, TNFRSF19, NGFR) in the stem cell compartment of the pulp (DPSC compartment; blue color in Figure 3A), which represents the 12% of the dental pulp tissue (4674 cells in a total number of 32,378 cells; Figure 3A,H) and highly express the FRZB, NOTCH3, THY1, and MYH genes [11]. In the DPSC compartment, 1523 cells expressed RTN4 in a total number of 4674 cells, which represent approximately the 33% of all DP-SCs ( Figure 3B,H). Cells expressing LINGO1, RTN4R, S1PR2, TNFRSF19, NGFR were less abundant in the DPSC compartment ( Figure 3C-H).

Nogo-A Effects on the In Vitro Osteogenic, Adipogenic, and Neurogenic Differentiation Potential of hDPSCs
The effect of Nogo-A on the multilineage differentiation potential of hDPSCs was assessed in cultures induced for osteogenic, adipogenic and neurogenic differentiation.
3.2.1. The Anti-Nogo-A Antibody Accelerates the Osteogenic Differentiation of hDPSCs hDPSCs obtained from healthy permanent teeth were cultured in osteogenic medium alone (control, group A), in osteogenic medium supplemented with the anti-Nogo-A antibody (group B), and in osteogenic medium supplemented with the active Nogo-A protein fragment ∆20 (group C). The osteogenic potential of hDPSCs in the three groups was assessed 7, 14 and 21 days upon culture by their ability to form extracellular calcium deposits (mineralized nodules), which were detected by Alizarin Red S staining ( Figure 4). hDPSCs from group A started to form mineralized nodules between 14 and 21 days of culture ( Figure 4E,H). hDPSCs from group B displayed early signs of osteogenesis and formed prematurely mineralized nodules as compared to hDPSCs of groups A and C (compare Figure 4B,E,H and Figure 4D,G,J with Figure 4C,F,I). In group A, the mineralized nodules were first observed at 14 days of culture ( Figure 4F). Their size and the extent of mineralization significantly and continuously increased up to the end of the culture period ( Figure 4I). In contrast, hDPSCs from group C delayed the onset of mineralization and reduced the overall mineralization capacity as compared to hDPSCs of groups A and B (compare Figure 4B,E,H and Figure 4C,F,I with Figure 4D,G,J). These observations were further confirmed by quantification of Alizarin Red S incorporated in the mineralized nodules, which showed its significant increase at 14 and 21 days in group B as compared to groups A and C ( Figure 4A). Conversely, hDPSCs from group C demonstrated significantly lower level of incorporated Alizarin Red S as compared to hDPSCs of groups A and B ( Figure 4A). Increased collagen type-I immunostaining was also observed in hDPSCs from group B ( Figure 4L) as compared to hDPSCs of groups A ( Figure 4K) and C ( Figure 4M).
(DPSC compartment; blue color in Figure 3A), which represents the 12% of the denta pulp tissue (4,674 cells in a total number of 32,378 cells; Figure 3A,H) and highly expres the FRZB, NOTCH3, THY1, and MYH genes [11]. In the DPSC compartment, 1,523 cell expressed RTN4 in a total number of 4,674 cells, which represent approximately the 33% of all DPSCs ( Figure 3B,H). Cells expressing LINGO1, RTN4R, S1PR2, TNFRSF19, NGFR were less abundant in the DPSC compartment ( Figure 3C-H).   We next analyzed by RT-qPCR the expression of stem cell (THY1/CD90 and END/CD105) and osteogenenesis-related (RUNX2, ALP and OSX) markers in these We next analyzed by RT-qPCR the expression of stem cell (THY1/CD90 and END/ CD105) and osteogenenesis-related (RUNX2, ALP and OSX) markers in these cultures ( Figure 5). After 7 days of culture, the expression of THY1/CD90 was significantly decreased in hDPSCs from group B, and it continuously declined until the end of culture ( Figure 5A). In contrast, hDPSCs from group B significantly increased the expression of END/CD105 ( Figure 5B). THY1/CD90 downregulation in hDPSCs from group B was accompanied by a significant increase of RUNX2 expression, indicating the initiation of osteogenic differentiation already at day 7 ( Figure 5C). A similar upregulation was observed in the expression of ALP ( Figure 5D) and SP7/OSX ( Figure 5E) after 14 and 21 days of culture in hDPSCs from group B, which marked their ongoing osteogenic differentiation. In contrast, the expression of THY1 was significantly increased and maintained high throughout the culture period in hDPSCs from group C ( Figure 5A). THY1 upregulation was associated with a considerable downregulation of RUNX2, ALP and OSX expression in hDPSCs from group C as compared to hDPSCs from groups A and B ( Figure 5C-E). These results suggest that Nogo-A is critical for maintaining the stemness of hDPSCs and inhibiting their differentiation into osteoblasts.
Cells 2022, 11, x FOR PEER REVIEW 11 o cultures ( Figure 5). After 7 days of culture, the expression of THY1/CD90 was significan decreased in hDPSCs from group B, and it continuously declined until the end of cult ( Figure 5A). In contrast, hDPSCs from group B significantly increased the expression END/CD105 ( Figure 5B). THY1/CD90 downregulation in hDPSCs from group B was companied by a significant increase of RUNX2 expression, indicating the initiation of teogenic differentiation already at day 7 ( Figure 5C). A similar upregulation was observ in the expression of ALP ( Figure 5D) and SP7/OSX ( Figure 5E) after 14 and 21 days culture in hDPSCs from group B, which marked their ongoing osteogenic differentiati In contrast, the expression of THY1 was significantly increased and maintained h throughout the culture period in hDPSCs from group C ( Figure 5A). THY1 upregulat was associated with a considerable downregulation of RUNX2, ALP and OSX express in hDPSCs from group C as compared to hDPSCs from groups A and B ( Figure 5C-These results suggest that Nogo-A is critical for maintaining the stemness of hDPSCs a inhibiting their differentiation into osteoblasts.  , and SP7/OSX (E) expression levels in hDPSCs cultured in osteogenic medium (Os) alone (black columns), in osteogenic medium supplemented with the anti-Nogo-A-antibody (red colored columns), and osteogenic medium supplemented with the Nogo-A ∆20 protein (green colored columns) for 7, 14 and 21 days. Values, normalized to Glyceraldaeyde-3-Phosphate Dehydrogenase (GAPDH), shown are the mean ± SD of triplicates. Statistical analysis used Student's t-test (significant difference at *, p < 0.05; **, p < 0.01).

The Anti-Nogo-A Antibody Accelerates the Adipogenic Differentiation of hDPCSs
hDPSCs can differentiate toward adipocytes if cultured at high cell density in presence of 10% FBS, Dexamethasone, Isobutyl-1-Methylxanthine, Indomethacin and Insulin. We tested the in vitro adipogenic differentiation potential of hDPSCs in adipogenic medium alone (control, group A), in adipogenic medium supplemented with the anti-Nogo-A antibody (group B), and in adipogenic medium supplemented with the Nogo-A ∆20 protein (group C), after 7, 14 and 21 days of culture ( Figure 6). The adipogenic differentiation of hDPSCs in the three groups was assessed by their capacity to form and accumulate intracellularly lipid droplets, which were identified by both Nile Red and Oil Red O staining ( Figure 6) [29]. Lipid globules in hDPSCs from group A were first detected with Nile Red staining at approximately day 10 and their number was increased by the end of the culture period (day 21) (Figure 6B,E,H). Nile Red and Oil Red O staining in group B hDPSCs was significantly higher at days 14 and 21 as compared to hDPSCs of group A (compare Figure 6C,F,I with Figure 6B,E,H). The level of Oil Red O staining in hDPSCs from group C was comparable to group A (compare Figure 6D,G,J with Figure 6B,E,H). Oil Red O staining quantification in hDPSCs showed a significant increase at all time points in group B as compared to groups A and C ( Figure 6A).
Higher adipogenesis in hDPSCs from group B was also associated with intense immunofluorescence staining for the fatty acid binding protein 4 (FABP4; Figure 6L), which was almost absent in hDPSCs from group C as compared to hDPSCs of group A (compare Figure 6M to Figure 6K).
RT-qPCR analysis of the LPL ( Figure 7C) and PPARG ( Figure 7D) adipogenesis-related genes showed significant upregulation of their expression in hDPSCs from group B already at day 7, indicating earlier onset of adipogenesis. High levels of LPL and PPARG expression in hDPSCs of group B were maintained throughout the 21 days culture period ( Figure 7C,D), corroborating the increased Oil Red O staining in these cultures ( Figure 6F,I). In contrast, LPL and PPARG mRNA expression in hDPSCs from group C was maintained at low levels throughout the culture period, which indicated delayed onset and inhibition of adipogenesis ( Figure 7C,D).
Nile Red staining at approximately day 10 and their number was increased by the end of the culture period (day 21) (Figure 6B,E,H). Nile Red and Oil Red O staining in group B hDPSCs was significantly higher at days 14 and 21 as compared to hDPSCs of group A (compare Figure 6C,F,I with Figure 6B,E,H). The level of Oil Red O staining in hDPSCs from group C was comparable to group A (compare Figure 6D,G,J with Figure 6B,E,H). Oil Red O staining quantification in hDPSCs showed a significant increase at all time points in group B as compared to groups A and C ( Figure 6A).  lated genes showed significant upregulation of their expression in hDPSCs from group B already at day 7, indicating earlier onset of adipogenesis. High levels of LPL and PPARG expression in hDPSCs of group B were maintained throughout the 21 days culture period ( Figure 7C,D), corroborating the increased Oil Red O staining in these cultures ( Figure  6F,I). In contrast, LPL and PPARG mRNA expression in hDPSCs from group C was maintained at low levels throughout the culture period, which indicated delayed onset and inhibition of adipogenesis ( Figure 7C,D). , LPL (C) and PPARG (D) in hDPSCs cultured in adipogenic medium (Ad) alone (black columns), in adipogenic medium supplemented with the anti-Nogo-A-antibody (red colored columns), and in adipogenic medium supplemented with the Nogo-A ∆20 protein (green colored columns) for 7, 14 and 21 days. Values, normalized to GAPDH, shown are the mean ± SD of triplicates. Statistical analysis used Student's t-test (significant difference at *, p < 0.05; **, p < 0.01).

The Anti-Nogo-A Antibody Accelerates the Neurogenic Differentiation of hDPSCs
hDPSCs undergo neurogenic differentiation if cultured under appropriate conditions (e.g., Neurobasal A medium supplemented with EGF and FGF) [13]. Addition of neurogenic medium (control, group A) induces generation of neuron-like cells within 14 days ( Figure 8A), positive for Neurofilement light chain (NFL) and Tubulin Beta-III (TUBB3) staining ( Figure 8D). Neurogenic medium supplemented with the anti-Nogo-A antibody (group B) affected both the cell bodies and the forming axons in differentiating hDPSCs ( Figure 8B), which were also NFL-and TUBB3-positive ( Figure 8E,H). Cell bodies in group B increased in size and presented a rounder, more neuron-like appearance ( Figure 8B), when compared to hDPSCs of group A (compare Figure 8B with Figure 8A). A similar increase was observed in the length of their axons and axon-like structures ( Figure 8B). In contrast, hDPSCs cultured in neurogenic medium supplemented with the Nogo-A ∆20 protein (group C) showed modest neurogenic differentiation ( Figure 8C) and less abundant NFL ( Figure 8F) and TUBB3 ( Figure 8I) staining as compared to hDPSCs from group A (compare Figure 8C,F,I with Figure 8A,D,G). B increased in size and presented a rounder, more neuron-like appearance ( Figure 8B), when compared to hDPSCs of group A (compare Figure 8B with Figure 8A). A similar increase was observed in the length of their axons and axon-like structures ( Figure 8B). In contrast, hDPSCs cultured in neurogenic medium supplemented with the Nogo-A Δ20 protein (group C) showed modest neurogenic differentiation ( Figure 8C) and less abundant NFL ( Figure 8F) and TUBB3 ( Figure 8I) staining as compared to hDPSCs from group A (compare Figure 8C,F,I with Figure 8A,D,G).  These changes were confirmed by RT-qPCR analysis for Doublecortin (DCX) Tubulin Beta-III (TUBB3), and NFL, which are markers for neuronal progenitors, early neuronal differentiation, and differentiated neurons, respectively. Expression of these genes, as well as that of CD105 but not of THY1, was significantly upregulated in hDPSCs from group B ( Figure 9A-E). A decrease in the values of TUBB3 ( Figure 9C) and NFL ( Figure 9E) was observed upon 21 days of culture, plausibly because the neuronal identity of the hDPSCs was acquired at earlier stages (i.e., day 14). Interestingly, THY1 expression was upregulated in hDPSCs of group C, suggesting THY1-expressing cells as a potential target of Nogo-A. differentiation, and differentiated neurons, respectively. Expression of these genes, as well as that of CD105 but not of THY1, was significantly upregulated in hDPSCs from group B ( Figure 9A-E). A decrease in the values of TUBB3 ( Figure 9C) and NFL ( Figure 9E) was observed upon 21 days of culture, plausibly because the neuronal identity of the hDPSCs was acquired at earlier stages (i.e., day 14). Interestingly, THY1 expression was upregulated in hDPSCs of group C, suggesting THY1-expressing cells as a potential target of Nogo-A. (E)) in hDPSCs cultured in neurogenic medium (Neur) alone (black columns), in neurogenic medium supplemented with the anti-Nogo-A antibody (red color columns), and in neurogenic medium supplemented with the Nogo-A ∆20 protein (green color columns) for 7, 14 and 21 days. Values, normalized to Glyceraldaeyde-3-Phosphate Dehydrogenase (GAPDH), shown are the mean ± SD of triplicates. Statistical analysis used Student's t-test (significant difference at *, p < 0.05; **, p < 0.01).

Discussion
Regenerative medicine offers alternative therapeutic approaches to the currently employed clinical practices for the repair or replacement of damaged tissues [16,23,46,[78][79][80]. Translation of experimental regenerative therapies to the patients is one of the biggest challenges of our time, which aim to advance a new medical discipline expecting to tackle major unmet healthcare needs [80][81][82][83]. Regenerative therapies may include both stem cell-based and molecule-based approaches to promote the repair of tissues and restore their structure and functions [23,82,84]. Mesenchymal stem cells (MSCs) have gained great attention in the past few decades due to their self-renewal, multilineage differentiation potentials, and immunomodulatory properties, which are important features for the effi-cacy of cell-based regenerative treatments [82]. Another advantage of MSCs is their fast availability in clinically relevant quantities [79,81]. Innovative approaches based on MSCs and specific molecules that are entering the clinical arena could also be used in dentistry in order to restore the structural integrity and physiology of teeth [23,30,38,47,79]. Pioneer in vitro and in vivo studies have already demonstrated the potential of human dental pulp stem cells (hDPSCs) for the regeneration of both dental and non-dental tissues in patients and in experimental animals [38,45,85]. However, a key challenge for tissue regenerative treatments is the efficient use of specific molecules since our knowledge about their way of action in the various cell types and contexts is mainly agnostic [46,79]. Indeed, most of the information is often based on in vitro experimental setups designed for distinct environments that do not necessarily reflect the in vivo patient-specific situation [82,83,86]. Several signaling pathways, including Wnt, Notch, TGF-β, EGF, VEFG and Neurotrophins have been used for controlling stem cell behavior in pathological conditions [25,[87][88][89][90]. The mode of action of these molecules can provide insights into their effects and adaptive behavior in specific cells and tissues during regenerative treatments.
Due to the beneficial effects of Nogo-A inactivation for the regeneration of injured nerve fibers [91][92][93][94], this molecule is the target in clinical trials aiming at the treatment of severe spinal cord lesions [72]. Antibodies or decoy receptors ('Nogo-traps') that block the function of Nogo-A could be tools of high potential for these therapies. Besides its role in neuronal regeneration, Nogo-A is a potential target for enhancing vascular regeneration and functionality since its inhibition restores vascularization in the central nervous system in Parkinson's disease [95] and upon stroke in mice [66]. These Nogo-A properties could also be applicable in dentistry for recovering tooth physiology upon injury by regenerating the damaged neuronal and vascular networks within the dental pulp.
Stem cell populations reside in specific niches that maintain stem cells in an undifferentiated status and guide their fates [3,5,78]. The dental pulp contains several stem cell niches in precise anatomic locations that are composed by fibroblasts, immune cells, vascular and neuronal structures [8,11,[18][19][20]89]. Our results show that these dental pulp compartments are characterized by high levels of Nogo-A expression and that the decision whether stem cells will remain in their quiescent, immature state or undergo differentiation is influenced by Nogo-A. Stem cell fate changes and cytodifferentiation are characterized by a shift in gene expression that involves the coordinated action of transcription factors, DNA methylation and chromatin remodeling activities [96][97][98]. The cell surface protein Thymocyte Antigen 1 (THY1, also known as CD90) is a marker for mesenchymal stem cells (MSCs) and hDPSCs and plays critical roles in the maintenance of their pluripotency and fate choices [31,99,100]. Our in vitro differentiation assays showed that addition of Nogo-A increases the levels of THY1 in hDPSCs in all experimental setups tested, suggesting that Nogo-A is critical for maintaining the stemness of hDPSCs and inhibiting their differentiation toward osteocytes, adipocytes, and neurons. In contrast, the anti-Nogo-A antibody significantly enhances hDPSCs differentiation, which is in accordance with previous studies showing increased differentiation potential of MSCs upon reduction of THY1 expression [101]. CD105 is the main glycoprotein of human vascular endothelium, which is commonly used as another MSC marker [10,40,102,103]. Antibody mediated blockade of Nogo-A decreased the levels CD105 in hDPSCs in all experimental groups. It appears that administration of the anti-Nogo-A antibody has opposite effects in the THY1and CD105-expressing hDPSCs, thus differently affecting the fate and differentiation potential of these two stem cell populations [11].
Our results highlight important, yet neglected, effects of Nogo-A in mineralized tissue formation that could improve the actual clinical treatments by accelerating the repair of dental or other hard tissues, such as dentin and bone. Enhancing bone regeneration is a big challenge for orthopedic research [42,81,84]. A faster mineralization allows quicker bone healing that is one of the critical steps in bone, craniofacial and dental regenerative medicine [16,42,79,84]. Nogo-A exerts an inhibitory role in hDPSCs osteogenic differentiation by reducing the expression of osteogenesis related genes, such as RUNX2, OSX and ALP, thus affecting the timing and the extent of mineralization. Indeed, onset of mineralization is accompanied by increased expression of these genes: RUNX2 is a major control gene for MSCs differentiation into preosteoblasts [104][105][106], OSX is involved in the transition of preosteoblasts to mature osteoblasts [107,108], and ALP is an early osteogenic gene essential for the mineralization process [109,110]. Their upregulation in hDPSCs cultured in presence of the anti-Nogo-A antibody correlates with the accelerated and extensive formation of mineralized nodules. While in vitro differentiation assays do not faithfully mimic the physiological environment and cannot be yet translated to the clinics, they point to exciting new possibilities for improving hard-tissues repair. Future follow-up in vivo studies using small or larger animal models will be very useful in order to obtain more robust conclusions concerning the biological functions of Nogo-A in osteogenesis and bone-tissue regeneration. However, studies involving animals have become extremely complex and demanding lately, and therefore require several years of work to achieve these results.
The present findings demonstrate that Nogo-A also regulates hDPSCs fate toward adipogenic differentiation. While hDPSCs do not exhibit extensive adipogenesis as other MSC populations (e.g., bone marrow stem cells) [13,29], previous studies showed they are able to generate a limited number of adipocytes evidenced by increased expression of the adipogenesis genes, including Peroxisome Proliferator-Activated Receptor-gamma2 (PPAR-γ2) and Lipoprotein Lipase (LPL), as well as increased Nile Red and Oil Red O staining of intracellular lipid droplets [29]. The transcription factor PPAR-γ2 is a master regulator of adipogenesis and glycose/lipid metabolism [111,112] and LPL plays a central role in lipid metabolism, lipid transport, and triglycerides concentration in preadipocytes [113,114]. Addition of the anti-Nogo-A antibody significantly upregulated LPL and PPAR-γ2 expression and enhanced Oil Red O staining in hDPSCs cultured in adipogenic conditions, although the shift shape of hDPSCs from fibroblastic to spherical that characterizes adipogenic differentiation is not observed [115].
The neurogenic potential of hDPSCs is assessed by the increased expression of the neuronal genes Doublecortin (DCX), Neurofilament light chain (NFL) and Tubulin β-III. DCX is important for modulating and stabilizing microtubules to ensure effective migration of neuronal progenitor cells [116,117]; NFL is a structural protein forming neurofilaments and plays a role in the regulation of synaptic transmission and organelle trafficking [118]; and Tubulin β-III expression could be a common feature of stem cells and neuronal cells [119], confirming hMSCs differentiation toward neurons [120]. Addition of the anti-Nogo-A antibody upregulated DCX, NFL and Tubulin β-III expression in hDPSCs cultured in neurogenic medium. Previous studies have shown that upregulation of these genes correlates with accelerated neuronal outgrowth [51,52,91,121]. A small interfering RNA (siRNA) technique has been explored both in vivo and in vitro to specifically suppress Nogo-A to promote axonal growth and repair in a multiple sclerosis murine model [122]. However, the efficiency of the Nogo-A silencing was only 75% in cultures of an established cell line (N2A), suggesting that the efficiency of siRNA Nogo-A in our experimental setup is likely to be even lower, since we used hDPSCs from primary cultures. Although a confirmation of our results by a siRNA knock down experiment would be interesting, a negative result would not be telling due to potentially insufficient knock down.

Conclusions
In conclusion, our results show that Nogo-A is expressed in the various cell populations composing the dental pulp of human teeth, including odontoblasts, fibroblasts, cells of the vascular and neuronal networks and hDPSCs. The present in vitro findings demonstrate that Nogo-A plays an important role in the fate choices and multilineage differentiation potential of hDPSCs by maintaining their stemness and regulating their differentiation toward adipocytes, osteocytes, and neurons ( Figure 10). Inhibition of its function results in faster cell differentiation events, pointing to Nogo-A as a suitable candi-date molecule for future regenerative therapies in clinics in order to restore tissue integrity and function.
In conclusion, our results show that Nogo-A is expressed in the various cell populations composing the dental pulp of human teeth, including odontoblasts, fibroblasts, cells of the vascular and neuronal networks and hDPSCs. The present in vitro findings demonstrate that Nogo-A plays an important role in the fate choices and multilineage differentiation potential of hDPSCs by maintaining their stemness and regulating their differentiation toward adipocytes, osteocytes, and neurons ( Figure 10). Inhibition of its function results in faster cell differentiation events, pointing to Nogo-A as a suitable candidate molecule for future regenerative therapies in clinics in order to restore tissue integrity and function. Informed Consent Statement: Informed consent was obtained from all subjects involved in the study, as mentioned in the materials and methods section  Informed Consent Statement: Informed consent was obtained from all subjects involved in the study, as mentioned in the materials and methods section. Data Availability Statement: Not applicable.