In vivo cell proliferation analysis and cell-tracing reveal the global cellular dynamics of periodontal ligament cells under mechanical-loading

Periodontal ligament (PDL) is a uniquely differentiated tissue that anchors the tooth to the alveolar bone socket and plays key roles in oral function. PDL cells can respond rapidly to mechanical stimuli, resulting in accelerated tissue remodeling. Cell proliferation is an initial event in tissue remodeling and participates in maintaining the cell supply; therefore, analyzing cell-proliferative activity might provide a comprehensive view of cellular dynamics at the tissue level. In this study, we investigated proliferating cells in mouse molar PDL during orthodontic tooth movement (OTM)-induced tissue remodeling. Our results demonstrated that the mechanical stimuli evoked a dynamic change in the proliferative-cell profile at the entire PDL. Additionally, cell-tracing analysis revealed that the proliferated cells underwent further division and subsequently contributed to tissue remodeling. Moreover, OTM-induced proliferating cells expressed various molecular markers that most likely arise from a wide range of cell types, indicating the lineage plasticity of PDL cells in vivo. Although further studies are required, these findings partially elucidated the global views of the cell trajectory in mouse molar PDL under mechanical-loading conditions, which is vital for understanding the cellular dynamics of the PDL and beneficial for dental treatment in humans.

www.nature.com/scientificreports/ In the continuously growing mouse incisor, the entire spectrum of developmental and adult stages of the cells is seen throughout life; thus, it is widely used to study the differentiation trajectory of stem cells. In contrast to the highly proliferative nature of stem cells under culture, stem cells remain quiescent in vivo and reside in a specialized microenvironment. In the mouse incisor, a small population of slow-cycling MSCs exists at the proximal end of the epithelial cervical loops 15 . Immediately distal to the stem cell compartment, rapidly cycling, undifferentiated, transit-amplifying cells (TACs) are present 16 . Rapid-cycling TACs in mouse incisor mesenchyme express specific genes, such as polycomb repressive complex 1 (PRC1) 17 . Targeted deletion of PRC1 components, such as Ring1a and Ring1b, leads to the loss of TACs and results in the arrest of incisor growth 16 . This suggests that the transition of slow-cycling MSCs to rapidly cycling progenitor cells is a critically important process for tissue maintenance. Therefore, analyzing the cell-proliferative activity might offer a comprehensive view of the cell-differentiation trajectory at the tissue level.
Although cell proliferation is a favorable indicator of cell-differentiation trajectory, mouse molar PDL, as well as human tooth PDL, are stationary in terms of proliferative activity under physiological conditions. A unique feature of PDL cells is a rapid response to mechanical stimuli, which is well documented during OTM 8,18 . Therefore, we conceived the idea that changes in mechanical-loading-induced cell proliferation activity would expose the cell-differentiation trajectory in mouse molar PDL. Here, we determined the detailed characteristics of mechanical-loading-induced proliferating cells and the fate of these cells in order to obtain a comprehensive view of the cellular dynamics in PDL.

Results
Cell-proliferative status during OTM. The proliferative status of the mouse molar PDL is stationary under physiological conditions, similar to human tooth PDL. To expose the proliferative activity of cells in mouse molar PDL, we used an OTM model, in which tissue remodeling is accelerated by constant mechanical-loading using a coil spring (Fig. 1a). In the control, the distance between the first and second molars was 0.0 ± 0.0 μm. At 1 and 2 weeks, the distance increased to 95.0 ± 52.0 μm and 238.6 ± 83.0 μm (n = 5), respectively (Fig. 1b,c), confirming that the first molar was moved to the mesial side, resulting in the mesial side PDL receiving compressive force, and the distal PDL suffering tension force. We chose the distal side PDL of the distal root (i.e., the side with tension) as the region of interest (Fig. 1c, white rectangle). Hematoxylin and eosin (H&E) staining showed that extension of the PDL fibers was evident at as early as 24 h of OTM. (Fig. 1d), with extension and dis-alignment of the PDL fiber persisting for up to 72 h. Proliferating cells among PDL cells were detected using 5-ethynyl-2′-deoxyuridine (EdU) incorporation and the PCNA antibody. EdU + cells were rarely detected in the control, but they exponentially increased at 36 h of OTM and then decreased as time progressed (n = 3) (Fig. 1e). The PCNA + cells showed the same tendency but did not fully overlap with the EdU + cells shown in the superimposed image at 36 h. The proliferating cells were distributed evenly in the PDL, and we observed no correlation with the architectural component, such as blood vessels and mineralized tissue boundaries. Moreover, following the early response of the proliferating cells to the OTM, the total cell number increased in PDL (n = 3) (Fig. 1f).
We then analyzed the cell-proliferative status using a Fluorescent Ubiquitination-based Cell Cycle Indicator (Fucci2) mouse 19,20 . In this mouse, the cells at the S/G2/M phase express mVenus (green fluorescence protein) under control of the hGem (1/110) promoter, and cells in the G1 phase express mCherry (red fluorescence protein) under control of the hCdt1(30/120) promoter, thereby allowing cell cycle status to be distinguished by two different fluorescence probes. However, the mVenus signal was not detected in the PDL of 8-week-old mice (Supp. Fig. 1). Therefore, we used EdU administration to label proliferating cells in Fucci2 mice. In the control, the majority of PDL cells expressed mCherry, indicating that they were in the quiescent phase ( Fig. 2a). At 36 h, the quiescent cells decreased dramatically, corresponding to an increase in the number of proliferating cells. At 2 weeks, the majority of cells returned to the quiescent phase, whereas proliferative activity persisted in some cells (n = 3) (Fig. 2b).

Characterization of the proliferating cells during OTM.
To further characterize the OTM-induced proliferating cells in PDL, we analyzed the expression of cellular markers by immunohistochemistry (Fig. 3). We chose the following markers: Gli1 for MSCs; Ring1b for TACs; and Nfic, Cbfa1, and Osx for root/PDL-forming cells. Representative images of the immunoreactivity to the respective antibodies (red) together with the proliferating cells (green) at 36 h of OTM are shown in Fig. 3a. At 36 h of OTM, the ratio of marker + cells among total PDL cells was analyzed, and no significant difference was observed compared to control (n = 3) (Fig. 3b). Additionally, the ratio of marker + cells within the EdU + proliferating cells showed a diverse expression pattern of the markers, indicating that the proliferating cells comprised a wide range of cell types (n = 3) (Fig. 3c).

Fate of OTM-induced proliferating cells.
EdU-labeled cells at 36 h after OTM were further traced for 4 weeks (Fig. 4a). Proliferative cells were detected at the distal side of the distal root, apical, and furcation areas of the PDL at 36 h after OTM (Fig. 4b). After 4 weeks of chasing, the EdU-labeled cells completely disappeared at the distal side PDL of the distal root, whereas EdU-labeled cells remained present at the apical and furcation areas of the PDL.
In vivo cell clonal analysis was performed using RGBow: UBC-CreER T2 (RGBow) mice, where all cells randomly expressed three different fluorescence probes in a tamoxifen-inducible manner (Fig. 5a, b). OTM was initiated in 8-week-old RGBow mice, and tamoxifen was injected 3 days prior to the OTM. After 2 weeks of OTM and an additional 2 weeks of tracing, cluster formation was analyzed histologically (Fig. 5c). Single-cell-derived cell clones were visualized with three pseudo colors (red, green, or blue). The number of cells in each cluster increased significantly from 1.5 ± 0.9 to 2.2 ± 1.6 by OTM, confirming further division of the OTM-induced  (Fig. 5d). Consistent with the distribution of proliferating cells, we observed a nonspecific distribution of the single-cell-derived clusters within PDL. Notably, the cell clusters tended to grow along the collagen fiber direction of the PDL.

Discussion
Cell proliferation is an initial event of tissue remodeling and plays a key role in maintaining the cell supply.
Although accelerated cell proliferation under mechanical-loading conditions has been reported previously 18,21 , it remains unclear which cells acquire proliferative ability and from where these cells arise. Here, we showed that proliferating cells exponentially increased at 36 h after OTM. OTM-induced proliferating cells appeared at the entire PDL and did not have a geological correlation to the vasculature, a putative stem cell location. Moreover, single-cell-derived clusters in RGBow mice, representing daughter cells of the proliferated cells, did not exhibit a specific distribution pattern. Changes in cell cycle status in response to OTM were also detected at the entire PDL, as confirmed in Fucci2 mice. Furthermore, OTM-induced proliferating cells expressed the TAC marker Ring1b, as well as various other molecular markers, including Gli1, Nfic, Cbfa1, and Osx. These results, including the non-preferential distribution and expression of a wide range of molecular markers, indicate that OTM-induced proliferating cells do not arise from specific cell types and locations. Although Ring1b expression is strictly restricted to TACs in mouse incisor mesenchyme, it is widely distributed in the mouse molar PDL, and the ratio of Ring1b-expressing cells was unaltered in response to OTM. Furthermore, it is likely that the differentiation trajectory mechanism represented by the dynamic change in cell-proliferative activity (i.e., the transition from slow-cycling MSCs to rapid-cycling TACs) does not exist in mouse molar PDL under mechanical-loading conditions. Nfic is a member of the nuclear factor I family and functions as a critical regulator of root formation, with this supported by defective root formation observed in Nfic-knockout mice while crown formation remains intact 22 . Nfic interacts with TGF-β signaling and is also a downstream target of Gli1-mediated hedgehog signaling 23 . Additionally, Nfic reportedly promotes the proliferation and osteogenic/cementogenic differentiation of dental follicle cells 24 . In the present study, we identified Nfic expression in ~ 50% of the cells in mouse molar PDL, as well as some of the proliferating cells. According to previous reports, Nfic might also play roles in PDL homeostasis, as seen during tooth root/PDL development 22,23 .
Although the number of Gli1-and Cbfa1-expressing cells reportedly increases at 7 days 8 and Osx-expressing cells at 48 h of OTM 25 , we did not observe any changes in these markers at 36 h of OTM. In general, terminal differentiation usually coincides with proliferation arrest; thus, regulation of cell cycle and cell-type-specific transcription has a unique inverse relationship 26 . In accordance with this concept, we observed the appearance of proliferating cells and the disappearance of quiescent cells concurrently in Fucci2 mice. Furthermore, these data highlight the involvement of temporal coordination between cell cycle and terminal differentiation in PDL cells.
In the present study, we selected the distal PDL of the distal root as ROI since mechanical-loading-induced cell proliferation was consistently evoked in this region. Besides, this region has a typical ligament structure (i.e., rich in oriented collagen fibers and less vasculature) and load-bearing environment. However, recent studies indicated that the PDL is not uniform; different regions have distinct molecular and mechanical characteristics 27,28 . For instance, the PDL at the furcation area is sparse in collagen fiber, resulting in lower stiffness and a distinct protein composition compared to the collar region 27 . Moreover, the force distribution of the furcation area is different, as confirmed by finite element analysis 29 . Since the extracellular environment crucially affects cell behaviors, different regions of PDL may have different cell-differentiation trajectories.
To analyze the fate of OTM-induced proliferating cells in PDL, EdU-labeled proliferated cells were further traced for 4 weeks. The EdU-labeled cells completely disappeared at the distal PDL of the distal root, indicating that the cell division was not transient, and that these cells undergo further division and contribute to tissue remodeling. Cluster formation of single-cell-derived cells in RGBow mice also supported this observation. Remnants of the EdU-labeled cells at the apical and furcation area suggested that these cells did not divide further, most likely due to the distinct molecular and mechanical environment in this region, as discussed earlier.
It has long been believed that a small number of tissue stem cells are responsible for tissue remodeling. Although we detected a dramatic change in the proliferative status in response to mechanical-loading, the transition of slow-cycling MSCs to rapidly cycling TACs confirmed in mouse incisor mesenchyme 17 was not evident in mouse molar PDL during OTM. Conversely, the proliferative cells arise from various cell types at entire PDL tissue, this may suggest the differentiation plasticity of the PDL cells. A recent study showed that distraction www.nature.com/scientificreports/ osteogenesis induces the focal adhesion kinase signaling pathway and activates a gene-regulatory program usually activated in primitive neural crest cells, which is a major stem cell population during craniofacial development 30 . Similar to distraction osteogenesis, OTM represents mechanical-loading-induced tissue remodeling and PDL cells also originate from the neural crest 31 , the proliferative cells during OTM might be similar to the primitive neural crest cell and capable of giving rise to multiple dental mesenchymal cells. Although PDL cells act as fibroblasts to maintain collagen-rich fibrous tissue at the physiological state 32 , they might also exhibit lineage plasticity according to feedback signals from the surrounding environment 33 . In line with this notion, the multidifferentiation potential of PDL-derived cells into non-dental cells, including chondrocytes 34 , cardiac myocytes 35 , pancreatic-islet cells 36 , and retinal ganglion-like cells 37 , has been well documented in culture conditions. In summary, we analyzed proliferating cells in mouse molar PDL during mechanical-loading-induced tissue remodeling. The results demonstrated that the mechanical stimuli evoked a dynamic change in the cellproliferative profile at the entire PDL, and the proliferated cells contributed to tissue remodeling. Additionally, these findings indicated the lineage plasticity of PDL cells in vivo. Although further studies are required, our findings partially elucidate the global views of the cell-differentiation trajectory in mouse molar PDL under mechanical-loading conditions. Understanding the cellular dynamics is vital to maintain tissue homeostasis in a clinical setting and to develop innovative regenerative strategies in humans.   www.nature.com/scientificreports/ Animals. Wild-type mice (C57BL/6J, male) were purchased from Charles River Laboratories Japan (Yokohama, Japan). An R26p-Fucci2 mouse (CDB0203T) 19,20 was obtained from the RIKEN Center for Life Science Technologies. A UBC-CreER T2 mouse (JAX007001) 38 and RGBow mouse (JAX028583) 39 were obtained from Jackson Laboratory (Bar Harbor, ME, USA). All mice were maintained in the animal facility of Niigata University.
OTM. The Ni-Ti coil (25 gf; TOMY, Tokyo, Japan) was placed between the maxillary bilateral incisors and the left first molar of 8-week-old male mice with stainless steel wire (φ 0.09 mm). The steel wire tied to the incisors was reinforced with a light cure composite resin (BEAUTIFIL Flow; Shofu, Kyoto, Japan) to prevent breakage during mastication. After placing the Ni-Ti coil for a defined period of time (24, 36, and 72 h; 1 and 2 weeks), both sides of the maxilla were harvested and fixed with 10% paraformaldehyde for 24 h. The upper right first molar served as a control.
Nucleoside incorporation for proliferating cells. EdU (Invitrogen, Carlsbad, CA, USA) was intraperitoneally injected (5 mg/100 g body weight) into the wild-type or Fucci2 mice at 6 h and 12 h prior to euthanasia.

Micro-CT.
Micro-CT images of the maxilla were obtained using 90-kV and 88-μA irradiation from a highresolution X-ray tomographic system (CosmoScan GX; Rigaku Corporation, Tokyo, Japan). Structural analysis was performed using a software tool (Analyze 12.0; Analyze Direct, Inc., Overland Park, KS, USA). The outline of the teeth was defined with a threshold intensity between 6,000 and 12,000 Hounsfield units. The distance between the closest point of the first and second molars was measured.

Histology and immunohistochemistry.
To prepare the paraffin-embedded histology sections, samples were decalcified with 10% EDTA for 3 weeks at room temperature after image collection by micro-CT. Sections were stained with H&E for morphological analysis. EdU-incorporated proliferative cells were detected using the Click-iT EdU Cell Proliferation Kit for Imaging Alexa Fluor 488 dye (Invitrogen) according to manufacturer instructions. Following the detection of EdU-labeled cells, molecular markers were co-labeled by immunohistochemistry, as previously described 40  Cell clonal analysis. Three days before applying the coil, tamoxifen (T5648; Sigma-Aldrich, St. Louis, MO, USA) was injected intraperitoneally (75 mg/kg body weight) into RGBow: UBC-CreER T2 mice. After 14 days of OTM, the coil was removed, and mice were sacrificed after an additional 14 days of tracing. Paraffin-embedded histology sections were prepared, as described, and fluorescence proteins (EGFP; mOrange2 and mKate2) were detected with the following primary antibodies: EGFP (chicken anti-GFP antibody: 1:2,000; 13970; Abcam), mOrange2 (rat anti-mCherry antibody: 1:200; M11217; Invitrogen), and mKate2 (rabbit anti-tRFP antibody; 1:200; AB233; Evrogen, Moscow, Russia). The secondary antibodies used were as follows: goat anti-rat IgG antibody Alexa Fluor 488 (1:400; Invitrogen), goat anti-chicken IgY antibody Alexa Fluor 555 (1:400; Invitrogen), and goat anti-rabbit IgG antibody Alexa Fluor 633 (1:400; Invitrogen). Image capturing. A system microscope (BX53; Olympus, Tokyo, Japan) equipped with a digital camera (DP80; Olympus) was used to capture images. Fiji/ImageJ (NIH, Bethesda, MD, USA) and cellSens (Olympus) were used for image processing. Statistical analysis. The distal side of the PDL on the distal root of the first maxillary molar was selected as the observation area. The number of EdU + cells, immuno-positive cells, and total cells within the observation area were counted, and the positive cell ratio was calculated. Cell counting was performed on at least three different samples. For statistical analyses, an unpaired t-test with Welch's correction was performed using GraphPad Prism (GraphPad Software, San Diego, CA, USA), and a p < 0.05 was considered significant.

Data availability
The data that support the findings of this study are available within the article. Additional data may be available from the corresponding author, [M.K.], upon reasonable request.