Dentoalveolar Alterations in an Adenine-Induced Chronic Kidney Disease Mouse Model

Chronic kidney disease (CKD) is characterized by kidney damage and loss of renal function. CKD mineral and bone disorder (CKD – MBD) describes the dysregulation of mineral homeostasis, including hyperphosphatemia and elevated parathyroid hormone (PTH) secretion, skeletal abnormalities, and vascular calci ﬁ cation. CKD – MBD impacts the oral cavity, with effects including salivary gland dysfunction, enamel hypoplasia and damage, increased dentin formation, decreased pulp volume, pulp calci ﬁ cations, and altered jaw bones, contributing to clinical manifestations of periodontal disease and tooth loss. Underlying mechanisms are not fully understood, and CKD mouse models commonly require invasive procedures with high rates of infection and mortality. We aimed to characterize the dentoalveolar effects of an adenine diet (AD)-induced CKD (AD – CKD) mouse model. Eight-week-old C57BL/6J mice were provided either a normal phosphorus diet control (CTR) or adenine and high-phosphorus diet CKD to induce kidney failure. Mice were euthanized at 15 weeks old, and mandibles were collected for micro – computed tomography and histology. CKD mice exhibited kidney failure, hyperphosphatemia, and hyperparathyroidism in association with porous cortical bone in femurs. CKD mice showed a 30% decrease in molar enamel volume compared to CTR mice. Enamel wear was associated with reduced ductal components, ectopic calci ﬁ cations, and altered osteopontin (OPN) deposition in submandibular salivary glands of CKD mice. Molar cusps in CKD mice were ﬂ attened, exposing dentin. Molar dentin/cementum volume increased 7% in CKD mice and pulp volume decreased. Histology revealed excessive reactionary dentin and altered pulp-dentin extracellular matrix proteins, including increased OPN. Mandibular bone volume fraction decreased 12% and bone mineral density decreased 9% in CKD versus CTR mice. Alveolar bone in CKD mice exhibited increased tissue-nonspeci ﬁ c alkaline phosphatase localization, OPN deposition, and greater osteoclast numbers. AD – CKD recapitulated key aspects reported in CKD patients and revealed new insights into CKD-associated oral defects. This model has potential for studying mechanisms of dentoalveolar defects or therapeutic interventions. © 2023 The Authors. Journal of Bone and Mineral Research published by Wiley Periodicals LLC on behalf of American Society for Bone and Mineral Research (ASBMR).


Introduction
C hronic kidney disease (CKD) is a global public health problem, with an estimated prevalence of over 13% in the USA, where there are more than 700,000 individuals with end-stage kidney disease requiring kidney transplant. [1,2] CKD is characterized by kidney damage and progressive loss of renal function skeletal abnormalities, and vascular calcification. Altered mineral metabolism features hyperphosphatemia, decreased calcium levels, dysregulated parathyroid hormone (PTH) secretion, and altered vitamin D signaling. CKD-MBD contributes to renal osteodystrophy (ROD), collected skeletal disturbances that can result in osteitis fibrosa (a condition characterized by high bone turnover with increased osteoclast and osteoblast activity and elevated PTH) or adynamic bone disease (with low turnover and reduced circulating PTH [3] ).
Mouse models of CKD have provided important insights into pathological mechanisms and make it possible to test therapeutic interventions. To date, studies have employed several approaches to inducing renal failure. [24,25] The 5/6 nephrectomy model (5/6Nx) and electrocautery (EC) injury models are limited by the requirement for two invasive surgeries that pose increased risk of injury and mortality, alterations related to surgical ablation rather than loss of renal function, nonreversibility of the condition, and a relatively large degree of variability and sometimes require the inclusion of an inducing agent (e.g., 1,25 (OH) 2 D 3 ) or use of gene-edited mouse models to achieve biochemical changes and vascular calcification. Using an EC-5/6NxCKD model combined with high phosphate diet, we previously documented decreased bone volume fraction (BV/TV) and disordered bone architecture in mouse mandibles. [26] A more recently introduced adenine diet (AD)-induced CKD (AD-CKD) mouse model was adapted from use in rats and has several advantages that include no requirement for surgeries or postoperative care, kidney damage more similar to human renal failure, consistent and severe changes in mineral metabolism, and potential for partial reversibility. [27,28] AD-CKD in mice replicated aspects of CKD-MBD and ROD observed in human subjects and other mouse models. [28][29][30] Despite the many gaps in our knowledge of how CKD impacts oral health, little research has been done on oral effects using mouse models of CKD.
To date, there have been no reports on whether AD-CKD in mice parallels 5/6Nx, EC, or genetic models of renal failure or replicates effects on oral tissues similar to reports in human patients. We hypothesized that AD-CKD could recapitulate key aspects of renal failure, CKD-MBD, and dentoalveolar defects observed in adult-onset CKD. We aimed to test this hypothesis and analyze the dentoalveolar effects of an AD-CKD mouse model using high-resolution micro-computed tomography (μCT), histology, and immunohistochemistry (IHC) and evaluate changes in comparison to the EC-5/6Nx model and reports from human CKD case reports and studies.

Materials and Methods
Mice Animal studies were approved by the Institutional Animal Care and Use Committee (IACUC; Sanford Burnham Prebys Medical Discovery Institute, La Jolla, CA, USA) and conducted according to Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines. Mice were maintained according to animal facility breeding standard operating procedures. The animal room had a 12-h light/dark cycle, and it was kept clean and pathogen-free. Mice were housed in groups of five maximum. Renal damage and associated CKD was induced in C57BL/6J mice at 8 weeks of age. Mice in the control (CTR) group were maintained on a normal phosphorus diet (0.7%) throughout the study (Teklad Global Diet 2018, Envigo, Indianapolis, IN, USA). The mice in the CKD group were administered a 0.2% adenine diet for 3 weeks (Teklad Custom Diet TD.09138, Envigo), which was additionally supplemented with a high (1.8%) phosphorus diet (Teklad Custom Diet TD.190831, Envigo) for 4 more weeks to induce kidney damage, uremia, and hyperphosphatemia (Fig. 1A). [31] Males and females were analyzed together in this study due to severe and similar effects of CKD on dentoalveolar tissues of both sexes. Mice were euthanized at 15 weeks of age, at which time blood was collected (n = 6 CTR and CKD), and femurs (n = 7 CTR, n = 4 CKD), mandibles (n = 5 CTR, n = 4 CKD), and submandibular salivary glands (n = 3 CTR, n = 2 CKD; all male due to sex differences in submandibular glands of mice) were harvested and fixed in 4% paraformaldehyde phosphate buffered saline (PBS) solution. Mice that experienced severe or chronic pain or distress that could not be relieved were painlessly euthanized as deemed appropriate by the veterinary staff or researchers; therefore, some CKD mice were harvested prior to the predetermined study endpoint.

Blood biochemistry
Blood urea nitrogen (BUN), serum phosphorus, and plasma PTH were performed to confirm renal damage and the establishment of the CKD condition (n = 6 CTR mice and n = 6 CKD mice, including three males and three females per group). BUN was measured from fresh whole blood using Abaxis rotors on a VetScan VS2 chemistry analyzer (Zoetis, Parsippany, NJ, USA). Phosphorus was measured using the Stanbio Phosphorus Liqui-UV kit (Stanbio Laboratory, EKF Diagnostics, Boerne, TX, USA). Plasma PTH was quantified by the RayBio PTH Enzyme Immunoassay (EIA) kit (RayBiotech, Inc., Peachtree Corners, GA, USA) according to the manufacturer's instructions.

Micro-computed tomography (μCT)
Femurs were scanned in a SkyScan 1172 scanner (Bruker, Ettlingen, Germany) at 55 kV, 181 μA, 0.5-mm Al filter, 280-ms integration time, and 10 μm voxel dimension. Hemimandibles were scanned in a μCT 50 scanner (Scanco Medical, Bassersdorf, Switzerland) at 70 kV, 76 μA, 0.5-mm Al filter, 900-ms integration time, and 6-μm voxel dimension. Reconstructed images were calibrated to five known densities of hydroxyapatite and analyzed using AnalyzePro version 1.0 (AnalyzeDirect, Overland Park, KS, USA). Femurs were analyzed according to standard trabecular and cortical bone analysis using a bone threshold of 400 and 550 mg HA/cm 3 , respectively. [32] The first mandibular molar and associated alveolar bone were quantitatively analyzed as previously described. [33,34] The alveolar bone region of interest (ROI) included the area between 240 μm mesial to the most mesial point of the first molar mesial root and 240 μm distal to the most distal point of the distal root. Enamel was segmented above 1,600 mg HA/cm 3 , while dentin/cementum and alveolar bone were segmented at 550-1,600 mg HA/cm 3 .

Histology
Hemimandibles were decalcified in an acetic acid/formalin/ sodium chloride solution, processed for paraffin embedding, and sectioned at 5-μm thickness in the coronal plane. Paraffin sections were stained with hematoxylin and eosin (H&E) and picrosirius red (viewed under polarized light microscopy) to assess the tooth and associated periodontium. [35] Submandibular salivary glands were processed for paraffin embedding, sectioned at 5-μm thickness, and stained by H&E, Masson's trichrome, Alcian blue, and von Kossa stains (described subsequently). Measurement of acellular cementum thickness was performed on H&E-stained coronal sections in the central portion of the mesial root of the first mandibular molar by measuring 300 μm apically from the cementum-enamel junction on the buccal aspect of the root. Measurement of cellular cementum area was performed on H&E-stained coronal sections in the central portion of the mesial root of the first mandibular molar by measuring areas of buccal and lingual aspects of cementum. Measurements were performed using ImageJ 1.53e (National Institutes of Health, Bethesda, MD, USA). Immunohistochemistry (IHC) procedures were performed as described previously. [33,36,37] Primary antibodies included polyclonal rabbit anti-mouse bone sialoprotein (BSP) immunoglobulin G (IgG) (Dr. Renny Franceschi, University of Michigan, Ann Arbor, MI, USA) [38,39] ; polyclonal rabbit anti-mouse osteopontin  [40] ; polyclonal rabbit anti-rat dentin matrix protein 1 (DMP1) IgG (M176; Takara Bio USA Inc., San Jose, CA, USA) [35] ; polyclonal rabbit anti-mouse dentin sialoprotein (DSP) IgG (LF-153; Dr. Larry Fisher) [41] ; and monoclonal rat antitissue nonspecific alkaline phosphatase (TNAP) IgG (R&D Systems, Minneapolis, MN, USA).
Tartrate-resistant acid phosphatase (TRACP) staining was used to identify and quantify osteoclast-like cells. [36,37] TRACP-positive (TRACP+) osteoclasts were enumerated on alveolar bone surfaces and bone interiors/marrow space and normalized to bone perimeter and area, respectively, using ImageJ 1.53e.
Masson's trichrome staining was performed on submandibular salivary glands according to the manufacturer's instructions (Electron Microscopy Sciences, Hatfield, PA, USA) with some modifications. Briefly, after deparaffinization, sample sections were stained for 15 min in Weigert's iron hematoxylin, then rinsed in tap water for 5 min and then in distilled water. Sections were incubated with Biebrich scarlet for 30 s and phosphotungstic/phosphmolybdic acid for 15 min and transferred directly into aniline blue staining for 15 min. For Masson's trichome staining, blue color indicates collagen, red color indicates cytoplasm, and black stains nuclei. Alcian blue staining was performed to selectively distinguish mucous acini with a blue-green color based on their high content of mucopolysaccharides. [42] Alcian bluestained histology sections were quantified by ImageJ. Four images captured at 20Â magnification per sample were converted to 8-bit type, a maximum threshold of 200 was applied to distinguish positively stained (mucous) acini, and percent area fraction of mucous acini was calculated and averaged for each sample. Alcian blue was used to quantify ductal cell components by freehand selection tracing of 10 round ducts per sample (avoiding elongated ducts that were sectioned in a different orientation) and calculating the average area of individual ducts. Quantification of the ductal lumina was accomplished by freehand selection tracing of 20 duct lumina per sample (avoiding elongated ducts that were sectioned in a different orientation) and calculating the average area of the lumen space per duct. Von Kossa staining was performed on deparaffinized submandibular salivary gland sections. Tissue sections were incubated in 1.5% freshly made silver nitrate solution (Thermo Fisher Scientific, Waltham, MA, USA) in the dark for 20 min at room temperature, rinsed in distilled water, and then incubated in 2% hydroquinone for 2 min. After washing in distilled water, sections were incubated in 5% sodium thiosulfate (Thermo Fisher Scientific) for 5 min and counterstained with nuclear fast red staining (Electron Microscopy Sciences) for 3 min. For von Kossa staining, black stain indicates calcification while the counterstain is red.

Statistical analyzes
Statistical analyses were performed using GraphPad Prism version 9.1.1 (GraphPad Software, San Diego, CA, USA). Data in graphs are expressed as box plots showing all individual data points, median, and upper and lower quartiles. Statistical analysis was performed by unpaired t-test to analyze CTR versus CKD mice. Statistical significance was determined by p < 0.05, and all p-values are shown in graphs.

Adenine and high phosphate diet induced CKD-MBD in mice
The AD-CKD model used is summarized in Fig. 1A. Induction of renal damage was initiated in mice at 8 weeks of age, equivalent to young adulthood in humans. Mouse molars have completed development and are in functional occlusion by this age, while the incisor is a continually erupting tooth that features all stages of odontogenesis at any given age. The control (CTR) group mice (n = 7) were maintained on a normal phosphorus diet (0.7%) throughout the study, whereas the mice in the CKD group (n = 6) were administered a 0.2% adenine diet for 3 weeks, which was additionally supplemented with high (1.8%) phosphorus diet for 4 more weeks. Mice were euthanized and tissues and blood harvested at 15 weeks of age.
Previous reports of AD-CKD in mice confirmed reduced renal function (i.e., increased BUN and increased serum creatinine levels), hyperphosphatemia, increased PTH, and increased FGF23. [28,30] Blood biochemistry was tested to confirm renal damage and onset of CKD in our study. BUN was increased in CKD versus CTR about threefold ( p < 0.0001) (Fig. 1B). Serum phosphorus levels were increased in CKD versus CTR mice more than fourfold (p < 0.0001) (Fig. 1C). Circulating PTH levels were increased in CKD versus CTR mice more than 11-fold ( p < 0.0001) (Fig. 1D). μCT analysis of trabecular and cortical bone parameters in femurs revealed substantial skeletal changes in CKD versus CTR mice, manifesting as thinner and more porous cortical bone (Fig. 1E,F). Quantification confirmed reduced trabecular thickness (Tb.Th; p = 0.0220) and a wider array of effects on cortical bone including reduced cortical bone volume fraction (BV/TV; p = 0.0004), reduced cortical thickness (Ct.Th;   Enamel damage in mice with CKD High-resolution μCT was performed to evaluate the effects of CKD on mineralized dentoalveolar tissues. Three alterations were apparent by observation of 3D and 2D μCT images ( Fig. 2A-D): (1) damage to molar enamel in CKD versus CTR mice; (2) increased dentin associated with reduction of the dental pulp space and pulp calcification in CKD versus CTR mice; and (3) periodontal changes, including altered alveolar bone organization, regions of tooth-bone ankylosis, and alveolar bone defects around the first molar. These three alterations are analyzed in more detail in what follows.
Three-dimensional μCT images revealed large regions of enamel missing from first molar crowns, focused on lingual aspects (Fig. 2E,F). μCT analyses confirmed an approximately 30% reduction in enamel volume in CKD versus CTR molars (p = 0.0013), with no differences in enamel mineral density (Fig. 2G,H). The mouse incisor is a continuously erupting tooth that features all stages of tooth development at any given time, which provides a model to test effects on tooth development processes, including enamel formation. The unerupted region of the mandibular incisor beneath the first molar was analyzed (Fig. 2I,J), and no changes were revealed in the incisor enamel volume or density in CKD versus CTR mice (p = 0.6125 and 0.7812, respectively) (Fig. 2K,L).
Salivary glands were investigated based on reported salivary gland dysfunction and hyposalivation resulting from CKD. [21,23,43,44] Harvested submandibular salivary glands were smaller and discolored in CKD versus CTR mice (data not shown). H&E and Masson's trichrome staining revealed acini and ductal cells in CTR salivary glands and notably reduced ductal component sizes in CKD mice (Fig. 3A-H). Alcian blue staining revealed trends toward altered acinar composition (increased blue-green staining localizing to acini suggestive of mucous acini) in CKD versus CTR and quantification of blue-green stained regions confirmed decreased sizes of the ductal components and lumina in CKD versus CTR salivary glands ( p = 0.0038 and p = 0.0183, respectively) ( Fig. 3I-L,U). Von Kossa staining of salivary glands revealed numerous and dispersed microcalcifications in CKD mice ( Fig. 3M-P). In salivary glands of CTR mice, OPN staining was intensely localized to acini with no or little staining in ducts, while in CKD mice, OPN localized heavily throughout all components, including ducts (Fig. 3Q-T).
Dentin-pulp disruptions in mice with CKD Two-dimensional μCT images revealed reduced dental pulp space and evidence of pulp calcifications in the molar (Fig. 4A,B). μCT analyses confirmed a 7% increase in dentin/ cementum volume in CKD versus CTR mice (p = 0.0129), with no difference between groups in mineral density (Fig. 4C,D). Conversely, pulp volume decreased by more than 30% in CKD versus CTR mice (p = 0.0320) (Fig. 4E).   By histology, increased tertiary reactionary dentin was apparent in molars of CKD versus CTR mice, particularly localized to pulp horns and significantly ablating the pulp space (Fig. 4F-I). Some tertiary dentin was also noted in CTR mice but comprised a much smaller pulp area. IHC was employed to investigate the extracellular matrix (ECM) composition in dentin and pulp. DMP1 and DSP showed abnormal localizations in the reactionary dentin in CKD versus CTR mouse molars (Fig. 4J-M). While CTR dentin did not have any detectable BSP, reactionary dentin in CKD featured increased BSP in odontoblasts (and possibly subodontoblast cell layers) adjacent to the increased and disorganized reactionary dentin (Fig. 4N,O). While OPN was widely localized in the dentin-pulp complex, including intense staining in reactionary crown dentin, CKD mouse molars also featured more intense OPN accumulation in dental pulp compared to the CTR group (Fig. 4P,Q).

Periodontal alterations in mice with CKD
While 3D μCT images did not indicate appreciable loss of alveolar bone crest height, 2D images revealed dramatically altered alveolar bone structure, evidence of subcrestal alveolar bone loss, and erratic alveolar bone-PDL surfaces in some locations (Fig. 2B,D). Bone volume fraction (BV/TV) of alveolar bone decreased by more than 10% in CKD versus CTR mice (p = 0.0025) (Fig. 6A). Additionally, alveolar bone mineral density was reduced by approximately 10% in CKD versus CTR mice (p = 0.0007) (Fig. 6B). Alveolar bone proper (ABP), the bone that interfaces with PDL and is most involved with tooth attachment, showed significant decreases in both volume and mineral density in CKD versus CTR mice (p = 0.0136 and p = 0.0023, respectively) (Fig. 6C,D). However, PDL volume was not different between groups (Fig. 6E).
Histology of periodontal tissues revealed severely disorganized alveolar bone with dramatically increased numbers and sizes of marrow spaces. Regions in the PDL space were lacking that appeared to be approaching ankylosis between alveolar bone and cellular cementum (Fig. 6L-O). These tended to be in the apical region of the tooth adjacent to cellular cementum. Picrosirius red stain observed under polarized light showed normal attachment and organization of PDL in the cervical root regions of CKD mice; however, staining confirmed apical PDL regions of some CKD mice displayed loss of collagen organization between bone and root ( Fig. 6P-S).
IHC showed osteoblast TNAP expression on alveolar bone surfaces in CTR mice; in contrast, CKD mice exhibited generalized elevated expression and specific regions of heightened TNAP localization (Fig. 7A-D). Increased TNAP localization in CKD was typically observed near regions of disrupted bone remodeling, either on bone surfaces or in the marrow and fibrous spaces within. Immunostaining indicated altered distribution of OPN in alveolar bone of CKD versus CTR mice, where OPN was increased and highly concentrated in some disorganized regions of bone (Fig. 7E-H).
TRACP staining allowed enumeration of osteoclasts in periodontal tissues (Fig. 7I-L). CKD mice exhibited greater than threefold increased numbers of total osteoclasts and osteoclasts normalized to total alveolar bone area measured (p = 0.0246 and p = 0.0318, respectively) (Fig. 7M,N). When osteoclast density was separated by area, it became evident that the cell density of osteoclasts along the alveolar bone-PDL interface was unchanged between groups (Fig. 7O), while there was an eightfold increased density of osteoclasts within alveolar bone and its marrow spaces in CKD versus CTR groups (p = 0.0042) (Fig. 7P).

Discussion
We tested an AD-CKD mouse model to determine whether this would recapitulate the primary aspects of renal failure, CKD-MBD, and oral and dentoalveolar effects. Blood biochemistry reflected altered mineral metabolism associated with renal damage, and analysis of femurs confirmed changes consistent with CKD-MBD and ROD, particularly thinning and increased porosity of cortical bone. Dentoalveolar changes in mice of the AD-CKD group included damage to enamel associated with dystrophic changes in salivary glands, increased dentin and decreased dental pulp volume with a tendency for pulp calcification, altered distribution of ECM markers in reactionary dentin and pulp, and porous bone with increased TNAP, rich in OPN, and harboring increased numbers of osteoclasts. These collective defects parallel reports of CKD-associated oral changes in humans. Therefore, the AD-CKD model represents a potentially valuable model of CKD-MBD and ROD for studying mechanisms underlying defects and testing therapeutic interventions.

Comparison of adenine-induced CKD-MBD with other mouse models of renal failure
We document in the mice biochemical changes and skeletal effects of AD-CKD consistent with CKD-MBD, ROD, and osteitis fibrosa, primarily manifesting as hyperphosphatemia, increased PTH, cortical bone thinning, and increased cortical bone porosity. Similar biochemical and skeletal changes have been reported in other CKD mouse models, including 5/6 nephrectomy model (5/6Nx), EC injury models, dietary induction/exacerbation, and transgenic models, as well as AD-CKD rat models. [28,[45][46][47][48][49][50] The changes from AD-CKD were consistently achieved with less than 2 months of treatment; conversely, 5/6Nx and EC models typically require longer periods of time and invasive surgeries that are accompanied by several limitations, including an increased risk of injury and mortality, alterations related to surgical ablation rather than loss of renal function, nonreversibility of the condition, a relatively large degree of variability, and, in some cases, the necessary use of an inducing agent (e.g., 1,25(OH) 2 D 3 ) and/or use of gene-edited mouse models to achieve biochemical changes and vascular calcification. [24,25] As concluded in previous publications with more comprehensive analyses of the biochemical, cardiovascular, and skeletal effects of the model, the AD-CKD model described here meets the criteria as a valid approach to inducing and studying CKD-MBD and ROD. [24,28,29,47] Dentoalveolar effects in a mouse model of AD-CKD Up to 90% of individuals with CKD present with oral abnormalities caused by CKD-MBD. [4] These include skeletal alterations (e.g., radiologic changes, expansion of maxilla and/or mandible, thinning of cortical bone, increased/altered trabecular bone, loss of lamina dura, and radiopaque lesions), hyposalivation and xerostomia, gingival hyperplasia, enamel wear, abnormal dentin accumulation, reduced dental pulp volume, widened PDL, and hypercementosis. [4][5][6][7][8][9]23] Dentin-pulp changes can lead to pulp obliteration and potential effects on tooth vitality, while periodontal changes can contribute to periodontal disease, tooth mobility, malocclusion, and tooth loss. Early-onset forms of CKD due to genetic or environmental disruptions of renal function can disrupt enamel formation in primary and/or secondary dentition (depending on age of onset and severity of renal failure), resulting in enamel hypoplasia and discoloration that have a negative impact on esthetics and quality of life. [16][17][18][19][20][21][22][23] The effects of CKD-associated enamel hypoplasia on dental caries are not entirely clear due to mixed results from case reports; notably, several case studies report reduced prevalence of caries in children with CKD compared to healthy controls. [16][17][18][19][20][21][22][23] While many studies have examined the effects of renal failure and CKD-MBD on ROD in the postcranial skeleton in mice, very few reports have analyzed the associated dentoalveolar defects. In the EC-5/6Nx model of CKD we previously described, 18-weekold dilute brown agouti/2 (DBA/2) mice underwent dietary alteration and two renal surgeries over the course of 14-15 weeks. [26] Quantitative analysis of mandibular bone identified reduced cortical thickness and increased trabecular number, thickness, and bone volume fraction (BV/TV) in CKD mice on a high phosphate diet. Histological observations included widened PDL, irregular alveolar bone surface, disrupted alveolar bone patterning, and increased bone marrow. In another EC-5/6Nx model of CKD, the effects of ovariectomy (OVX) on mandibular bone loss were analyzed in 28-week-old C57BL/6 mice. [51] CKD induced decreased cortical thickness and BV/TV, with OVX exacerbating these effects of ROD on mandibles. Histological observations included increased PDL and bone marrow spaces. Other studies have examined additional relationships between oral and renal health, albeit in indirect ways and without detailed analyses of dentoalveolar tissues. Oral swabbing of periodontal pathogen Porphyromonas gingivalis was found to provoke alveolar bone loss and exacerbate kidney injury in hypertensive Nos3 knockout mice. [52] Investigation of a rat model with autosomal dominant inherited polycystic kidney disease (Cy/+) found calcium gluconate and/or zoledronate could partially ameliorate alveolar bone loss associated with CKD. [53] The AD-CKD mouse model exhibited severe defects in multiple oral and dentoalveolar tissues. We document dramatic damage to enamel in the molar teeth of mice in the CKD group that is consistent with reports of damage to enamel in adults with CKD. [10,13,15,54] To our knowledge, this is the first such report of enamel damage arising from CKD in mice. The pathogenesis of enamel damage in individuals with CKD is unknown. There has been speculation about regurgitation and acid damage to teeth, suggesting erosion of enamel. Mice and rats lack the regurgitation response, but gastroesophageal reflux can be elicited by overeating, surgical, or gene-editing approaches. [55,56] CKD can cause salivary gland dysfunction, reduced salivary flow, and xerostomia in human patients as well as in CKD mouse models where salivary gland metabolism has been investigated. [14,21,23,43,44] This suggests that changes in salivary composition, pH, or reduced saliva output directly or indirectly (as in allowing accumulation and damage from cariogenic bacteria) promote enamel erosion. Salivary flow is understood to be protective against enamel erosion in humans. [57][58][59][60] Salivary gland hypofunction was shown to cause enamel erosion in several mouse and rat models; similar observations of enamel absence or thinning, particularly on the lingual aspect, have been reported. [61] Studies of the effects of diet on dental tissues in mice and rats also showed severe enamel erosion patterns when acids overwhelmed the capacity for salivary flow. [62][63][64] The connection between enamel wear and salivary glands in this study is supported by observations of reduced submandibular salivary gland size, reduced duct and lumina sizes, altered acinar composition, ectopic calcification, and dystrophic OPN distribution in AD-CKD mice. Damage, calcification, and increased OPN localization have been reported in kidneys of mice with CKD. [65] Parallel changes in salivary glands in the AD-CKD model may point to similar pathological mechanisms, and in particular the question of whether OPN plays a protective role against salivary gland calcification should be investigated. Reports have linked oxidative stress arising from severe CKD with salivary gland damage, reduced salivary protein content and amylase activity, hyposalivation, and xerostomia. [43,44] Some have speculated that pharmacotherapy for CKD may contribute to salivary gland dysfunction, [43] but results in AD-CKD mouse model suggest that dramatic changes result directly from the onset of CKD.
CKD has also been linked to developmental enamel defects in primary and secondary teeth. [17][18][19][20][21]66] Children with early-onset CKD were found to have increased debris, calculus, and incidence of developmental defects of enamel but lower incidence and severity of caries in several studies. [11,12,[17][18][19][20][21] The timing of induction of kidney damage is not critical when evaluating the mouse incisor, as this continuously erupting tooth presents all stages of odontogenesis at any given time. In this study, incisor enamel showed no apparent developmental defects from the induction of CKD, indicating that developmental effects on enamel were not recapitulated by the AD-CKD mouse model.
Dentin was responsive to the induction of CKD, showing increased dentin volume and reduced dental pulp space, consistent with reports from CKD patients. [8,9] Incisors showed trends similar to those of molars, pointing to strong and consistent effects of CKD on the pulp-dentin complex regardless of developmental stage. Some mouse molars in the CKD group exhibited frank and extensive pulp calcification, indicating a predisposition to ectopic calcification of soft tissues associated with CKD-MBD. Calcification of the pulp is likely the result of hyperphosphatemia. Increased dentin, reduced dental pulp volume, and ectopic calcifications in pulp have been observed in other conditions that increase circulating phosphate levels, for example, hyperphosphatemic familial tumoral calcinosis (HFTC) caused by mutations in FGF23, GALNT3, KLOTHO, or FGF23 autoantibodies [67,68] ; hyperphosphatemia arising from genetic ablation of Fgf23 in mice [69] ; and pseudohypoparathyroidism caused by mutations in GNAS, STX16, or GNASAS1. [70] In a novel insight revealed by μCT and histology, mice in the CKD group showed increased production of reactionary dentin. Primary dentin describes the rapidly formed dentin during tooth formation, and secondary dentin includes the slowly growing dentin following completion of odontogenesis. [71] Tertiary dentin arises as a local reaction to an insult, including bacterial, thermal, chemical, or mechanical varieties. Tertiary dentin arising from milder stimuli and produced by preexisting odontoblasts is reactionary dentin, while more traumatic stimuli can cause cell death and require new odontoblasts that produce reparative dentin. Tertiary dentin often exhibits an altered structure and composition compared with primary and secondary varieties. In the CKD mice, excessive dentin in the pulp horns of the crown appeared to be reactionary dentin, likely resulting, in part, from loss of enamel and traumatic occlusion (thus, a mechanical insult).
The reactionary dentin in the CKD group showed abnormal localization of several ECM proteins. DSP is a major regulator of dentin mineralization and is richly present in the dentin ECM. [71,72] Reactionary dentin in CKD mice showed disrupted and reduced DSP localization, similar to a previous report where lesions were created in rat molars. [73] DMP1 is a dentin and bone ECM protein and player in systemic mineral metabolism via FGF23-PTH-vitamin D regulation of systemic phosphate. [71,74] Altered bone DMP1 expression has been linked to CKD-MBD. [75] In a previous study, reactionary dentin showed reduced DMP1, [73] contrary to what we observed in mice with CKD. BSP is expressed in bone and cementum and not thought to play a major role in dentinogenesis. [38] Dramatically increased BSP expression was specifically localized to only reactionary dentin and may provide insights into cell lineage differentiation in reactionary dentinogenesis, which often results in a disorganized, more bonelike dentin ("osteodentin") matrix lacking organized dentinal tubules. [73,76] This is consistent with a report that dentin in CKD presented numerous ultrastructural abnormalities. [77] As outlined earlier, OPN is a regulator of mineralization found in bones and teeth, but also with a wider distribution in soft tissues. [78,79] OPN was present in normal and abnormal dentin in CKD, but also increased in the dental pulp, in parallel to changes in salivary glands, which also show ectopic calcification like some pulps.
Gradual narrowing of the dental pulp and increased prevalence of pulp stones or diffuse calcification in the pulp is associated with increasing age. [80] However, more extreme pulp obliteration or extended ectopic calcification is the pathological result of trauma or genetic conditions that predispose to loss of dentin-pulp regulation, such as dentinogenesis imperfecta. [81] Pulp calcification or obliteration can be asymptomatic; however, some cases may lead to loss of tooth vitality and pulp necrosis. [82] Teeth with substantial pulp calcification can be difficult or impossible to treat with endodontics. It is unclear whether altered pulp-dentin distribution of ECM proteins in CKD versus CTR was associated with altered mineral metabolism of CKD-MBD, increased insult to dentin (due to damage to enamel) spurring increased tertiary dentinogenesis, or some combination of the two. The effects of CKD on dentinogenesis deserve additional attention to determine direct and indirect effects and how these contribute to changes in oral health and dental function. These will be important considerations for dental treatment of the CKD patient population.
Alveolar bone changes in this model of AD-CKD meet or exceed those described in previous studies, including disorganization and increased porosity mirroring the effects in cortical bone of the femur, increased TNAP, increased and abnormal deposition of OPN, and many-fold increased numbers of osteoclasts. Substantially reduced alveolar bone density may also indicate mineralization defects. A recent study linked metabolic acidosis with reduced bone density and increased bone disorganization in CKD patients. [83] Increased OPN has been documented in the kidney as a response to hyperphosphatemia and is thought to be secreted to maintain urinary phosphate solubility. [65] Increased OPN expression has been associated with worsening or more severe stages of CKD in mice and humans, [84][85][86] and polymorphisms are linked to cardiovascular events in CKD patients. [87] Furthermore, OPN is suspected to act as an inhibitor of ectopic calcification in CKD-associated atherosclerosis and other pathological conditions. [78] As phosphate is a potent inducer of increased Spp1/OPN expression in skeletal and dental cells, it is likely that increased OPN in mineralized tissues is partially an effect of hyperphosphatemia. [88][89][90][91][92] The role of increased OPN in CKD-MBD is not well understood. Notably, OPN is a mineralization inhibitor that exerts effects on bone and dentin, and increased OPN has been associated with hypomineralization defects in other pathological conditions, including hypophosphatemia and hypophosphatasia. [40,79,[93][94][95][96][97][98] OPN, in part via its arginine-glycine-aspartic acid (RGD) integrinbinding or other signaling domains, also contributes to osteoclast migration, adhesion, and/or function. [99][100][101] Thus, increased OPN could contribute to both defective mineralization and increased osteoclast resorption in CKD-MBD. The roles of OPN in CKD-MBD and ROD require further study, and this model of AD-CKD appears to be appropriate for this purpose.
Ankylosis refers to the loss of PDL space, leading to a pathological connection between tooth (usually cementum) and alveolar bone. Ankylosis in erupted teeth is associated with negative impacts, including malocclusion and root resorption, sometimes contributing to tooth loss. [102] Frank ankylosis was noted in some μCT scans of mice in the CKD group. Histology indicated regions where PDL space was reduced, organized collagen fibers appeared reduced or absent, and an osteoid-like material took the place of the PDL. These regions are likely approaching an ankylotic union. Ankylosis can result from changes in cementum and/or bone. While hypercementosis was described in previous CKD case reports in humans, neither acellular nor cellular cementum was apparently affected in the AD-CKD model. Though cementum is sensitive to disturbances in phosphate metabolism (e.g., X-linked hypophosphatemia), [103][104][105][106][107] regulators of inorganic pyrophosphate (e.g., progressive ankylosis protein [ANK] and ectonucleotide pyrophosphatase phosphodiesterase 1 [ENPP1]) exert control over cementum apposition and represent one of several lines of defense against pathological cementum expansion. [108][109][110] These collected observations point to alveolar bone as the periodontal tissue primarily targeted by CKD-MBD. We did note that increased bone resorption and formation, hyperphosphatemia, and altered ECM composition may all contribute to regions of ankylosis associated with CKD, though the respective contributions and sequence of events require further study.

Conclusions
The AD-induced CKD mouse model holds many advantages over other approaches to inducing renal failure, including no required surgeries, more rapid disease onset, and consistent and severe changes in mineral metabolism. We confirm in this AD-CKD mouse model rapid and dramatic dentoalveolar changes, including loss of enamel, increased dentin and reduced dental pulp, and severe alveolar bone dysregulation. These alterations recapitulate a range of hard tissue changes in individuals with CKD.
This AD-CKD mouse model therefore represents a promising model for additional investigations of pathological mechanisms and potential interventions.

Acknowledgments
We thank Buddy Charbono, Diana Sandoval, and Andy Vasquez in the Animal Facility at Sanford Burnham Prebys Medical Discovery Institute for assistance with animal care. We thank Dr. Isabelle Lombaert at the University of Michigan School of Dentistry for feedback on salivary gland analysis.

Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.

Funding Information
This work was funded by Grant R01DE12889 from the National Institute of Dental and Craniofacial Research (NIDCR) of the National Institutes of Health (NIH) to JLM, Grants R03DE028411, R01DE027639, and R01DE032334 from NIDCR/NIH to BLF, Grant R00DE031148 to EYC, research grants from Soft Bones to FAO and FFM, and a grant from the Endocrine Fellows Foundation to FAO.