Targeted Alkaline Phosphatase Therapy Enhances Alveolar Bone Healing in X‐Linked Hypophosphatemia in Mice

ABSTRACT

Aim

X‐linked Hypophosphatemia (XLH), caused by PHEX mutations, hinders skeletal and dental mineralization and contributes to tooth loss. While XLH is associated with dental implant‐related complications, no clinical or preclinical studies have investigated socket healing. XLH secondarily disrupts local mineral metabolism by increasing levels of the mineralization inhibitors, osteopontin (OPN) and inorganic pyrophosphate (PP_i). Tissue‐nonspecific alkaline phosphatase (TNAP) promotes mineralization by dephosphorylating OPN and hydrolyzing PP_i. In this proof‐of‐principle study, we hypothesized that alveolar bone socket healing defects in the Hyp mouse model of XLH would be improved by exogenous TNAP.

Methods

Maxillary first molars were extracted from wild‐type (WT) and Hyp mice at 6 weeks, and collagen gel ± mineral‐targeted TNAP (TNAP‐Fc‐D₁₀; asfotase alfa) was placed in sockets. Submucosal injections of TNAP‐Fc‐D₁₀ or saline were delivered at 7 and 14 days post‐procedure (dpp) in some mice. Maxillae were collected at 21 dpp for micro‐computed tomography, histology, and RT‐qPCR.

Results

Untreated Hyp mice showed impaired socket healing compared to WT mice in bone volume and density. TNAP delivered at the time of extraction was unable to improve healing in Hyp mice. However, additional local TNAP delivery increased both alveolar bone volume and density in Hyp mice. Histology indicated repeated TNAP increased both woven and mature bone in Hyp mouse sockets. Immunostaining for osteopontin and bone sialoprotein suggested partial resolution of osteoid accumulation.

Conclusion

TNAP enhanced socket healing in Hyp mice, overcoming inherent bone healing defects in XLH. These results provide new insights into bone healing with implications beyond alveolar bone in XLH.

Effects of exogenous tissue‐nonspecific alkaline phosphatase (TNAP) were analyzed in the Hyp mouse model of X‐linked hypophosphatemia (XLH). Maxillary first molars were extracted from wild‐type (WT) and Hyp mice at 6 weeks, and collagen gel ± TNAP was placed in sockets. In one group of Hyp mice, submucosal injections of TNAP or saline were delivered at 7 and 14 days post‐procedure (dpp). Maxillae were collected at 21 dpp, and bone healing was analyzed by micro‐computed tomography and histology.

Summary.

• Background

  • ○
    X‐linked hypophosphatemia (XLH) is a genetic condition that leads to weak bones and teeth. Patients experience frequent dental issues, including pulp infections and early tooth loss. Healing of the sockets after tooth loss is necessary to allow for dental implant placement, but this aspect of XLH has not been well studied.
• Added value of the study

  • ○
    This study used a mouse model of XLH to examine how the sockets heal after tooth extraction. We tested a treatment based on an enzyme called TNAP, which supports bone hardening. We found that giving TNAP more than once after the procedure significantly improved bone healing in the jaw. This is the first study to show that repeated TNAP treatment can help overcome the bone healing challenges seen in XLH.
• Clinical relevance

  • ○
    These results reveal a critical role for PHEX in alveolar bone socket healing that is deficient in XLH. For clinicians treating patients with XLH, dental procedures can be complicated by poor bone healing. This research suggests that TNAP therapy could be considered an adjunct treatment option to support better recovery after extractions or implants, potentially improving long‐term dental outcomes. The results may have implications beyond alveolar bone healing in XLH.

1. Introduction

X‐linked hypophosphatemia (XLH), the most common form of inherited rickets, is caused by inactivating mutations in the PHEX gene. Global incidence is estimated to be 1:20 000 to 70 000 [1], with regional variations calculated at 1:20000 in the United States [2], 1:25 000 to 1:60 000 in the United Kingdom and Europe [3, 4, 5], 1:20 000 in Japan [6], and about 1:62 500 to 75 000 in Australia and New Zealand [7]. Individuals with XLH exhibit elevated serum levels of fibroblast growth factor 23 (FGF23), leading to renal inorganic phosphate (P_i) wasting and decreased serum levels of 1,25 dihydroxyvitamin D₃ (1,25D), both of which contribute to impaired skeletal and dental mineralization [8, 9, 10].

Between 60% and 80% of adult patients diagnosed with XLH experience dental manifestations [9, 11, 12, 13, 14, 15, 16]. The most common dental problem is spontaneous dental pulp infection often leading to necrosis and abscesses; this is associated with thin enamel and defective dentin mineralization, particularly accumulation of interglobular dentin defects [17, 18, 19]. Increased prevalence of periodontal disease is associated with XLH, potentially related to reduced cementum and osteomalacic changes in alveolar bone, though this requires further study [19, 20]. Dental manifestations of XLH contribute to increased frequency of tooth loss and dental extractions, adding to the burden of disease. There are few case reports on dental implant placement in individuals with XLH [21, 22, 23, 24, 25]. Multiple of these publications describe increased instability and failure of implants [22, 23, 24], suggesting challenges in implant osseointegration. Clinical practice recommendations direct the use of one of the therapies for affected individuals for at least 3 months before and 6 months after dental implant placement, but the evidence basis remains weak due to so few clinical reports [26]. Based on current gaps in knowledge, it is imperative to understand how XLH influences alveolar bone healing and to explore potential strategies to enhance healing because those affected by XLH require extensive restorative dental procedures, possibly including dental implants.

While previous reports on dentoalveolar features of the Hyp mouse model of XLH have described the periodontal phenotype and response to induced periodontitis [27, 28, 29, 30], studies to date have not investigated alveolar bone socket healing in the context of XLH, either in human patients or in Hyp mice. Prior studies from our lab reported increased Phex gene expression in alveolar bone following tooth removal in mice [31, 32], suggesting PHEX plays a role in alveolar bone healing and raises the question of whether PHEX loss‐of‐function impairs healing. Understanding the underlying mechanisms and the potential to improve alveolar bone healing may contribute significantly to enhancing the quality of life of affected individuals.

Tissue‐nonspecific alkaline phosphatase (TNAP) is a critical enzyme for biomineralization of bones and teeth [33]. TNAP enzyme activity is unimpaired by PHEX inactivation in XLH and circulating TNAP levels (reported as ALP measurements) are increased in XLH and other osteomalacic disorders. However, TNAP has been reported to be locally suppressed in the mineralized tissues of individuals with XLH due to excessive FGF23 [34, 35]. This suppression occurs via Klotho‐independent signaling mediated by fibroblast growth factor receptor‐3 (FGFR3) in osteoblasts [36]. As a result, there is an accumulation of inorganic pyrophosphate (PP_i), a substrate of TNAP that acts as a mineralization inhibitor [37, 38]. Accumulation of PP_i is the primary pathological factor in hypophosphatasia (HPP; OMIM#146300, 241 500, and 241 510), which is caused by mutations in the ALPL gene that encodes TNAP [33, 39]. PP_i is thought to be a secondary contributor to hypomineralization in XLH, though its impact is not well understood. A recombinant enzyme replacement therapy for HPP was designed by adding a deca‐aspartate (D₁₀) tail to the human TNAP dimer, which effectively directs the enzyme to the mineralization front of bones and teeth where the enzymatic activity is required [40, 41, 42]. TNAP‐Fc‐D₁₀ (Asfotase alfa or Strensiq) has been an FDA‐approved treatment for HPP since 2015 [43].

Osteopontin (OPN) is a highly phosphorylated protein found in mineralized tissues and body fluids such as milk, saliva, and urine [44, 45]. OPN contains Acidic Serine Aspartate‐rich Motifs (ASARM) that can bind to hydroxyapatite, thereby regulating the mineralization process [46, 47]. OPN is encoded by the SPP1/Spp1 gene, located on chromosome 4 in humans and chromosome 5 in mice [48]. Previous studies from our laboratory and others have demonstrated that Spp1 knockout (Spp1 ^−/−) mice exhibit increased mineral volume, density, content, and/or crystallinity [49, 50, 51]. The ASARM portion of OPN (and related proteins, including matrix extracellular phosphoglycoprotein; MEPE) is a substrate for PHEX, which cleaves and inactivates the sequence, abolishing its mineral inhibitory properties [46, 47]. This is particularly relevant for those with XLH, who have an elevated accumulation of OPN ASARM fragments proposed to inhibit mineralization [18, 46, 47]. Additionally, OPN has been observed to be upregulated in the mineralized tissues of XLH patients, including dentin and alveolar bone, which is thought to contribute to defects in dentoalveolar mineralization and alveolar bone healing. The phosphorylation of OPN is essential for inhibiting the mineralization process; TNAP can dephosphorylate OPN, and consequently, phosphorylated OPN accumulates in HPP and contributes to its pathophysiology [47, 52]. Though TNAP is not deficient in XLH, this enzyme has the potential to reduce levels of both PP_i and phosphorylated OPN, two factors thought to contribute to mineralization defects in XLH.

We hypothesized that alveolar bone socket healing would be impaired in the Hyp mouse model of XLH. We used an established mouse socket healing model as a challenge to conduct a proof‐of‐principle study to determine if administration of TNAP‐Fc‐D₁₀, which dually reduces PP_i levels and dephosphorylates OPN, could improve bone healing. Evidence of enhanced alveolar bone healing in Hyp mice by exogenous TNAP would support the pathological roles of excess OPN and/or PP_i in bone healing in XLH and suggest a novel therapeutic approach. Furthermore, these experiments provide insights about PHEX in alveolar bone healing and may have implications for bone healing beyond the craniofacial region in XLH.

2. Methods

2.1. Mice

Animal studies were approved by the Institutional Animal Care and Use Committee (IACUC) at The Ohio State University (Columbus, OH) and conducted in accordance with ARRIVE 2.0 guidelines. Hyp mutant mice have been described previously [27, 29, 30, 53, 54, 55]. WT males and females, Hyp (hemizygous Phex ^mut) males, and heterozygous (Phex ^+/mut) females were included in the study (n = 4–7 mice/experimental group). Additional details are in the Appendix S1.

2.2. Tooth Extraction

Mice at 6‐weeks‐old were anesthetized with isoflurane using a nose cone. Bilateral extraction of maxillary first molars (M1) was achieved (Figure 1A) as previously described [31, 32]. After extraction, blood flow from the socket was stopped by using sterile gauze and paper points (Figure 1B,C). Liquid type I collagen solution was administered by a 30‐gauge needle with a 1 mL syringe into empty sockets with or without 20 mg of TNAP‐Fc‐D₁₀ (asfotase alfa) (Figure 1D). The preparation of collagen gel is described below. In two additional experimental groups at 7 and 14 dpp, we administered 30 mg/kg TNAP‐Fc‐D₁₀ or 0.9% sterile saline in a total volume of 2 μL locally to the buccal submucosal area near the maxillary first molar. Experimental groups are summarized in Table 1. In total, we performed tooth extractions on 55 mice, and 3 mice were excluded from analyses. All tissues were collected for the analysis of bone healing at 21 dpp. Additional details are in the Appendix S1.

FIGURE 1.

Mouse molar extraction and treatment. First maxillary molars (M1) are bilaterally extracted at 7 weeks. (A) M1 sockets show slow bleeding immediately after extraction. (B, C) Bleeding is stopped using sterile paper points to staunch blood flow. (D) Collagen ± TNAP‐Fc‐D₁₀ is injected by syringe into the socket. Immunohistochemistry (IHC) at 1 day post‐procedure (dpp) shows positive (red‐brown) localization of (E) tissue‐nonspecific alkaline phosphatase (TNAP) and (F) collagen type I (COL1A1) in M1 socket. Socket borders in E and F are outlined with a yellow dotted line. AB, alveolar bone.

TABLE 1.

Experimental groups.

Genotype	0 dpp		7 and 14 dpp
	Collagen type I	TNAP‐D10	Saline	TNAP‐D10
Wild‐type (WT)	X
Hyp	X
WT	X	X
Hyp	X	X
Hyp	X	X	X
Hyp	X	X		X

Note: First maxillary molar socket healing was analyzed at 21 days post‐procedure (dpp) in the six experimental groups summarized below.

2.3. Collagen Gel Preparation

Collagen gel was prepared on ice to prevent collagen from gelling before delivering it into tooth sockets, as previously described [31, 32, 56]. The retention of TNAP‐Fc‐D₁₀ gel was confirmed by performing immunohistochemistry with primary antibodies specific for TNAP or COL1A1 at 1 dpp, as demonstrated in Figure 1E,F. Additional details are in the Appendix S1.

2.4. Micro‐Computed Tomography (Micro‐CT)

Hemi‐maxillae were fixed with 10% neutral buffered formalin for 24 h and stored in 70% ethanol until scanning. Tissues were scanned in a μCT 50 (Scanco Medical, Bassersdorf, Switzerland) at 70 kVp, 76 μA, 0.5 mm Al filter, with 1200 ms integration time and 6‐μm voxel dimension. DICOM files were constructed from raw data, exported, and calibrated to a standard curve calculated from five known densities of hydroxyapatite (mg/cm³ HA). Reconstructed images were uploaded and analyzed using Analyze 15.0 (AnalyzeDirect, Overland Park, KS) as previously described [31, 32]. Socket healing was evaluated by calculating bone volume fraction (BV/TV), bone mineral density (BMD), and tissue mineral density (TMD) [57, 58]. Additional details are in the Appendix S1.

2.5. Histology

Tissues were decalcified in acetic acid/formalin/sodium chloride (AFS) solution for 5 weeks [59]. Tissue sections were used for Hematoxylin and Eosin (H&E) and Masson's trichrome staining and immunohistochemistry (IHC) for bone sialoprotein (BSP) and OPN [49, 60]. Additional details are in the Appendix S1.

2.6. Reverse Transcription Quantitative Polymerase Chain Reaction (RT‐qPCR)

RNA was collected from healing alveolar bone as previously described [31, 32]. Reverse transcription and qPCR (RT‐qPCR) was performed for 11 genes, as shown in Table S1, with Gapdh as the housekeeping gene. Additional details are in the Appendix S1.

2.7. Statistical Analysis

Mean ± standard deviation is shown in graphs. Data were analyzed by one‐way ANOVA and post hoc Tukey test using Prism v10.0 (GraphPad). Additional details are in the Appendix S1.

3. Results

3.1. Repeated Delivery of Exogenous TNAP Improves Defective Alveolar Bone Healing in Hyp Mice

In order to investigate the alveolar bone healing process in the Hyp mutant mouse model of XLH, we bilaterally extracted maxillary first molars (M1) and analyzed healing at 21 dpp. We first compared WT and Hyp mice receiving only collagen in sockets at the time of extraction. Hyp mice exhibited profound alveolar bone healing defects compared to WT mice (Figure 2A vs. Figure 2B). Quantitative micro‐CT analysis revealed a 60% decrease in bone volume fraction (BV/TV), an 80% decrease in bone mineral density (BMD), and a 50% decrease in tissue mineral density (TMD) in Hyp vs. WT controls (p < 0.0001 for all comparisons) (Figure 2G–I).

FIGURE 2.

Repeated Delivery of Exogenous TNAP Improves Defective Alveolar Bone Healing in Hyp Mice. Micro‐computed tomography analysis of alveolar bone (AB) socket healing at 21 days post‐procedure (dpp). (A–F) Both 3D (left) and 2D (right) images of first maxillary molar (M1) sockets are shown. New bone in 3D panels is shown in red and root sockets are outlined in yellow dotted lines in 2D images. Experimental groups are defined in Table 1 (n = 5–7/group). Quantitative analyses include (G) bone volume fraction (BV/TV), (H) bone mineral density (BMD), and (I) tissue mineral density (TMD). M2, second maxillary molar; M3, third maxillary molar. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

We next included TNAP‐Fc‐D₁₀ with collagen gel at the time of M1 extraction and assessed WT and Hyp mouse healing at 21 dpp. Exogenous TNAP improved alveolar bone healing in WT mice, with a 40% increase in BV/TV (p < 0.0001), a 20% increase in BMD (p < 0.05), and no difference in TMD compared to WT mice receiving only collagen (Figure 2A vs. Figure 2C,G–I). In contrast, the inclusion of TNAP with collagen at the time of M1 extraction did not improve any measured parameters of alveolar bone healing in Hyp mice (Figure 2B vs. Figure 2D,G–I).

We next investigated whether additional doses of TNAP‐Fc‐D₁₀ delivered submucosally near the M1 socket at later stages of healing had the potential to improve bone healing defects in Hyp mice. Compared to Hyp mice receiving repeated saline injections, those receiving additional TNAP‐Fc‐D₁₀ at 7 and 14 dpp showed approximately doubled BV/TV (p < 0.0001), whereas BMD did not show a significant change and TMD showed a 10% increase (p < 0.01) (Figure 2E vs. Figure 2F,G–I). Indeed, compared to all other Hyp experimental groups, those receiving repeated TNAP administration showed significantly increased BV/TV (p < 0.01 to 0.0001 for pairwise comparisons).

3.2. Repeated Delivery of Exogenous TNAP Increases Woven and Compact Bone in Sockets of Hyp Mice

We performed histology and immunohistochemistry (IHC) to characterize alveolar bone organization and composition in healing sockets. Histological assessments using H&E and Masson's Trichrome staining revealed collagenous matrix deposition indicative of woven bone formation beginning at the apical region of the sockets in Hyp mice, while WT mice displayed a greater presence of mature bone within the socket (Figure 3A,B). IHC for bone markers, OPN and BSP, showed labeling of newly formed bone in WT mice. New bone in Hyp mouse sockets showed regions of reduced OPN and BSP immunostaining, suggestive of hypomineralized osteoid that accompanies the osteomalacia documented in alveolar and appendicular bone in XLH.

FIGURE 3.

Repeated Delivery of Exogenous TNAP Increases Woven and Compact Bone in Sockets of Hyp Mice. (A, B) At 21 days post‐procedure (dpp), hematoxylin and eosin and (H&E) and Masson's Trichrome (MT) staining show woven alveolar bone (AB) formation beginning at the apical region of the sockets in Hyp mice, while WT mice display greater quantities of mature bone within the socket. Immunohistochemistry (IHC) for bone markers, osteopontin (OPN) and bone sialoprotein (BSP) labels new AB in WT mice compared to reduced OPN and BSP immunostaining in new AB in Hyp mice in osteoid‐like regions (red stars). (C, D) TNAP treatment increases quantities of mature bone in WT but not in Hyp mice, where osteoid‐like regions (red stars) persist. (E, F) Hyp mice receiving additional TNAP‐Fc‐D₁₀ show increased quantities of both woven and mature, compact‐like bone, and reduced appearance of osteoid‐like bone by OPN and BSP staining.

Whereas TNAP‐Fc‐D₁₀ treatment appeared to contribute to increased quantities of compact bone in WT mice, Hyp mice administered TNAP showed little change and still presented osteoid‐like regions by IHC (Figure 3C,D). Compared to Hyp mice receiving saline injections at 7 and 14 dpp, Hyp mice receiving additional TNAP‐Fc‐D₁₀ showed increased quantities of both woven and compact‐like bone and also reduced appearance of osteoid‐like bone indicated by more consistent OPN and BSP staining in new bone (Figure 3E,F).

3.3. No Significant Changes in the Expression of Mineralization‐Associated Genes After TNAP Treatment

We performed RT‐qPCR analysis on RNA harvested from healing socket alveolar bone in all experimental groups and measured mRNA for 11 genes implicated in the mineralization process and bone formation: Dspp, Dmp1, Ibsp, Mepe, Spp1, Sost, Phex, Fgf23, Alpl, Ank, and Enpp1. Our analyses indicated no significant differences between WT and Hyp mouse healing alveolar bone across treatment modalities, with the exception of Sost and Phex expression (Figure 4). Sost levels were found to be significantly elevated in the WT + TNAP group when compared to the Hyp group receiving weekly TNAP‐Fc‐D₁₀ injections and the saline injection group. The expression of the Phex gene exhibited a substantial increase following the delivery of TNAP in WT mice. Notably, our findings indicated a significant elevation in Phex expression in the WT + TNAP group when compared to all other groups.

FIGURE 4.

No Significant Changes in the Expression of Mineralization‐associated Genes after TNAP Treatment. RNA harvested from the healing socket alveolar bone was used to perform RT‐qPCR on 11 mineralization‐associated genes. No significant differences are observed between WT and Hyp mouse healing alveolar bone across treatment modalities, with the exception of Sost and Phex expression. Sost mRNA levels are significantly elevated in the WT + TNAP group vs. Hyp receiving weekly TNAP injections and the saline injection group. The expression of the Phex gene increases following the delivery of TNAP‐Fc‐D₁₀ in WT mice. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

4. Discussion

In this proof‐of‐principle study, we investigated alveolar bone socket healing in the Hyp mouse model of XLH. We administered exogenous TNAP, known to affect the levels of the mineralization inhibitors, PP_i and phosphorylated OPN. We discovered a substantial socket healing defect in Hyp mice compared to controls, identifying PHEX as playing an important role in alveolar bone healing. A single dose of exogenous, mineral‐targeted TNAP‐Fc‐D₁₀ at the time of extraction was insufficient to affect bone healing in Hyp mice, though it improved healing in WT mice. However, local administration of TNAP‐Fc‐D₁₀ during healing improved bone quantity in Hyp mice sockets, promoting increased woven and mature bone. These findings support that XLH‐associated PP_i and/or OPN inhibit new bone formation in this tooth extraction challenge model and that increased enzymatic activity from exogenous TNAP can partially ameliorate this inhibition. These results provide new insights into pathological mechanisms underlying XLH, as well as pointing to novel therapeutic possibilities.

4.1. Alveolar Bone Healing Defects in Hyp Mice Following Tooth Extraction

XLH affects mineral metabolism in several ways. PHEX loss‐of‐function promotes elevated circulating FGF23, which in turn causes systemic disturbances including renal P_i wasting, hypophosphatemia, reduced serum 1,25D, and secondary hyperparathyroidism, all contributing to defective mineralization. Conventional therapy for XLH, consisting of oral 1,25D and P_i, improves mineralization when initiated early, but has limited capacity to correct skeletal and dental defects, driving FGF23 levels even higher and leading to hypercalcemia and nephrocalcinosis [15, 20, 61, 62, 63]. Burosumab (Crysvita), a recently developed FGF23Ab, is poised to become standard‐of‐care. Burosumab demonstrates positive outcomes by raising serum P_i levels and reducing rickets severity, though long‐term outcomes remain uncertain [62, 63, 64, 65]. Conflicting dental outcomes have been reported for FGF23Ab treatment, with studies variably reporting no substantial improvement, a reduction in dental abscesses, or increased prevalence of dental abscesses in patients treated with burosumab compared to conventional treatment [64, 66, 67, 68, 69, 70]. However, studies to date are limited to relatively small cohorts over insufficient periods of time. An international working group provided recommendations to begin treatment with conventional therapy or burosumab early to provide the best oral health outcomes, but the level of evidence remains weak and there is no clear indication which treatment provides better outcomes [63]. Clinical practice recommendations suggest treatment of individuals with XLH for at least 3 months before and 6 months after dental implant surgery, but the evidence basis for this advice remains weak [26]. Larger, better controlled, and longer‐term studies will be necessary to provide detailed insights into effects of burosumab on oral health. Details such as timing of intervention and severity of disease will likely dictate response to treatment. FGF23Ab has also shown divergent effects on dentoalveolar tissues in Hyp mice [29, 71]. Limitations of both conventional therapy and burosumab to resolve dentoalveolar defects suggest underlying mechanisms of XLH that are unrecognized and continue unchecked with current therapies.

This study investigated the effects of XLH on alveolar bone socket healing after tooth extraction. This is a challenge model used to investigate stages of bone healing, consequences of gene knockout on healing, and potential approaches to improve healing [31, 32, 56, 72, 73, 74, 75, 76]. The Hyp mouse model of XLH carries inactivating mutations in Phex and phenocopies key biochemical, skeletal, and dentoalveolar manifestations of XLH [27, 30, 77, 78]. In this first study of alveolar bone socket healing in the Hyp model, we found profound defects in new bone formation, with deficiencies in both bone quantity and quality. Histology demonstrated the accumulation of osteoid and woven bone in the healing sockets of Hyp versus WT mice. The deficiencies in healing bone in Hyp mice closely match the developmental alveolar bone phenotype, including osteomalacia (reduced bone mineralization) and disorganization (inability for woven bone to remodel to mature, compact bone) [27, 30]. A prior study used a model of ligature‐induced periodontitis in Hyp mice, reporting delayed alveolar bone formation and impaired mineralization, marked by the accumulation of osteoid and increased OPN accumulation [27]. Those results parallel our findings in socket healing in the current study.

The importance of examining the post‐extraction healing process is underscored by the prevalence of dental infections and periodontal disease among XLH patients, which often culminate in tooth loss [26, 79, 80]. While dental implants represent a viable treatment for replacing lost teeth, case reports have reported complications and negative outcomes when using dental implants in XLH patients [23, 24, 25, 81]. Moreover, impairment in bone healing has been observed in XLH patients after femur and hip fractures and procedures, supporting the concept that XLH negatively affects bone healing [82]. The effects of burosumab on bone healing are deemed positive, leading to an international consortium to recommend the use of burosumab to improve fracture and pseudofracture healing [62, 83]. Lack of reports on alveolar bone healing and conflicting reports of the efficacy of FGF23 inactivation to improve dental mineralization in XLH [26, 68, 69, 71, 84] have prompted additional studies of unidentified pathological mechanisms [54, 55]. Identifying additional strategies to improve bone healing could substantially enhance the quality of life for individuals affected by XLH by increasing the success of dental implant placement, integration, and long‐term retention.

4.2. Exogenous TNAP Improved Alveolar Bone Healing in Hyp Mice

Two factors have emerged as potential contributors to mineralization defects in XLH, OPN and PP_i. The secreted, matrix protein OPN regulates and inhibits mineralization in vitro and in vivo [44, 50]. OPN particularly affects dentin and alveolar bone formation in mice [49]. PHEX cleaves and inactivates OPN [46, 47, 85, 86, 87, 88]. Increased deposition of OPN in bones and teeth in XLH likely contributes to mineralization defects [18, 27, 46, 55, 89, 90]. Prior research showed that genetic ablation of Spp1 in Hyp mice results in enhanced bone volume fraction (BV/TV) and reduced osteoid in the tibiae of Hyp mice [55], while a concurrent study demonstrated that dentoalveolar tissues were not similarly responsive to Spp1 knockout [54].

PP_i is a potent inhibitor of mineralization in the body [33, 37, 38, 91, 92, 93, 94, 95]. PP_i is locally increased by progressive ankylosis protein (ANKH/ANK) and ectonucleotide pyrophosphatase phosphodiesterase 1 (ENPP1/ENPP1) [96, 97, 98, 99, 100]. Levels of PP_i are reduced at sites of mineralization to allow hydroxyapatite crystal nucleation and growth, and the enzyme TNAP is expressed by mineralizing cells for this purpose [33, 96, 100, 101]. Increased PP_i results in hypomineralization disorders, the prime example being hypophosphatasia (HPP), resulting from loss‐of‐function mutations in ALPL [33, 39, 43, 102, 103]. Increased PP_i levels are found in Hyp mice due to dysregulation of PP_i metabolism, including increased ANK and ENPP1 and decreased TNAP [35, 55]. Thus, disruptions at both systemic (high FGF23, low 1,25D and P_i) and local (increased OPN and PP_i) levels likely contribute to mineralization defects in XLH. Local mineralization inhibitors OPN and PP_i have not been targeted by XLH treatment approaches to date.

We delivered exogenous mineralized tissue‐targeted tissue‐nonspecific alkaline phosphatase (TNAP‐Fc‐D₁₀) with a bovine collagen type I carrier to Hyp mice following molar extraction. TNAP‐Fc‐D₁₀ was selected as an optimal agent for proof‐of‐principle experiments because TNAP reduces PP_i levels [33] and inactivates OPN through dephosphorylation [52], and the recombinant enzyme localizes to mineralized tissues [40, 41, 43, 104] and is FDA‐approved (asfotase alfa, Strensiq), presenting a potential translational path. We employed TNAP‐Fc‐D₁₀ as a treatment to dually inactivate OPN and reduce PP_i levels in alveolar bone tissues.

Use of the collagen carrier likely aids in the retention of TNAP‐Fc‐D₁₀ in the socket, as demonstrated by IHC specific for TNAP and COL1A1. We found a substantial improvement in the quantity and quality of the new bone formation after TNAP treatment in WT mice. It is possible that TNAP may reduce the level of PP_i in the healing socket or might have effects on other mineralization regulators that accelerate bone formation even in healthy animals. Previous studies have shown that the P_i/PP_i ratio and PP_i regulators modulate osteogenesis [105, 106]. In contrast to the WT outcome, TNAP delivered only at the time of extraction failed to improve the alveolar bone healing defects in Hyp mice. Lack of improvement could be the result of several factors, but a likely problem is that this delivery is too soon. A single intravenous bolus dose of 5 mg/kg in adult mice indicated the half‐life of asfotase alfa in blood is approximately 34 h, with prolonged retention in bone [41]; the half‐life within bone is unknown and not easily determined. Inclusion of TNAP only at the onset of healing might result in the loss of the majority of activity by the time the enzyme is most useful in bone healing, that is, at stages after initial inflammation has resolved and new bone matrix is being generated by intramembranous ossification [32, 76]. To counteract this possibility, we added additional injections at 7 and 14 dpp in Hyp mice, which successfully enhanced alveolar bone healing in Hyp mice. The most obvious explanation for this different outcome is that TNAP activity was now increased at appropriate stages of bone healing, which could make a difference in the delayed healing process in Hyp mice. The underlying mechanisms require additional experiments to examine dose–response, expanded times of intervention, and additional healing time points. The evidence is consistent with increased TNAP counteracting OPN and PP_i inhibition of bone mineralization, though this requires additional confirmation.

It is important to note that TNAP cannot be considered an alternative to conventional therapy or burosumab, though results from this study support a potential role as an adjunct therapy that should be further explored. Asfotase alfa is the FDA‐approved mineral‐targeting form of recombinant human TNAP used in these experiments [41, 42]. The safety profile of asfotase alfa is considered acceptable and manageable, though it is not without risks. Adverse effects include most commonly injection site reactions, and more rarely, hypersensitivity and lipodystrophy. Ectopic calcifications in eyes have been linked to the treatment [107] and a preclinical model provided evidence that the drug may bind to pre‐existing vascular calcifications, potentially exacerbating those sites [104].

We note important limitations in this first proof‐of‐principle study. We include both male and female Hyp and WT mice in experimental groups to enhance scientific rigor and reproducibility and noted that XLH‐associated healing defects surpassed any potential sex‐related differences. However, the potential for sex‐related differences in disease severity and bone healing due to the X‐linked nature of PHEX suggests additional experiments should consider sex differences. Our RT‐qPCR interrogation of local mineralization‐associated genes was not able to capture any changes that would explain the improved healing outcomes with multiple deliveries of TNAP‐Fc‐D₁₀. Notably, our investigation was limited to 11 genes. It remains plausible that alterations may occur in other locally expressed genes implicated in alveolar bone healing, expression of factors from other tissues, and/or regulation of factors other than through transcription, which we did not measure here.

5. Conclusions

This study documented a substantial socket healing defect in the Hyp mutant mouse model of XLH, providing new insights into the importance of PHEX function during alveolar bone healing. This preclinical model corroborates case series reporting challenges and failures in dental implant integration in individuals with XLH. We demonstrated that multiple administrations of mineral‐targeted TNAP‐Fc‐D₁₀ during healing successfully improve bone quantity in sockets of Hyp mice, promoting increased woven and mature bone and partial resolution of accumulated osteoid. These findings support that XLH‐associated PP_i and/or OPN may inhibit new bone formation in this tooth extraction challenge model and that increased enzymatic activity from TNAP can partially overcome or ameliorate this inhibition. These results provide new insights into pathological mechanisms underlying XLH, as well as point to novel therapeutic possibilities in jaws or other skeletal tissues affected by XLH.

Author Contributions

A.P.: Writing – review and editing, writing – original draft, validation, methodology, investigation, formal analysis, data curation, conceptualization. B.D.: Writing – review and editing, methodology, investigation, formal analysis, data curation. J.L.M.: Writing – review and editing, conceptualization. B.L.F.: Writing – review and editing, writing – original draft, validation, supervision, project.

Disclosure

Artificial Intelligence Statement: This manuscript did not use artificial intelligence in any capacity.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Appendix S1: jre70044‐sup‐0001‐AppendixS1.docx.

Phanrungsuwan A., Donnelly B., Millán J. L., and Foster B. L., “Targeted Alkaline Phosphatase Therapy Enhances Alveolar Bone Healing in X‐Linked Hypophosphatemia in Mice,” Journal of Periodontal Research 61, no. 3 (2026): 309–322, 10.1111/jre.70044.

Data Availability Statement

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