Effects of Infantile Hypophosphatasia on Human Dental Tissue

To characterize the effect of HPP on dental tissues, eleven exfoliated deciduous teeth (two incisors, three canines, and six molars) were collected from a 24-year-old female patient (patient 1) with a compound-heterozygous mutation in the ALPL gene and an infantile onset HPP. Patient 1 presented with muscle weakness, respiratory complications and early loss of deciduous teeth. During adolescence she had suffered from fractures with delayed healing. The detailed patient characterization and description of the musculoskeletal apparatus has been published previously [22]. All teeth exfoliated between the age of 3 and 6 years. For comparison, ten corresponding control deciduous teeth after regular tooth loss (6 to 10 years age) from healthy children of the institute staff were analyzed.

For investigation of possible effects of asfotase alfa treatment on dental tissues, a 3 years and 2 months old male patient was analyzed (patient 2). He had been diagnosed with HPP because of low serum alkaline phosphatase (ALP) activity with consecutively elevated PLP levels and an increased deoxypyridinoline excretion in urine (Table 1). At the initial presentation he showed a height of 88.5 cm (1st centile), weight of 11.6 kg (3rd centile), and head circumference of 49 cm (8th centile). Further, he presented with an asymmetrical face, muscular hypotonia leading to a waddling gait, delayed motor development, rachitic changes, impaired bone microstructure (Table 1), intermittent abdominal pain, chronic constipation, and early loss of numerous deciduous teeth within the preceding 2 years (nine teeth until the age of 4 years and 8 months). To specify the underlying variant, molecular genetic analysis including gene panel sequencing with reference sequence ENST00000374840 was performed, revealing the missense mutation p.A176T.

At an age of 4 years and 1 month, treatment with asfotase alfa was initiated. Laboratory values were assessed at baseline visit and during asfotase alfa treatment and included ALP, bone ALP, PLP, calcium, phosphate, vitamin D status (25-OH-D3), and parathyroid hormone (PTH). Osteocalcin and desoxypyridinolin (DPD) were assessed during treatment. At baseline high resolution peripheral quantitative computed-tomography (HR-pQCT, XtremeCT I, Scanco Medical, CH) was performed at the right distal tibia in a previously described standardized procedure [31]. By this, cortical bone mineral density (Ct.BMD, mgHA/mm³), cortical thickness (Ct.Th, mm), bone volume over tissue volume (BV/TV, %), trabecular thickness (Tb.Th, mm), trabecular number (Tb.N, 1/mm), and trabecular bone mineral density (Tb.BMD, mg HA/mm³) were assessed according to previously published guidelines [32].

Shortly prior to initiation of therapy, patient 2 lost one deciduous tooth at the age of 4 years. During asfotase alfa treatment another tooth was lost at the age of 4 years and 8 months. Additionally, two control teeth were collected to match the teeth of patient 2. Clinical data were collected retrospectively, and all procedures performed in this study were approved by the local ethics committee (WF-059/19).

All teeth, i.e., HPP and controls, were halved and dehydrated in an increasing alcohol series. One half was embedded in methyl methacrylate for mineralized tissue assessment using quantitative backscattered electron imaging [33], while the other half was embedded in glycol methacrylate (Technovit 7200, Heraeus Kulzer GmbH, GER) for histological analysis, which were ground to a thickness of 100 µm and subsequently stained with toluidine blue.

To assess the mineralization degree of enamel, dentin, and cementum of the collected teeth, embedded, co-planar and polished samples were analyzed using a scanning electron microscope (LEO 435 VP, LEO Electron Microscopy Ltd., Cambridge, England) with a backscattered electron detector (BSE Detector Type 202, K.E. Developments Ltd., Cambridge, England) to acquire grey scale images [33]. The following parameters were maintained during the measurement; 680 pA current (confirmed by Faraday cup), 20 mm working distance, 20 keV voltage and ×100 magnification. Grey values were calibrated on aluminum and carbon standard (MAC Consultant Ltd., GB). For each region of interest, enamel, dentin, and cementum, eight images were obtained per region and tooth and evaluated using a custom-made Matlab code (MATLAB R2014a, MathWorks®, USA). Of note, acellular cementum was imaged starting at around 200 µm distal of the enamel-cementum junction to avoid analyzing tissue exposed to deep pockets. To differentiate between cementum and dentin, the regions of interest were selected based on Tomes’ granular layers separating both tissues as well as their different visual appearance. Based on the tissue mineral density distribution five different parameters describing the mineral pattern were determined; (i) the calcium mean representing the average calcium weight percentage (CaMean, wt%), (ii) the calcium peak as the highest calcium weight % observed in the region of interest (CaPeak, wt%), (iii) the standard deviation of the bone mineral density distribution which describes the heterogeneity of the mineralization pattern (CaWidth, wt%), (iv) the percentage of the tooth area with low mineralized tissue regions (CaLow, %) and (v) with high mineralized tissue regions (CaHigh, %). The impact of asfotase alfa on the mineralization degree of cementum and dentin was determined by normalizing the results of patient 2 teeth obtained prior and during treatment to matched control teeth.

For the analysis of structural parameters of the dental tissue, histological analysis on toluidine blue stained ground blocks was performed using Osteomeasure (Osteometrix, USA). For the staining, the surface of the tooth was immersed in 10% hydrogen peroxide for 10 min, washed in water and subsequently stained for 30 min using a 1% toluidine blue solution at pH 4.5. The following parameters were obtained; (i) the thickness of the acellular cementum (Ac.cem.Th., µm), (ii) the thickness of the repaired cementum (Repair cem.Th., µm) and (iii) the resorption surface proportional to the measured tooth surface (Resorption S./tooth S., %). To evaluate structural changes of tooth properties under asfotase alfa therapy, the cellular cementum thickness (Cellular Cem. Th., µm), the resorption surface over tooth surface and the dentin thickness (Dentin Th., µm) were determined and normalized to values of control teeth.

Within this study, teeth from two patients have been analyzed in comparison with matched control teeth. The characterization and skeletal phenotype of patient 1 has already been described in detail including HR-pQCT values [22]. Similar to patient 1, the enzyme replacement therapy with asfotase alfa had a beneficial effect on the skeletal apparatus of patient 2 with infantile onset of HPP. Beginning at age 3, patient 2 suffered from premature loss of his deciduous teeth, while one tooth was also lost during treatment (Fig. 1A). Of note, a right lateral crossbite caused by a narrowed upper jaw was noted, which has potential implications for inappropriate biomechanical stress (Fig. 1B, C). At age 4 years and 8 months (following 1 year of ERT), the exfoliated tooth 61 appeared with less signs of resorption in comparison to the tooth 81, which was lost right before start of the ERT administration. Additionally, the radiographic images indicate a densification and mineralization of trabecular bone near growth plates, e.g., at wrist and knee (Fig. 1E). Serum parameters before and during asfotase alfa treatment confirmed an increase of ALP & bone-specific ALP and a normalization of PTH and PLP, suggesting that mineralization was systemically promoted (Table 1).

To determine the effect of HPP on the mineralization of cementum, dentin, and enamel, qBEI images were evaluated in teeth from patient 1 and compared to the control cohort (Fig. 2A). The mineral density distributions of the dental tissues indicated differences between both groups in cementum and dentin, whereas enamel of the HPP patient 1 and control group showed a similar distribution (Fig. 2B).

Specifically, the mean (CaMean, 26.1 ± 1.2 wt% vs. 21.8 ± 4.3 wt%) and peak (CaPeak, 26.5 ± 1.0 wt% vs. 21.6 ± 5.0 wt%) calcium content of the cementum were significantly lower in the HPP patient 1 compared to controls (Fig. 2C), whereas CaWidth (2.9 ± 0.4 wt% vs. 3.1 ± 0.4 wt%), CaLow (5.0 ± 2.6% vs. 41.9 ± 40.4%), and CaHigh (4.9 ± 3.6% vs. 2.1 ± 2.7%) were similar between the two groups.

In dentin, a comparable reduced mineralization was observed with a significantly lower CaMean (27.5 ± 1.0 wt% vs. 25.9 ± 1.0 wt%) and CaPeak (27.8 ± 1.0 wt% vs. 26.4 ± 1.0 wt%), while CaLow (5.0 ± 2.2% vs. 9.3 ± 5.5%) and CaHigh (5.0 ± 3.6% vs. 2.0 ± 2.4%) showed a higher amount of hypomineralized regions with concurrent smaller amount of hypermineralized regions. The mineralization heterogeneity (CaWidth, 2.9 ± 0.3 wt% vs. 2.9 ± 0.7 wt%) was similar in HPP patient 1 and control group.

In contrast to cementum and dentin, in enamel the CaMean (36.7 ± 0.3 wt% vs. 36.3 ± 0.4 wt%) and CaPeak (36.9 ± 0.3 wt% vs. 36.8 ± 0.2 wt%) were similar between groups, while CaWidth (1.6 ± 0.3 wt% vs. 2.2 ± 0.7 wt%) was significantly higher in the HPP group. CaLow (4.9 ± 3.8% vs. 8.5 ± 5.3%) and CaHigh (4.9 ± 4.7% vs. 2.3 ± 2.6%) of enamel were similar in both groups.

To analyze the differences in tooth structures, histological evaluation in ground sections was performed. The entire teeth for control group and HPP patient 1 (Fig. 3A) were analyzed by evaluating unique structures of the cementum. Acellular cementum thickness was determined and showed similar values for HPP and control group (17.68 ± 10.26 µm vs. 21.20 ± 13.49 µm, respectively). Of note, four out of ten control teeth and two out of eleven HPP teeth showed absent acellular cementum. However, the absence of acellular cementum in control teeth was due to regular complete resorption of the tooth root, whereas in HPP teeth acellular cementum was absent in the above-mentioned cases despite existing tooth roots. A significantly higher repair cementum thickness was observed in HPP teeth compared to control teeth (38.61 ± 5.47 µm vs. 3.92 ± 8.66 µm, respectively), which was observed in three out of ten control teeth and in nine out of eleven HPP teeth. The resorption surface over tooth surface in HPP teeth was similar in HPP with 19.69 ± 11.91% compared to control teeth with 14.30 ± 10.36% (Fig. 3B).

QBEI measurements before and during treatment with asfotase alfa were used to analyze its effect on the three dental tissues, i.e., cementum, dentin, and enamel, in deciduous teeth (Fig. 4A–C). In the tooth of patient 2, cementum mineralization slightly adapted to control level by a reduction of CaLow (221% to 79% normalized to the control) and elevation of CaHigh (1% to 130% normalized to the control). The dentin mean and peak mineralization of patient 2 slightly changed from control level to 106% and 107%. Simultaneously, hypomineralized areas became more and hypermineralized areas less frequent. Of note, changes in mineralization pattern of enamel were not observed.

Regarding histological analysis (Fig. 4D), the resorption surface per tooth surface of patient 2 was reduced during treatment from 79 to 6% normalized to control, while the normalized dentin thickness showed slightly higher values during asfotase alfa treatment in patient 2 (89% and 92%; before and during treatment, respectively; data not shown) (Fig. 4E). Cellular cement thickness was not detectable in patient 2 and matched control tooth.