Gene Therapy Using Recombinant AAV Type 8 Vector Encoding TNAP-D10 Improves the Skeletal Phenotypes in Murine Models of Osteomalacia

Abstract

Hypophosphatasia (HPP), caused by loss-of-function mutations in the ALPL gene encoding tissue-nonspecific alkaline phosphatase (TNAP), is characterized by skeletal and dental hypomineralization that can vary in severity from life-threatening to milder manifestations only in adulthood. PHOSPHO1 deficiency leads to early-onset scoliosis, osteomalacia, and fractures that mimic pseudo-HPP. Asfotase alfa, a life-saving enzyme replacement therapy approved for pediatric-onset HPP, requires subcutaneous injections 3 to 6 times per week. We recently showed that a single injection of an adeno-associated virus vector serotype 8 harboring TNAP-D₁₀ (AAV8-TNAP-D₁₀) effectively prevented skeletal disease and prolonged life in Alpl ^-/- mice phenocopying infantile HPP. Here, we aimed to determine the efficacy of AAV8-TNAP-D₁₀ in improving the skeletal and dental phenotype in the Alpl ^Prx1/Prx1 and Phospho1 ^-/- mouse models of late-onset (adult) HPP and pseudo-HPP, respectively. A single dose of 3 × 10¹¹ vector genomes per body (vg/b) was injected intramuscularly into 8-week-old Alpl ^Prx1/Prx1 and wild-type (WT) littermates, or into 3-day-old Phospho1 ^-/- and WT mice, and treatment efficacy was evaluated after 60 days for late-onset HPP mice and after 90 days for Phospho1 ^-/- mice. Biochemical analysis showed sustained serum alkaline phosphatase activity and reduced plasma PP_i levels, and radiographic images, micro-computed tomography (micro-CT) analysis, and hematoxylin and eosin (H&E) staining showed improvements in the long bones in the late-onset HPP mice and corrected scoliosis in the Phospho1 ^-/- mice. Micro-CT analysis of the dentoalveolar complex did not reveal significant changes in the phenotype of late-onset HPP and pseudo-HPP models. Moreover, alizarin red staining analysis showed that AAV8-TNAP-D₁₀ treatment did not promote ectopic calcification of soft organs in adult HPP mice after 60 days of treatment, even after inducing chronic kidney disease. Overall, the AAV8-TNAP-D₁₀ treatment improved the skeletal phenotype in both the adult HPP and pseudo-HPP mouse models. This preclinical study will contribute to the advancement of gene therapy for the improvement of skeletal disease in patients with heritable forms of osteomalacia. © 2022 The Authors. JBMR Plus published by Wiley Periodicals LLC on behalf of American Society for Bone and Mineral Research.

Keywords: ANALYSIS/QUANTITATION OF BONE; ANIMAL MODELS; BONE QCT/MICRO‐CT; DENTAL BIOLOGY; DISEASES AND DISORDERS OF/RELATED TO BONE; GENE THERAPY; GENETIC ANIMAL MODELS; OSTEOMALACIA AND RICKETS; PRECLINICAL STUDIES.

Fig. 1

Biochemical analysis in serum/plasma from adult hypophosphatasia (HPP) mice and wild‐type (WT) littermates under AAV8‐TNAP‐D₁₀ treatment or AAV8‐GFP control, 60 days after injection. (A, B) Body weight. (C, D) Serum alkaline phosphatase activity. (E, F) Plasma PP_i levels. (G, H) Serum calcium concentrations. (I, J) Serum phosphorus concentrations. (K, L) Blood urea nitrogen (BUN) levels in serum. Sample size of the cohorts was as follows: (i) HPP: AAV8‐TNAP‐D₁₀ female n = 6, male n = 6; (ii) HPP: AAV8‐GFP female n = 3, male n = 4; (iii) WT: AAV8‐TNAP‐D₁₀ female n = 3, male n = 4; (iv) WT: AAV8‐GFP female n = 3, male n = 4. Statistical analysis was performed by one‐way ANOVA followed by Tukey's multiple comparison test. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

Fig. 2

Biochemical analysis in serum/plasma from Phospho1 knockout (KO) mice under AAV8‐TNAP‐D₁₀ treatment or AAV8‐GFP as control after 45 and 90 days of injection. Wild‐type (WT) littermates were treated with the AAV8‐TNAP‐D₁₀ vector. (A, B) Body weight. (C, D) Serum alkaline phosphatase activity. (E, F) Plasma PP_i levels. (G, H) Serum calcium concentrations. (I, J) Serum phosphorus concentrations. (K, L) Blood urea nitrogen (BUN) levels in serum. Sample size of the cohorts was as follows: (i) Phospho1 KO: AAV8‐TNAP‐D₁₀ female n = 3, male n = 6; (ii) Phospho1 KO: AAV8‐GFP female n = 5, male n = 3; (iii) WT: AAV8‐TNAP‐D₁₀ female n = 3, male n = 4. Statistical analysis was performed by one‐way ANOVA followed by Tukey's multiple comparison test. *p < 0.05; **p < 0.01; ***p < 0.001. ****p < 0.0001.

Fig. 3

Radiographic findings and hematoxylin and eosin (H&E) staining of long bones in adult hypophosphatasia (HPP) mice bone phenotype. X‐ray images from (A–D) female and (E–H) male HPP mice and wild‐type (WT) littermate control treated with AAV8‐TNAP‐D₁₀ or AAV8‐GFP after 60 days of injection. Impaired long bones from (C, G) HPP mice with defects in the proximal and distal femur along with the patellar joint surface and proximal tibia. (D–H) TNAP treatment partially rescued the epiphyseal and metaphyseal bone regions (highlighted in red). (I) Femur and (J) tibia length measurement. (K‐O) H&E staining showing the growth plate in AAV8‐GFP and (L‐P) in AAV8‐TNAP‐D₁₀ ‐treated WT mice. (M–Q) H&E staining showed the disorganization of the growth plate in AAV8‐GFP‐treated HPP mice and (N–R) a substantial improvement of the bone morphologic parameters in AAV8‐TNAP‐D₁₀‐treated HPP mice. H&E images were acquired in tiling mode and 20× magnification. Statistical analysis was performed by one‐way ANOVA followed by Tukey's multiple comparison test. *p < 0.05. **p < 0.01. ***p < 0.001. ****p < 0.0001.

Fig. 4

Micro‐CT analysis of femur from adult hypophosphatasia (HPP) mice and wild‐type (WT) littermates. 2D and 3D micro‐CT images of femurs from (A–D) females and (E–H) males treated with AAV8‐TNAP‐D₁₀ or AAV8‐GFP after 60 days of injection. (I–T) Micro‐CT analysis of bone parameters in the femurs showing reduced cortical bone volume fraction (Ct.BV/TV) associated with a reduced cortical thickness (Ct.Th) and increased marrow area (Ma.Ar) in untreated HPP mice compared with AAV8‐TNAP‐D₁₀ or AAV8‐GFP WT controls. These cortical bone defects were partially rescued after 60 days of AAV8‐TNAP‐D₁₀ treatment. Statistical analysis was performed by one‐way ANOVA followed by Tukey's multiple comparison test. Differences were statistically significant at *p < 0.05. **p < 0.01. ***p < 0.001. ****p < 0.0001.

Fig. 5

Representative X‐ray images and histological sections of the spine from female and male 90‐day‐old Phospho1 knockout (KO) mice. A single dose of AAV8‐TNAP‐D₁₀ or AAV8‐GFP, as control (3 × 10¹¹ vg/body), was intramuscularly administered in 3‐day‐old mice. (A, B) The control AAV8‐GFP‐treated Phospho1 KO mice showed scoliosis. (C, D) Gene therapy using the AAV8‐TNAP‐D₁₀ vector corrected the scoliosis deformity of the spine. (E) Cobb angle measurements. The threshold of scoliosis (Cobb angle >10°) was indicated with a dashed red line. Hematoxylin and eosin (H&E) staining showed a different tissue organization of vertebrae from mice treated with (F, G) AAV8‐GFP or (H, I) the vector encoding TNAP, revealing increased trabecular bone and less bone marrow area in AAV8‐TNAP‐D₁₀‐treated mice. H&E images were acquired in 10× magnification.

Fig. 6

Micro‐CT analysis of tibias from Phospho1 knockout (KO) mice and wild‐type (WT) littermates. (A) 2D micro‐CT images of tibias from females and males treated with AAV8‐TNAP‐D₁₀ or AAV8‐GFP 90 days after injection. (B–I) Quantification of trabecular bone parameters showing increased bone volume and trabecular bone connectivity in AAV8‐TNAP‐D₁₀‐treated Phospho1 KO compared with WT and untreated Phospho1 KO mice. (J–M) quantification of cortical bone parameters showed no statistically significant difference among the groups. Statistical analysis was performed by one‐way ANOVA followed by Tukey's multiple comparison test. Differences were statistically significant at *p < 0.05 (p = 0.0298) and **p < 0.01 (p = 0.0071).

Fig. 7

Micro‐CT analysis of dentoalveolar complex from adult hypophosphatasia (HPP) mice and wild‐type (WT) littermates. 2D and 3D micro‐CT images of (A–D) first molar (M1) and incisor (INC) from AAV8‐TNAP‐D₁₀‐ or AAV8‐GFP‐treated mice 60 days after injection. (E) Quantification of molar enamel, dentin, alveolar bone, and pulp parameters. (F) Quantification of incisor enamel and dentin volumes and densities. (G–J) Hematoxylin and eosin (H&E) staining of tissue organization of teeth. No obvious difference in tissue organization, periodontal attachments, or acellular cementum between AAV8‐TNAP‐D₁₀‐ or AAV8‐GFP‐treated mice. Statistical analysis was performed by one‐way ANOVA followed by Tukey's multiple comparison test. Differences were statistically significant at *p < 0.05. AB = alveolar bone; CC = cellular cementum; CEJ = cemento‐enamel junction; DE = dentin; EN = enamel; PDL = periodontal ligaments.

Fig. 8

Micro‐CT analysis of dentoalveolar complex from Phospho1 knockout (KO) and wild‐type (WT) littermates. 2D and 3D micro‐CT images of (A–C) first molar (M1) and incisor (INC) from AAV8‐TNAP‐D₁₀‐ or AAV8‐GFP‐treated mice 90 days after injection. (D) Quantification of molar enamel, dentin, alveolar bone, and cellular cementum parameters. No obvious difference in tissue organization, or periodontal attachments, except for higher dentin and acellular cementum volumes from AAV8‐TNAP‐D₁₀‐treated WT mice when compared with Phospho1 KO mice treated with AAV8‐TNAP‐D₁₀ and AAV8‐GFP. Statistical analysis was performed by one‐way ANOVA followed by Tukey's multiple comparison test. Differences were statistically significant at *p < 0.05 and **p < 0.01. AB = alveolar bone; DE = dentin; EN = enamel.