- Email: [email protected]
Studies of osteoclast pathogenesis of craniometaphyseal dysplasia (CMD) in a mouse model and in patient-specific IPS cells
548
conditions. Case reports and small series suggest that radiation and medical therapies (predominantly interferon and/or bisphosphonates) can stabilize progressive disease. These studies are limited by inconsistent phenotyping, variation in length of therapy and follow-up, and publication bias. We performed a retrospective study of 102 patients who have been referred to our center (24 with G-SD and 78 with GLA). Our data suggest medical therapies may allow remineralization; however, we do not know whether all patients benefit or whether those that do have sustained improvement. Important next steps include a thorough study of natural history and responses to therapy in large patient cohorts, prospective interventional trials with clearly defined outcome measures, and discovery of the genetic cause(s). Recent technologic and analytic advances in DNA and RNA sequencing, improvements in recreating human disease-causing mutations in model organisms, and in high-throughput screening for new therapeutic agents create optimism that the scientific and medical community will soon achieve a detailed understanding of the causes of G-SD and GLA and devise improved therapies for patients.
Melorheostosis . . . By Michael Whyte, MD; Shriners Hospital, Washington University School of Medicine, St. Louis, MO
Melorheostosis (MEL) refers to “flowing hyperostosis” (dense bone), typically in the limbs, appearing radiographically like wax dripping down a candle. Reports of 200 cases show sporadic occurrence (not inherited), although MEL can appear in the genetic “spotted bone” disorders, osteopoikilosis (OPK) and Buschke-Ollendorff syndrome (BOS). MEL typically presents during childhood in one limb, and otherwise is asymmetrical. Skin changes may overlie dense bone and can include scar-like tissue, excessive hair, and small blood vessels. The collagen appears normal, i.e., linear melorheostotic scleroderma. Pain and stiffness are major symptoms. Affected joints can contract and deform. Leg-length inequality sometimes occurs from soft-tissue contractures or premature fusion of growth plates. The skeletal lesions seem to progress most during childhood. In adults, MEL may gradually extend, but pain is especially frequent. Thickening of the inner cortical bone occurs during childhood, and then at the surface during adulthood. Irregular, eccentric, osteosclerosis is the radiographic consequence. Any bone may be affected, but most commonly within lower limbs. Ectopic bone can develop, particularly near joints. MEL bone is hyperemic and causes a “hot” bone scan. Routine biochemical studies are unremarkable. Its anatomic
Conference Proceedings
distribution in sclerotomes, myotomes, and dermatomes suggests a segmentary defect during embryogenesis. Linear scleroderma may reflect the primary abnormality that descends into bone. Affected skin has an altered expression of several adhesion proteins. Germline mutation of LEMD3 causes OPK and BOS, but not classic MEL. Contractures or neurovascular compression can require surgery, but it is challenging and recurrent deformity is common. Distraction techniques have been promising.
Studies of Osteoclast Pathogenesis of Craniometaphyseal Dysplasia (CMD) in a Mouse Model and in Patient-specific IPS Cells . . . By I.-Ping Chen, DDS, PhD, Liping Wang, MD, Keiichi Fukuda, MD, PhD, Noemi Fusaki, PhD, Akihiro Iida, PhD, Alexander Lichtler, PhD, and Ernst J. Reichenberger, PhD; University of Connecticut, Farmington, CT
Rare genetic bone disorders are of significant clinical relevance because of their number and their life-time debilitating impact on patients. Treatment options are often limited due to insufficient knowledge of their pathogenesis. Studies have been plagued by the unavailability of primary cells/tissues and suitable animal models. Patient-specific induced pluripotent stem (iPS) cells offer new avenues for studying bone cells from patients with rare diseases. We study craniometaphyseal dysplasia (CMD) utilizing a knock-in mouse model and patientspecific iPS cells. CMD is characterized by hyperostosis of craniofacial bones concurrent with widened metaphyses in long bones. Mutations for autosomal dominant CMD have been identified in the ANK gene (ANKH). A knockin (KI) mouse model expressing a human Ank mutation (Phe377del) replicates many features of CMD. We observed defects in AnkKI/KI osteoclast (OC) cultures including (1) decreased OC formation; (2) reduced mineral resorption; (3) reduced OC migration shown by live-cell time-lapse imaging; and (4) altered podosome organization. The bone mass phenotype of AnkKI/KI mice is partially rescued by wild-type bone marrow transplants. We hypothesize that CMD-causing ANKH mutations decrease the osteoclast activity by negatively affecting the actin cytoskeleton. Our ultimate goal is to test this hypothesis in the human system using patientspecific inducible pluripotent stem cells (iPSCs). We derived iPSCs from peripheral blood mononuclear cells of CMD patients and healthy controls with four separate Sendai-virus vectors encoding OCT3/4, SOX2, KLF4, and c-MYC. The Sendai virus, a cytoplasmic RNA vector, can produce iPSCs free of vector integration into chromosomes. The pluripotency of these iPSCs is tested by (1) expression of hES cell markers; (2) embryoid body
Conference Proceedings
formation; and (3) teratoma formation and normal karyotypes are confirmed. iPSCs from a normal control have already been differentiated into multinucleated TRAP-positive OC-like cells. We are currently differentiating OCs from CMD iPSCs and comparing OCs derived from CMD to normal iPSCs. We expect that combining mouse data with findings from human iPS cells will significantly increase our understanding of the CMD pathology. If successful, I believe that this model can serve as paradigm to study other rare genetic skeletal disorders.
The Phosphate-lowering Effect of Nicotinamide is Offset by Reduced Fgf23 Levels in a Murine Model of Familial Tumoral Calcinosis . . . By Shoji Ichikawa, PhD, and Austin M. Reilly; Department of Medicine, Indiana University School of Medicine, Indianapolis, IN
Familial tumoral calcinosis is caused by mutations in the GALNT3 gene. Lack of GalNAc transferase 3, encoded by GALNT3, destabilizes FGF23, a key hormone that suppresses phosphate reabsorption and 1,25-dihydroxyvitamin D synthesis in the kidney. The destabilized FGF23 is more susceptible to proteolytic cleavage, thereby reducing the secretion of biologically active intact FGF23. Persistent hyperphosphatemia due to decreased FGF23 concentrations leads to often large, ectopic calcific masses in soft tissues, which are usually removed surgically. However, the calcific masses often recur, requiring a more permanent solution to the problem. Nicotinamide is reported to lower serum phosphate by decreasing type IIb sodium-dependent phosphate cotransporter in the gut. This effect of nicotinamide has promoted its use in treatment of limited cases of tumoral calcinosis, though its effectiveness remains largely unclear. Therefore, we investigated nicotinamide as a potential therapy for tumoral calcinosis, using a murine model of the disease – Galnt3 knockout mice. Initially, nicotinamide (doses 0, 2.5, 5, 7.5, and 10 mmol/kg/day) was given to normal mice for 2 weeks. Treatment had no effect on serum phosphate levels; however, Fgf23 was decreased in a dose-dependent manner. Subsequently, high-dose nicotinamide (10 mmol/kg/day) was tested for 4 weeks in Galnt3 knockout mice on a high-phosphate diet. The radiographic data pretreatment and posttreatments showed that the treatment did not eliminate the calcification, but retarded its growth, while in the untreated mice, calcifications increased in size. The therapy did not change serum phosphate levels despite moderately increased phosphate excretion, likely due to decreased serum intact Fgf23 levels. Quantification of calcium and phosphate contents in hearts and kidneys revealed
549
that the treated mice had significantly high calcium in the heart. In summary, nicotinamide did not alter serum phosphate levels because the phosphate-lowering effect of nicotinamide was diminished by reduction in intact Fgf23 concentrations. The increased calcium in the heart suggests that nicotinamide therapy may also have an adverse effect.
Enchondroma resulting from loss of a Stk11-dependent switch of proliferative chondrocytes to a postmitotic fate . . . By Lick Pui Lai, PhD; University of Southern California, Los Angeles, CA
Stk11 (also known as liver kinase b1 (Lkb1)) is a serinethreonine protein kinase that acts upstream of the AMPactivated protein kinase (AMPK) family in coupling energy homeostasis to cell growth, proliferation and survival. Through chondrocyte-specific removal of Lkb1 activity, we showed that Stk11 is required for the normal switch of mitotic chondrocytes to a postmitotic hypertrophic chondrocyte fate. Consequently, it led to a dramatic overgrowth of the growth plate in the Stk11-mutant postnatal skeletal elements. To determine the molecular mechanisms underlying Stk11 action, we examined the mTOR pathway, which is inhibited through AMPK in growth regulation. Strikingly, rapamycin treatment of the pregnant mouse was able to rescue the delay in chondrocyte hypertrophy in Stk11mutant embryos, suggesting that Stk11 inhibition of mTOR signaling is critical for the switch in chondrocyte fate. Since the dramatic overgrowth of the growth plate is characteristic of enchondroma, we also examined the tumorigenicity of the mutant chondrocytes both in vitro and in vivo. In contrast to wild-type chondrocytes isolated from the postnatal day-30 growth plate, Stk11-mutant chondrocytes proliferated and formed colonies in monolayer and anchorage-independent agar cultures, indicative of a neoplastic transformation in vitro. Similarly, allotransplantation of mutant chondrocytes into immune-deficient NOG mice also resulted in tumor formation in vivo. Gene Ontology analysis of gene expression profiles indicated an augmented activity of cell proliferation and cell cycle regulators within the enchondroma-chondrocyte population compared to chondrocytes in the normal growth plate. Taken together, these data demonstrated an unexpected role of Stk11 in balancing proliferative and nonproliferative hypertrophic states of chondrocyte development through the regulation of mTOR signaling. Stk11 is a known tumor suppressor; our findings raise the possibility that loss of Stk11 may play a role in human enchondroma, a possibility that we are investigating.
conditions. Case reports and small series suggest that radiation and medical therapies (predominantly interferon and/or bisphosphonates) can stabilize progressive disease. These studies are limited by inconsistent phenotyping, variation in length of therapy and follow-up, and publication bias. We performed a retrospective study of 102 patients who have been referred to our center (24 with G-SD and 78 with GLA). Our data suggest medical therapies may allow remineralization; however, we do not know whether all patients benefit or whether those that do have sustained improvement. Important next steps include a thorough study of natural history and responses to therapy in large patient cohorts, prospective interventional trials with clearly defined outcome measures, and discovery of the genetic cause(s). Recent technologic and analytic advances in DNA and RNA sequencing, improvements in recreating human disease-causing mutations in model organisms, and in high-throughput screening for new therapeutic agents create optimism that the scientific and medical community will soon achieve a detailed understanding of the causes of G-SD and GLA and devise improved therapies for patients.
Melorheostosis . . . By Michael Whyte, MD; Shriners Hospital, Washington University School of Medicine, St. Louis, MO
Melorheostosis (MEL) refers to “flowing hyperostosis” (dense bone), typically in the limbs, appearing radiographically like wax dripping down a candle. Reports of 200 cases show sporadic occurrence (not inherited), although MEL can appear in the genetic “spotted bone” disorders, osteopoikilosis (OPK) and Buschke-Ollendorff syndrome (BOS). MEL typically presents during childhood in one limb, and otherwise is asymmetrical. Skin changes may overlie dense bone and can include scar-like tissue, excessive hair, and small blood vessels. The collagen appears normal, i.e., linear melorheostotic scleroderma. Pain and stiffness are major symptoms. Affected joints can contract and deform. Leg-length inequality sometimes occurs from soft-tissue contractures or premature fusion of growth plates. The skeletal lesions seem to progress most during childhood. In adults, MEL may gradually extend, but pain is especially frequent. Thickening of the inner cortical bone occurs during childhood, and then at the surface during adulthood. Irregular, eccentric, osteosclerosis is the radiographic consequence. Any bone may be affected, but most commonly within lower limbs. Ectopic bone can develop, particularly near joints. MEL bone is hyperemic and causes a “hot” bone scan. Routine biochemical studies are unremarkable. Its anatomic
Conference Proceedings
distribution in sclerotomes, myotomes, and dermatomes suggests a segmentary defect during embryogenesis. Linear scleroderma may reflect the primary abnormality that descends into bone. Affected skin has an altered expression of several adhesion proteins. Germline mutation of LEMD3 causes OPK and BOS, but not classic MEL. Contractures or neurovascular compression can require surgery, but it is challenging and recurrent deformity is common. Distraction techniques have been promising.
Studies of Osteoclast Pathogenesis of Craniometaphyseal Dysplasia (CMD) in a Mouse Model and in Patient-specific IPS Cells . . . By I.-Ping Chen, DDS, PhD, Liping Wang, MD, Keiichi Fukuda, MD, PhD, Noemi Fusaki, PhD, Akihiro Iida, PhD, Alexander Lichtler, PhD, and Ernst J. Reichenberger, PhD; University of Connecticut, Farmington, CT
Rare genetic bone disorders are of significant clinical relevance because of their number and their life-time debilitating impact on patients. Treatment options are often limited due to insufficient knowledge of their pathogenesis. Studies have been plagued by the unavailability of primary cells/tissues and suitable animal models. Patient-specific induced pluripotent stem (iPS) cells offer new avenues for studying bone cells from patients with rare diseases. We study craniometaphyseal dysplasia (CMD) utilizing a knock-in mouse model and patientspecific iPS cells. CMD is characterized by hyperostosis of craniofacial bones concurrent with widened metaphyses in long bones. Mutations for autosomal dominant CMD have been identified in the ANK gene (ANKH). A knockin (KI) mouse model expressing a human Ank mutation (Phe377del) replicates many features of CMD. We observed defects in AnkKI/KI osteoclast (OC) cultures including (1) decreased OC formation; (2) reduced mineral resorption; (3) reduced OC migration shown by live-cell time-lapse imaging; and (4) altered podosome organization. The bone mass phenotype of AnkKI/KI mice is partially rescued by wild-type bone marrow transplants. We hypothesize that CMD-causing ANKH mutations decrease the osteoclast activity by negatively affecting the actin cytoskeleton. Our ultimate goal is to test this hypothesis in the human system using patientspecific inducible pluripotent stem cells (iPSCs). We derived iPSCs from peripheral blood mononuclear cells of CMD patients and healthy controls with four separate Sendai-virus vectors encoding OCT3/4, SOX2, KLF4, and c-MYC. The Sendai virus, a cytoplasmic RNA vector, can produce iPSCs free of vector integration into chromosomes. The pluripotency of these iPSCs is tested by (1) expression of hES cell markers; (2) embryoid body
Conference Proceedings
formation; and (3) teratoma formation and normal karyotypes are confirmed. iPSCs from a normal control have already been differentiated into multinucleated TRAP-positive OC-like cells. We are currently differentiating OCs from CMD iPSCs and comparing OCs derived from CMD to normal iPSCs. We expect that combining mouse data with findings from human iPS cells will significantly increase our understanding of the CMD pathology. If successful, I believe that this model can serve as paradigm to study other rare genetic skeletal disorders.
The Phosphate-lowering Effect of Nicotinamide is Offset by Reduced Fgf23 Levels in a Murine Model of Familial Tumoral Calcinosis . . . By Shoji Ichikawa, PhD, and Austin M. Reilly; Department of Medicine, Indiana University School of Medicine, Indianapolis, IN
Familial tumoral calcinosis is caused by mutations in the GALNT3 gene. Lack of GalNAc transferase 3, encoded by GALNT3, destabilizes FGF23, a key hormone that suppresses phosphate reabsorption and 1,25-dihydroxyvitamin D synthesis in the kidney. The destabilized FGF23 is more susceptible to proteolytic cleavage, thereby reducing the secretion of biologically active intact FGF23. Persistent hyperphosphatemia due to decreased FGF23 concentrations leads to often large, ectopic calcific masses in soft tissues, which are usually removed surgically. However, the calcific masses often recur, requiring a more permanent solution to the problem. Nicotinamide is reported to lower serum phosphate by decreasing type IIb sodium-dependent phosphate cotransporter in the gut. This effect of nicotinamide has promoted its use in treatment of limited cases of tumoral calcinosis, though its effectiveness remains largely unclear. Therefore, we investigated nicotinamide as a potential therapy for tumoral calcinosis, using a murine model of the disease – Galnt3 knockout mice. Initially, nicotinamide (doses 0, 2.5, 5, 7.5, and 10 mmol/kg/day) was given to normal mice for 2 weeks. Treatment had no effect on serum phosphate levels; however, Fgf23 was decreased in a dose-dependent manner. Subsequently, high-dose nicotinamide (10 mmol/kg/day) was tested for 4 weeks in Galnt3 knockout mice on a high-phosphate diet. The radiographic data pretreatment and posttreatments showed that the treatment did not eliminate the calcification, but retarded its growth, while in the untreated mice, calcifications increased in size. The therapy did not change serum phosphate levels despite moderately increased phosphate excretion, likely due to decreased serum intact Fgf23 levels. Quantification of calcium and phosphate contents in hearts and kidneys revealed
549
that the treated mice had significantly high calcium in the heart. In summary, nicotinamide did not alter serum phosphate levels because the phosphate-lowering effect of nicotinamide was diminished by reduction in intact Fgf23 concentrations. The increased calcium in the heart suggests that nicotinamide therapy may also have an adverse effect.
Enchondroma resulting from loss of a Stk11-dependent switch of proliferative chondrocytes to a postmitotic fate . . . By Lick Pui Lai, PhD; University of Southern California, Los Angeles, CA
Stk11 (also known as liver kinase b1 (Lkb1)) is a serinethreonine protein kinase that acts upstream of the AMPactivated protein kinase (AMPK) family in coupling energy homeostasis to cell growth, proliferation and survival. Through chondrocyte-specific removal of Lkb1 activity, we showed that Stk11 is required for the normal switch of mitotic chondrocytes to a postmitotic hypertrophic chondrocyte fate. Consequently, it led to a dramatic overgrowth of the growth plate in the Stk11-mutant postnatal skeletal elements. To determine the molecular mechanisms underlying Stk11 action, we examined the mTOR pathway, which is inhibited through AMPK in growth regulation. Strikingly, rapamycin treatment of the pregnant mouse was able to rescue the delay in chondrocyte hypertrophy in Stk11mutant embryos, suggesting that Stk11 inhibition of mTOR signaling is critical for the switch in chondrocyte fate. Since the dramatic overgrowth of the growth plate is characteristic of enchondroma, we also examined the tumorigenicity of the mutant chondrocytes both in vitro and in vivo. In contrast to wild-type chondrocytes isolated from the postnatal day-30 growth plate, Stk11-mutant chondrocytes proliferated and formed colonies in monolayer and anchorage-independent agar cultures, indicative of a neoplastic transformation in vitro. Similarly, allotransplantation of mutant chondrocytes into immune-deficient NOG mice also resulted in tumor formation in vivo. Gene Ontology analysis of gene expression profiles indicated an augmented activity of cell proliferation and cell cycle regulators within the enchondroma-chondrocyte population compared to chondrocytes in the normal growth plate. Taken together, these data demonstrated an unexpected role of Stk11 in balancing proliferative and nonproliferative hypertrophic states of chondrocyte development through the regulation of mTOR signaling. Stk11 is a known tumor suppressor; our findings raise the possibility that loss of Stk11 may play a role in human enchondroma, a possibility that we are investigating.