Development of Novel Therapies, Models, and Biomarkers for Osteoclast-Related Diseases

Abstract: Osteoclasts are derived from hematopoietic stem cells (HSCs) and form upon stimulation of osteoclast precursors by macrophage colony-stimulating factor (M-CSF) and receptor activator of nuclear factor κΒ ligand (RANKL). Osteoclasts resorb bone by secreting hydrochloric acid, matrix metalloproteinases (MMPs) and cathepsin K. Failure of bone resorption or bone formation results in metabolic bone diseases. Osteopetrosis (excess bone) is caused by mutations impairing osteoclastogenesis or osteoclast function, and can only be treated with a hematopoietic stem cell (HSC) transplant (HSCT). Infantile malignant osteopetrosis (IMO) is typically lethal without HSCT. 50% of IMO patients have mutations in TCIRG1, a component of the osteoclast V-ATPase that is crucial for acidification. The first aim of this dissertation was to develop a clinically applicable TCIRG1-targeting gene therapy, to circumvent the need for HSC donors, and assess its efficacy and safety.Osteoclasts have also been implicated as mediators of bone and cartilage degradation in rheumatic diseases. The increased bone degradation in rheumatoid arthritis and osteoarthritis has been strongly linked to increased osteoclast presence and activity, but the role of osteoclasts in the pathological degradation of cartilage—and type II collagen, its main component—remains unclear. Therefore, the second aim of this dissertation was to develop novel models and biomarkers for the investigation of osteoclast-mediated cartilage degradation.In Paper I, we demonstrated that a proof-of-concept lentiviral vector, with a viral promoter, used to transduce IMO HSCs only induces TCIRG1 expression in osteoclasts, not in precursors or macrophages, and that a small percentage of treated cells is sufficient to restore resorption. In Paper II, we developed a clinically applicable vector, with a human promoter, and demonstrated that transduced human IMO HSCs transplanted into mice maintained the ability to generate functional osteoclasts ex vivo long-term. These studies are promising progress for the development of IMO gene therapy. In Paper III, we developed a mouse model with human HSCs expressing human M-CSF—to bypass the incompatibility of the human M-CSF receptor and mouse M-CSF—by transducing the HSCs with a human M-CSF vector and transplanting them into immunodeficient mice. These mice had improved human monocyte and myeloid reconstitution, but human osteoclasts could not be detected.In Paper IV, we developed a cell culture model of human osteoclasts resorbing cartilage; using protease inhibitors and the C2M biomarker of type II collagen degradation, we demonstrated that cartilage resorption was mediated by MMPs and, to some extent, cathepsin K. In Paper V, we developed the type II collagen degradation biomarker assay GPDPLQ1237, targeting an elongated version of the CTX-II neo-epitope that has been speculated to be osteoclast- and cathepsin K-specific. Both MMPs and cathepsin K contributed to GPDPLQ1237 release from cartilage resorption, but only MMPs contributed to CTX-II release. Pro-inflammatory stimulation of cartilage without osteoclasts also resulted in MMP-mediated GPDPLQ1237 and CTX-II release. Hence, GPDPLQ1237 is a multi-protease biomarker of osteoclast- and inflammation-mediated cartilage degradation. However, GPDPLQ1237 could not be detected at sufficient levels in blood or urine from humans or rats to validate the assay in vivo. These studies highlight the potential contribution of osteoclastic cartilage degradation to rheumatic diseases, and that novel biomarkers of cartilage degradation covering a wider range of cartilage degradation processes are needed to better understand relevant disease mechanisms.