Short summary

In this cross-sectional clinical study, we will examine the bones of 111 Type 1 Diabetes (T1D) patients and 37 age-matched healthy controls with the aim of describing a T1D Bone Phenotype. The main Objectives of the study is a) to determine if the material properties of the bones are affected in diabetic bone disease and b) to determine if the mitochondrial function in osteoclasts and osteoblasts is impaired in T1D.

Rationale

Evidence supports that T1D is complicated by lower bone mass[3] and impaired bone microarchitecture and strength[4]. This results in an increased fracture risk[5] including a 5-times higher hip fracture risk[6]. Besides costs associated with treatment, fractures carry substantial morbidity and mortality, emphasising the relevance of addressing bone disease in T1D. Bone is a dynamic organ that is remodelled throughout life. Previously formed bone is resorbed by osteoclasts (OC) and new bone is formed by osteoblasts (OB) in a coupled and highly regulated process known as bone remodelling. Although the mechanisms whereby T1D exerts deleterious effects on bone are not entirely clear, hyperglycaemia and hypoinsulinaemia as observed in T1D may influence bone cells and ultimately bone mass and strength. Thus, glucose is essential for OB and OC differentiation and activity.[7] While Glut 1 and 3 expressions are stabile during osteoblast differentiation; Glut 4 expression is increased during transition from pre-osteoblasts to mature OBs.[8] OBs mainly metabolize glucose to lactate through oxidative glycolysis[9] despite presence of mitochondria that exhibit intact oxidative phosphorylation.[10] Mitochondrial respiration in OBs also relies on glutamine and fatty acids, which are essential for normal osteoblast activity.[11, 12] By contrast, glycolysis, lactate production and oxidative phosphorylation increase during osteoclastogenesis[13], which is promoted by pyruvate but decreased by inhibition of oxidative phosphorylation indicating that mitochondrial respiration is key to osteoclastogenesis.[13] The activity of mature OCs is energy demanding as bone tissue is dissolved by use of ATP-dependent H+-pumps and removal of collagen by proteases and reactive oxygen species (ROS).[14] These processes are fuelled primarily through glucose metabolisation.[15] By contrast, hyperglycaemia inhibits bone cell activity. Thus, hyperglycaemia promotes differentiation of mesenchymal stem cells (MSCs) to adipocytes rather than OBs, reduces Wnt signalling, effectively limiting osteoblastogenesis and bone formation[16-20], and impairs OC differentiation and function[21]. In addition, hyperglycaemia stimulates expression of sclerostin, which is an endogenous inhibitor of Wnt signalling and subsequent bone formation that is secreted by osteocytes.[20] Hyperglycaemia causes mitochondrial dysfunction with decreased ATP production in kidney cells[22] and increases ROS and decreases respiration in adipose tissue-derived MSCs[23]. This indicates that T1D is complicated by mitochondrial dysfunction in bone cells, which may inhibit bone remodelling, mass and strength. In addition, insulin promotes OB proliferation, normalises bone remodelling in animal models and increases bone mass in T1D patients[24-27] indicating that insulinopenia may have deleterious effects on bone in T1D. The effects of insulin on bone cell metabolism including mitochondrial function are largely unknown, and data on bone cell metabolism in T1D patients are unavailable. Levels of circulating markers of bone formation and resorption are lower and sclerostin levels higher in T1D,[28] lending credence to the notion that bone remodelling and bone formation in particular are compromised in T1D. Histological studies of bone tissue in adult T1D patients have demonstrated both normal and impaired bone formation, but these studies have included a mere ≈20 bone biopsies, failed to include well-characterised controls and assessment of cortical bone remodelling[29, 30]. Multiple factors may contribute to development of bone disease including disease-related complications such as microvascular disease and accumulation of advanced glycation endproducts[31, 32]. However, current data support that changes in bone cell metabolism including mitochondrial function contributes to development of bone disease in T1D. With the emergence of pharmacological opportunities of restoring mitochondrial function such as NAD+ modulators, bone cell metabolism has the potential to emerge as a treatment target in T1D patients with bone disease.

Description of the cohort

Type 1 Diabetic subjects recruited from out-patient clinics and internet pages such as Facebook. Controls from community i.e. through advertisement on internet sites such as Facebook

Data and biological material

The clinical investigation will include 111 male and female patients aged 18-80 years with childhood-onset T1D and 37 controls. Inclusion criteria (cases): Insulin-only treated male and female T1D patients aged 18-80 years with documented autoimmune diabetes and insulinopenia (C-peptide < 200 pmol L-1) and HbA1c 42-80 mmol mol 1. Exclusion criteria: Diseases or treatments considered to influence bone (e.g. stage 3 CKD or higher). Inclusion criteria, controls: Similar age-, sex- and BMI. Exclusion criteria: Any disease (except mild conditions such as mild hypertension). To characterize the skeletal phenotype in T1D, we will measure areal bone mineral density (BMD) at the hip and spine (by dual X-ray absorptiometry (DXA)), volumetric BMD and bone microarchitecture and geometry at the distal radius and tibia (by high-resolution peripheral quantitative CT scan (HR-pQCT)[4] and bone strength by use of microindentation (Osteoprobe) in T1D patients and controls. In addition, information on glucose control and diabetes-related complications will be retrieved in T1D patients to determine if these variables are determinants of the bone phenotype. In addition, we will investigate if skin levels of advanced glycated endproducts (AGE, AGE Reader), biochemical markers of bone turnover including sclerostin and circulating miRNAs reflect the bone phenotype in T1D patients. Following bone labelling with oral tetracycline treatment, an 8 mm trans iliac bone biopsy will be performed in 24 cases and 12 controls. Bone specimens will be plastic embedded and used for histomorphometry to determine cortical and trabecular bone remodelling in T1D to establish the T1D bone phenotype at the histological level[33]. These samples will also be used to calculate parameters for trabecular and cortical bone, micro-finite element analysis to assess the mechanical properties of the bone and ultra-high resolution measurements to characterize the number, density and connection between osteocytes in cortical bone[34]. In addition, using immunohistochemistry and imaging, we will investigate if the number and morphology of mitochondria in bone cells T1D patients are impaired. Bone cell metabolism in T1D A bone marrow aspirate will be collected in conjunction with the bone biopsy and used to compare the characteristics of OB and OCs in T1D patients and controls. Following isolation and culturing, we will characterise mesenchymal stromal cells (MSCs) including presence of surface markers that identify the cell type (e.g. CD44, DC73) and reflects their metabolic activity (e.g. expression of the insulin receptor, GLUT4) and their ability to proliferate and differentiate to OBs and adipocytes in vitro using established methods[35], which will clarify if the lineage commitment of MSCs to osteoblasts is defective in T1D, which could explain decreased bone formation. Monocytes isolated from the bone marrow will be used to study osteoclastogenesis and in vitro bone resorption activity (resorption assays) to evaluate if osteoclasts are dysfunctional in T1D, explaining in part decreased bone remodelling in T1D. As bone remodeling in vivo depends on OB-OC interactions, in vitro co-cultures of these cells will be used to study defects in bone cell communication using in-house methods.[36] Single cell RNA sequencing will be used to demonstrate gene signatures of MSCs, OBs and OCs to determine differences in cell complexity and heterogeneity as well as cell lineage relationships (MSCs) and differences in pathway expression (e.g. bone- and metabolism-related pathways such as Wnt and insulin signalling) between T1D patients and controls to determine if the cell composition and gene expression demonstrate differences that correspond to lower bone cell differentiation, activity and metabolism. Furthermore, using cryosectioned bone tissue, we will use spatial transcriptomics[37] to determine if transcriptional profiles of individual cells depend on the localisation in bone. This will demonstrate if bone disease in T1D is related to differences in expression of bone-specific loci of the bone. In addition, proteomics will be used to assess if OB and OC-secreted factors including factors known to couple OB and OC activity (e.g. sphingosine-1) differ between T1D and controls indicating impaired cell communication. Following the characterisation of bone cells in T1D, we will investigate the metabolic profile of MSCs, mature OBs and OCs using extracellular flux analyses (Agilent Seahorse). For these experiments we will use different concentrations of energy sources such as glucose to map mitochondrial function and bone cell metabolism including the metabolic flexibility in T1D. Proteomics data will be used to study levels of ≈500 mitochondria-related proteins[38] in OB and OC cell cultures to demonstrate if T1D associates with impaired levels of proteins related to mitochondrial function. In addition, we will investigate if the mitochondrial function relates to expression of genes involved in bone cell activity (e.g. Wnt- and RANK signalling) and expression of miRNA. Finally, in situ hybridization and immunohistochemistry will be used to further validate the effects of the cellular changes in the bone tissue using established methods[33].

Collaborating researchers and departments

Medical dep., Hospital of South West Jutland

• Jeppe Gram

Medical dep., Odense University Hospital, Odense

• Stinus Hansen

KMEB Laboratory

• Sebastian Zanner

Steno Diabetes Center Odense

• Kurt Højlund

Steno Diabetes Center Aarhus

Dep. of Endocrinology, Aarhus University Hospital

• Torben Harsløf

Steno Diabetes Center Nordjylland

• Peter Vestergaard
• Inge Brandt