Role of neutrophil extracellular traps in delayed fracture healing of type 2 diabetics

Abstract

Diabetes is a global disease with a strongly increasing prevalence and incidence. In Germany, about 9 million people are living with diabetes, most of them with type 2. Type 2 diabetes impairs vascularization, nerve signal transfer, and kidney function. Additionally, type 2 diabetic patients have impaired healing abilities: Wounds and fractures need more time to heal, and complications occur more often. After a fracture, the immune system is rapidly activated in the fracture gap. The induced inflammatory process is essential for fracture healing, starting with the arrival of neutrophils. Neutrophils have strong phagocytic abilities and can release large amounts of cytokines, thus inducing strong inflammation. They can also release their DNA as a defense mechanism, which then organizes in large net-like structures covered with antimicrobial proteins and proteases—so-called neutrophil extracellular traps (NETs). One of the main proteins involved in NET formation is peptidyl arginine deiminase type IV (PAD4), which is responsible for the induction of chromatin decondensation. NETs have been shown to impair wound healing in diabetic mice and are released after trauma. The negative effect on wound healing could be improved by PAD4 knockout or DNase treatment. In this study, the possible role of NETs in diabetic fracture healing was investigated. To investigate NET release in diabetic conditions, neutrophils were isolated from patients or healthy volunteers, and NET release was measured by using the Sytox Green Assay. Neutrophils from healthy volunteer were stimulated with controlled diabetic conditions in vitro (high glucose, high insulin). Reaction to different stimulants (phorbol 12-myristate 13-acetate [PMA], calcium ionophore A23187 [CI], lipopolysaccharide [LPS], H2O2) was investigated in diabetic conditions. Variants of PADI4 and their effect on NET release were analyzed. Intracellular processes were analyzed by western blot, and reactive oxygen species (ROS) and myeloperoxidase (MPO) activity were measured. Isolated NETs from stimulated neutrophils were added to a mesenchymal stem cell (MSC) line (SCP-1 cells), and viability (mitochondrial activity, total protein, lactate dehydrogenase release) and functional parameters (migration, osteogenicdifferentiation) were analyzed. The reaction of monocytes to isolated NETs was analyzed by measuring activation and viability. Neutrophils from diabetic patients released more NETs at baseline and after stimulation with PMA or H2O2 but not CI. They had increased PAD4 levels, but the MPO and citrullinated histone H3 levels were not altered. Neutrophils from healthy volunteers did not show NET release in response to high glucose or high glucose with insulin. Combination of diabetic conditions with CI did not alter NET release, but high insulin delayed NET release in response to PMA or LPS by 2-3 h (+3.26 h with PMA, +2.09 h with LPS). Insulin induced high levels of ROS with and without PMA whereas high glucose did not induce ROS. Further, insulin with PMA significantly induced two mitogen-activated proteins kinases, with increased phosphor-extracellular signal-regulated kinase and phospo-p38, but slightly reduced MPO activity. In a PADI4 minor variant haplotype, NET formation was significantly accelerated, and PAD4 levels increased. Isolated NETs were very toxic to SCP-1 cells, and migration was already reduced at non-toxic concentrations of NETs. Single-dose exposure of NETs at the beginning of osteogenic differentiation significantly reduced alkaline phosphatase activity and matrix formation. Recovery from NET exposure regarding viability could be achieved between days 10 and 14 of differentiation. NETs activated Toll-like receptor 4. Monocytes were activated at lower NET doses, and their viability was decreased with higher NET concentrations. In conclusion, NET formation in diabetic patients was increased, but not because of high glucose. Insulin had a strong regulatory function on NET formation, possibly deregulating NET formation. A minor PADI4 haplotype could increase the deregulation in type 2 diabetic patients. The increased NET formation could then have a strong negative effect on MSCs and contribute to delayed fracture healing in type 2 diabetic patients.