Proliferation of 3.6Col1a1 Lineage Cells From 3-10 Days After Fracture is Essential for Normal Callus Formation in Mice

Introduction: Approximately 5-15% of fractures don't heal properly, depending on factors such as fracture location and severity. An impairment in fracture healing can lead to the formation of an atrophic nonunion, characterized by a failure in osteochondral callus formation. Atrophic nonunions are difficult to treat, as little is known of their mechanism and there’s a lack of a clinically relevant model. Proliferation of periosteal cells is known to be a key feature of fracture healing. We have previously shown that proliferation of periosteal cells during the first 2 weeks following a fracture is essential for normal healing. This was determined using mice expressing a thymidine kinase (TK+) ‘suicide gene’ driven by the 3.6Col1a1 promoter, which is active in osteoblast lineage periosteal cells. In this mouse model, proliferating TK+ cells die when dosed with ganciclovir (GCV), a guanosine mimetic that incorporates into the DNA of TK+ cells during replication and induces an apoptotic process, without affecting non-dividing cells or TK-negative cells. In this study, we sought to investigate the role of proliferating 3.6Col1a1-expressing cells, or early osteoblast lineage cells, in the early stages of fracture healing and determine their critical proliferation window. We used the 3.6Col1a1-TK model to study how fracture healing is affected when GCV is administered to ablate 3.6Col1a1-expressing cells for different durations and evaluating fracture healing at 21, 10, and 5 days post fracture. We hypothesize the critical window for osteoblast lineage cell proliferation lies within the first 10 days following fracture.

Materials and

Methods: To determine the role of proliferating 3.6Col1a1-expressing cells in fracture healing, we performed full femur fractures in 12-week-old TK+ mice and their wildtype (WT) littermates. A total of 90 mice were fractured using an established model of mid-diaphyseal femur fracture. The fracture surgeries consisted of placing a guidewire in the right femur marrow space, creating the fracture by applying a transverse force with a custom three-point bending set-up, and stabilizing the fracture with a stainless-steel tube inserted over the guidewire. Following fracture, mice were dosed with GCV daily for 3, 7, or 14 days, starting on the day of fracture, and healing was evaluated at 21 days post fracture (DPF) through radiographs, µCT and histology outcomes. Following initial results, we focused on evaluating healing at 5DPF and 10DPF to better understand the role of proliferating early osteoblast lineage cells in the first 10 days after fracture. In the 5DPF endpoint, mice were administered GCV daily for 3 or 5 days following full femur fracture, with fracture healing evaluated through histology. In the 10DPF endpoint, mice were administered GCV daily for 3, 5, 7, or 10 days following full femur fracture, with healing evaluated by µCT and histology outcomes. Weekly radiographs were performed to confirm fracture stability and monitor healing until euthanasia, when both the fractured femur and intact contralateral femur were dissected and fixed, scanned using µCT, and then embedded for histology.

Results, Conclusions, and Discussions: Radiographs, µCT and histology at 21DPF showed that TK+ mice receiving GCV for 3 days showed no significant differences in callus bone volume (Fig. 3) and callus composition compared to WT mice, forming similar calluses consisting mainly of woven bone (95% of callus area) with no cartilage and negligible fibrous tissue. In contrast, TK+ mice receiving 7 or 14 days of GCV dosing had impaired callus formation (Fig. 1) with significantly lower callus bone volume and callus area, forming calluses composed of significantly less woven bone (70 & 20%), more cartilage (10 & 15%), and more fibrous tissue (10 & 30%) than wildtype mice (Fig. 2). Histology at 5DPF showed GCV dosing for 3 or 5 days resulted in TK+ mice with decreased cell proliferation and slightly smaller fracture calluses composed of 15% less woven bone compared to WT mice (Fig. 1 & 2). µCT analysis at 10DPF showed a trend where TK+ mice developed smaller calluses with less mineralization as GCV dosing duration increased. Histology at 10DPF showed TK+ mice dosed with GCV for 3, 5, 7 or 10 days formed calluses with progressively decreasing cell proliferation and composed largely of cartilage (30%) and fibrous (40%) tissues, with only ~15% woven bone (Fig. 1 & 2). In contrast, WT mice formed larger calluses composed mostly of woven bone (>40%) with cartilage (25%) and fibrous (15%) tissues composing less of the fracture callus. The data found is consistent with our hypothesis stating the critical window for proliferation of osteoblast lineage periosteal cells lies within the first 10 days following fracture and ablation of proliferating 3.6Col1a1-TK cells during this window impairs fracture healing. The impairment to healing means these cells and their proliferation play a critical role in normal healing and in preventing atrophic nonunion formation. These results are important for having a greater understanding of atrophic nonunion fracture formation and this knowledge will also allow us to use the critical timeframe of proliferation of these cells in therapies to improve fracture healing.

Acknowledgements (Optional): Thanks to Robert Jilka and Charles O’Brien (UAMS) for sharing the 3.6Col1a1-TK mice. Thanks to Crystal Idleburg and Samantha Coleman for histological sections.