Identification of Novel Sub-Populations of Resident and Inflammatory Myeloid Cells and Osteogenic Progenitor Cells in Musculoskeletal Trauma
Michael Sorkin, MD, Amanda K. Huber, PhD, Charles Hwang, Noelle D. Visser, MS, Rajasree Menon, PhD, Benjamin Levi.
University of Michigan, Ann Arbor, MI, USA.
PURPOSE: Despite the high frequency of musculoskeletal extremity trauma, little is known about the dynamics of the complex inflammatory response that can result in pathologic healing in the form of heterotopic ossification (HO). While posttraumatic inflammation is a critical element of normal wound healing, aberrant inflammatory processes in response to musculoskeletal trauma have been described to initiate ectopic bone formation through endochondral ossification. Therefore, accurate identification of cells present at the injury site is critical to understanding the pathophysiology and cellular interactions between inflammatory cells and resident mesenchymal progenitors (MSC). Here, we utilize single cell RNA sequencing to obtain an unbiased analysis of the cellular composition at the injury site and to identify dynamic changes in these cell subpopulations over time in a model of HO. METHODS: Male C57BL/6 mice were subjected to a musculoskeletal extremity trauma model of HO formation involving a 30% total body surface area dorsal burn and Achilles tenotomy (Fig A). Single cell RNA sequencing (10X genomics) and downstream unsupervised clustering analyses were performed on the extremity injury HO site at baseline (D0), during inflammation (D3), and mesenchymal progenitor cell condensation (D7) (n=3/group). RESULTS: Canonical correlation analysis yielded 14 transcriptionally unique cell clusters identifiable at the injury site with characteristic profiles attributable to phenotypically distinct cell types (Fig B). While recruited granulocytes significantly reduced in numbers from 36.1% on day 3 to only 10.7% at day 7, macrophages/monocytes and dendritic cells were the predominant cell populations and constituted over 35% of total cells on day 7. Subpopulation analysis revealed distinct monocyte/macrophage clusters with M2 characteristic gene expression including Mrc1 (CD206), H2-Eb1 (MHC II) and Cd163 as well as Arg1 (cluster 1, 3, 5 and 8). While cluster 1 showed expression of all three markers, cluster 3 showed high expression of Arg1 but not Cd163 indicating that these are phenotypically unique cell populations with distinct functions (Fig. D). We further observed a significant increase in HO progenitor subpopulations (cluster 2, 4, and 6) on day 7 which were almost entirely absent on day 3. Clusters 2, 4, 6 and 11 showed high expression of Pdgfra, a marker known to identify mesenchymal progenitor cells (Fig. E). Cluster 11 completely disappeared on day 7, suggesting that these cells likely differentiated into HO forming cells. Interestingly, mesenchymal cell clusters further demonstrated high expression of chondrogenic differentiation genes Acan and Sox9 as early as on day 3 indicating early cell fate determination (Fig. F). CONCLUSION: To elucidate HO pathophysiology, it is critical to characterize the intricate interactions of inflammatory cells and progenitor cells. Using single cell RNA sequencing as a novel tool, we identify the presence of yet unidentified and functionally distinct subpopulations of monocytes and progenitor cells with unique transcriptional characteristics. Isolation of cells based on these findings may allow for a more granular understanding of HO and future design of targeted treatment.
Back to 2019 Abstracts