Reply: Questioning the existence of Aif1 +, Csf1r + mesenchymal stem cells

Abstract

We thank Dr. Hume for his interest in our Hepatology article.Dr. Hume argued that colony stimulating factor 1 receptor (CSF1R) and allograft inflammatory factor 1 (AIF1) are specifically expressed in mononuclear phagocytes, whereas we and others have shown that CSF1R and AIF1 are expressed elsewhere in addition to mononuclear phagocytes. Poudel et al. demonstrate that IL-34 induces epithelial cell transformation through CSF1R. [1] The CSF1/CSF1R axis was involved in the regulation of osteoblast proliferation and differentiation. [2] Furthermore, AIF-1 was associated with inflammation in cells other than macrophages, such as endothelial and peritoneal mesothelial cells. [3] A similar debate was focused on the discussion of Kupffer cells contaminated with liver sinusoidal endothelial cells. It was concluded that "researchers need to go beyond the concept of absolute lineage markers and address data with an open mind."Additionally, Dr. Hume believed that AIF1 + CSF1R + MSCs (mesenchymal stem cells) should be defined as PDGFRβ + macrophages, whereas no evidence supports this notion in the literature. Conversely, our study identified mesenchymal stem-cell-like cells (MSCLCs) by wellrecognized markers (such as Cd44, Nt5e [CD73], and others) often used to identify MSCs. We noticed a unique MSCLC subpopulation that, in addition to expressing common MSC markers, also highly expressed inflammation-related genes Csf1r, Aif1, and Cd74. We also ruled out the interference of cell doublets by DoubleTFinder in R. Furthermore, we validated our results by multicolor immunohistochemical staining and confirmed this MSC subgroup as AIF1 + CSF1R + MSCs. In fact, the gene expression profiles of AIF1 + CSF1R + MSCs exhibited similar patterns as MSCs and stromal cells, rather than that of macrophages.Finally, Dr. Hume states that our AIF1 + CSF1R + MSCs may be an artifact attributable to contamination by macrophage disaggregation. Though single-cell sequencing technology may have certain technical defects, we believe that if there is a contamination of fragmented macrophages, the majority of cells from the tissue should be affected as shown, [4] not just in this MSC subpopulation. It should be noted that MSCs and macrophages closely interact in tissue niches, thus their transcriptional profiles or surface markers could have partially converged, which may explain why some markers were shared, as found in our study and others. [5] In any case, subsequent studies using single-cell sequencing technology should pay attention to detection methods, such as single-nuclei sequencing and imaging flow cytometry, to further verify single-cell RNA-sequencing-based findings. CONFLICTS OF INTERESTNothing to report.