CORD-19:9fb5ca87ca47d35d87161158ae37fc0ca642c51e JSONTXT 8 Projects

A Mac2-positive progenitor-like microglial population survives independent of CSF1R signaling in adult mouse brain Abstract 10 Microglia are the primary innate immune cells in the CNS, capable of mounting inflammatory 29 responses and phagocytosis. They can be distinguished from other CNS cell types by the 30 distinctive ramified morphology and expression of common myeloid markers including Cd11b 31 and ionized calcium binding adaptor molecule 1 (Iba1). In addition to immune functions, microglia 32 carry out a multitude of neurotrophic functions during CNS development and homeostasis 33 (Kierdorf and Prinz, 2017) . Microglia also play critical pathological roles in a wide spectrum of 34 neurodegenerative conditions, including Alzheimer's disease, Parkinson's disease, Huntington's 35 disease, and Amyotrophic lateral sclerosis (Hickman et al., 2018) . A number of disease genes were 36 found to be highly expressed in microglia (Hansen et (Haynes et al., 2006) . Interestingly, 51 we recently discovered that adult newborn microglia follow a similar maturation path (Zhan et al., 52 2019), suggesting that the developmental plasticity of microglia in the adult brain might be an 53 underlying feature of microglial homeostasis (Santambrogio et al., 2001) . Unlike other tissue 54 myeloid populations such as monocytes and macrophages, the resident microglial pool receives 55 no significant replenishment from circulation and is internally maintained by self-renewal ( To examine the transcriptome profiles of the Csf1r inhibitor-resistant microglial population, we 86 performed scRNA-seq on the remaining microglia from C57/BL6J mice that were treated with 87 PLX diet (D0) (Fig. 1A) . To investigate early stage adult newborn microglia, we also included 88 mice that were switched to a control diet for 2 days after PLX treatment (D2) (Fig 1A) . Microglia 89 from age-matched non-treated mice were used as control (Ctrl) (Fig 1A) . Similar to our previous 90 results (Zhan et al., 2019) , oral dosing of 1200 mg/kg PLX in C57/BL6J mice for two weeks 91 resulted 92.93 -96.2% removal of CD11b+ myeloid cells in the CNS (Fig S1) . Single cell 92 suspensions were stained with CD11b antibody for myeloid population purification via 93 fluorescence activated cell sorting (FACS) (Fig S1) , followed by the 10x genomics single cell 94 Ctrl brains (Fig 2H) . Iba1 was used as a microglial marker. The fluorescence intensity of P2ry12 119 and Tmem119 per Iba1+ cell was significantly decreased in D0 microglia compared with Ctrl 120 microglia (Fig 2I, J) . 121 122 We next applied differentially expressed gene (DEG) analysis to obtain the transcriptomic profile 124 of each cluster. DEGs were selected based on log fold-change ratio either above 0.5 or below -0.5 125 (Fig 3A and Table S1 ). Notably, while some of the marker genes were exclusively expressed in 126 certain clusters, many were shared among multiple clusters. Among the most upregulated genes, 127 each cluster could be identified with distinctive markers (Fig 3B) . 128 In order to analyze the cell populations from each cluster in more detail, we manually compared 129 the most prominently expressed markers from each cluster with the tabula-muris dataset, a single-130 cell atlas that contains annotation data for 100,000 cells from 20 different organs and tissues 131 (Tabula Muris et al., 2018) ( Fig 3C) . Cluster-1 and Cluster-2 resembled homeostatic microglia, 132 with Cluster-1 expressing high-levels of homoeostatic genes including P2ry12, Tmem119 and 133 Trem2. Cluster-2 showed a similar homeostatic gene signature but also exhibited a modest increase 134 of NF-kB target genes including Ccl3, Ccl4, and Tnf (Fig. 3C ). Microglia in Cluster-3 expressed 135 a high level of Mrc1, which encodes the meningeal macrophage marker CD206 (Goldmann et al., 136 2016), and apolipoprotein E, the major risk factor in AD. Remarkably, this population was 137 completely wiped out by PLX treatment, as Cluster-3 was absent in both the D0 and D2 groups 138 ( Fig 1H) . Immunohistochemistry analysis validated this observation, showing complete loss of 139 CD206+ meningeal macrophages after 2 weeks of PLX treatment (Fig 3D) . 140 Despite reduced expression of homeostatic microglial signatures, Clusters-4, -5, -8, -9 141 expressed myeloid markers as expected (Fig. S3A) , with transcriptional profiles that resembled 142 monocytes Fig S3B) ; granulocytes Fig S3B) ; and dendritic cells 8, Fig S3F) . Specifically, cells in Cluster-8 expressed CD209a, and MHC genes such as H2-Ab1, 144 H2-Eb1, and CD74, a cell surface receptor for the cytokine macrophage migration inhibitory factor 145 (MIF). Immunohistochemistry analysis showed over-representation of CD74-positive microglia in 146 D0 brains, confirming that this cell population is resistant to csf1r inhibitors (Fig. 3E ). In Cluster-147 9, the top-10 most upregulated genes were involved in mitosis, including Marker of Proliferation 148 Ki-67 (Mki-67), DNA Topoisomerase 2a (Top2a), Cyclin A2 (Ccna2), and Cyclin B2 (Ccnb2) 149 ( Fig 3C) . Consistent with this population representing proliferative microglia, the relative 150 frequency of cells in Cluster 9 roughly doubled its size from 30% (D0) to 69% in just two days of 151 microglial repopulation (Fig 1N) . 152 Cells in Cluster-6 and Cluster-7 lacked myeloid markers including Cd11b, Csf1r, and Cd14 did 153 not (Fig. S3A) . Instead, Cluster-6 showed high expression of Klrb1c and Nkg7, which are normally 154 found in natural killer (NK) cells (Fig S3D) , while Cluster-7 expressed T-cell markers Ccr7 and 155 Il7r ( Fig S3E) . We then performed microglial lineage mapping using the Cx3cr1-CreERT2 and 156 inducible RFP reporter, and showed that microglia under Csf1r inhibition did not express either 157 the T cell marker CD3 or the NK cell marker Nkg7 (Fig. S4) . Thus, it is unlikely that these are 158 microglia with altered states under Csf1r inhibition. Instead, it is more likely that certain subsets 159 of NK and memory T cells were captured by cd11b-based capture during FACS since they can proliferative, and PLX-resistant ( Fig 3F) . As expected, Clusters-6 and -7 were placed on the PLX-166 resistant branch (Fig 3G) , because lymphocytes were not affected by PLX5622 (Wheeler et al., 167 2018 ). Among the myeloid clusters, Cluster-8 was evenly distributed along the PLX-resistant and 168 proliferative branches, while Clusters-4 and -5 were largely distributed along the PLX-resistant 169 branch ( Fig 3G) . 170 To spatially map the underlying PLX-resistant microglia from Cluster-4 and -5, we next searched 173 for markers that distinguish them via immunohistochemistry. Marker gene analysis in identified Mac2 antigen, also known as Galectin 3 (Lgals3), as a potential marker (Fig 4A) . Lgals3 175 was highly expressed in Clusters-4, -5, -8, and -9 (Fig 4B, C) . Immunofluorescence staining of 176 brain tissues from C57/BL6J mice that were treated with PLX diet for 2 weeks (D0) revealed a 177 subset of Iba1+ microglia expressing Mac2 in both Ctrl and D0 groups (Fig 4D) , with increased 178 Mac2 intensity and more extensive ramification in the D0 mice. After 2 weeks of PLX treatment, 179 while the majority of Iba+ cells was depleted, showing 88% loss (Fig. 4E ), the number of 180 Iba1+Mac2+ cells was modestly increased (D0) in the hippocampus ( Fig 4F) . Notably The copyright holder for this preprint (which was not peer-reviewed) is the author/funder. It . https://doi.org/10.1101/722090 doi: bioRxiv preprint repopulation day 4 (D4) (Fig 5I) , accounting for ~5% of total Iba1+ microglia ( Fig 5J) . Roughly 210 50% of the Iba1+Mac2+ cells were EdU+ at repopulation Day 4 ( Fig 5K) . Their highly 211 proliferative property was further supported by the presence of mitotic marker Ki67 (Fig S5) . 212 Notably, newborn cells in the Iba1+Mac2+ population failed to retain Mac2+ expression over long 213 term, and the number of Iba1+Mac2+ cells dropped back to control level after repopulation day 214 14 (Fig 5I) . 215 216 To further characterize the Mac2+ microglial population, we selected the Mac2+ cells from our 218 scRNA-seq dataset based Lgals3 expression one standard deviation above the average log UMI 219 counts. A total of 504 cells were manually distinguished and binned together as the Mac2+ cluster 220 ( Fig 6A) . Overlay of the Mac2+ cells on the tSNE plot showed that they were largely derived from 221 Clusters -4, -5, -8, -9, with only a few sprinkled in the homeostatic Clusters-1 and -2 ( Fig 6A) . To 222 investigate the gene signatures in Mac2+ cells, we next performed DEG analysis in comparison to 223 homeostatic Cluster-1 ( Fig 3C) . DEGs were identified that showed log fold-change ratio above 0. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder. It . https://doi.org/10.1101/722090 doi: bioRxiv preprint notion that a Csf1r-independent microglial population in the adult CNS might reflect a similar 291 paradigm to the developmental progenitors in which alternative survival pathways are engaged. 292 In characterizing the remaining microglia after acute Csf1r inhibition, we recovered a subset 293 of cells that can be distinguished by the Mac2 antigen that was also detectable under homeostatic 294 conditions. Unlike the vast majority of microglial population that is sensitive to Csf1r inhibitor, 295 the Mac2+ population modestly increased in response to the drug (Fig 4F) . The copyright holder for this preprint (which was not peer-reviewed) is the author/funder. It . https://doi.org/10.1101/722090 doi: bioRxiv preprint removed; 3) cells that have greater than 10% mitochondrial genes were removed; (Supplementary 405 figure 2 a-d) . In addition, we also removed genes that showed expression in less than 10 cells. 406 After data filtering, we obtained a total 2,100 cells for subsequent analysis (Fig 1c) . The data were 407 normalized by log transformation followed by regression based on total UMI counts and 408 visualization was achieved with R package ggplot2 (Wickham, 2009 The copyright holder for this preprint (which was not peer-reviewed) is the author/funder. It . https://doi.org/10.1101/722090 doi: bioRxiv preprint The copyright holder for this preprint (which was not peer-reviewed) is the author/funder. It . https://doi.org/10.1101/722090 doi: bioRxiv preprint 694 . CC-BY-NC-ND 4.0 International license is made available under a The copyright holder for this preprint (which was not peer-reviewed) is the author/funder. It . https://doi.org/10.1101/722090 doi: bioRxiv preprint The copyright holder for this preprint (which was not peer-reviewed) is the author/funder. It . https://doi.org/10.1101/722090 doi: bioRxiv preprint 709 . CC-BY-NC-ND 4.0 International license is made available under a The copyright holder for this preprint (which was not peer-reviewed) is the author/funder. It . https://doi.org/10.1101/722090 doi: bioRxiv preprint 725 . CC-BY-NC-ND 4.0 International license is made available under a The copyright holder for this preprint (which was not peer-reviewed) is the author/funder. It . https://doi.org/10.1101/722090 doi: bioRxiv preprint 742 . CC-BY-NC-ND 4.0 International license is made available under a The copyright holder for this preprint (which was not peer-reviewed) is the author/funder. It . https://doi.org/10.1101/722090 doi: bioRxiv preprint 771 . CC-BY-NC-ND 4.0 International license is made available under a The copyright holder for this preprint (which was not peer-reviewed) is the author/funder. It . https://doi.org/10.1101/722090 doi: bioRxiv preprint The copyright holder for this preprint (which was not peer-reviewed) is the author/funder. It . https://doi.org/10.1101/722090 doi: bioRxiv preprint