Abstract Bone marrow-derived mesenchymal stem cells (BM-MSCs) are multipotent stromal cells that have a critical role in the maintenance of skeletal tissues such as bone, cartilage, and the fat in bone marrow. In addition to providing microenvironmental support for hematopoietic processes, BM-MSCs can differentiate into various mesodermal lineages including osteoblast/osteocyte, chondrocyte, and adipocyte that are crucial for bone metabolism. While BM-MSCs have high cell-to-cell heterogeneity in gene expression, the cell subtypes that contribute to this heterogeneity in vivo in humans have not been characterized. To investigate the transcriptional diversity of BM-MSCs, we applied single-cell RNA sequencing (scRNA-seq) on freshly isolated CD271^+ BM-derived mononuclear cells (BM-MNCs) from two human subjects. We successfully identified LEPR^hiCD45^low BM-MSCs within the CD271^+ BM-MNC population, and further codified the BM-MSCs into distinct subpopulations corresponding to the osteogenic, chondrogenic, and adipogenic differentiation trajectories, as well as terminal-stage quiescent cells. Biological functional annotations of the transcriptomes suggest that osteoblast precursors induce angiogenesis coupled with osteogenesis, and chondrocyte precursors have the potential to differentiate into myocytes. We also discovered transcripts for several clusters of differentiation (CD) markers that were either highly expressed (e.g., CD167b, CD91, CD130 and CD118) or absent (e.g., CD74, CD217, CD148 and CD68) in BM-MSCs, representing potential novel markers for human BM-MSC purification. This study is the first systematic in vivo dissection of human BM-MSCs cell subtypes at the single-cell resolution, revealing an insight into the extent of their cellular heterogeneity and roles in maintaining bone homeostasis. Keywords: single-cell RNA sequencing (scRNA-seq), mesenchymal stem cell (MSC), bone marrow, osteogenesis, chondrogenesis, adipogenesis Introduction The human bone tissue is a complex system that consists of diverse cell types including osteoblast/osteocyte, osteoclast, and chondrocyte (collectively known as “bone cells”), together with various supporting cells such as adipocyte, fibroblast, and hematopoietic cells among others. A delicate balance of bone formation/resorption is critical for maintaining bone health, and therefore bone cells must work together to maintain bone strength and mineral homeostasis. Despite the extensive study of bone cells, their underlying biology remains poorly understood. While osteoclasts are of hematopoietic origin and derived from the “monocyte/macrophage-preosteoclast-osteoclast” differentiation trajectory [67]^1, the detailed origins of osteoblast/osteocyte and chondrocyte are not as well characterized. Currently, cells that give rise to osteoblast/osteocyte, chondrocyte, and adipocyte are generally referred to as “mesenchymal stromal/stem cells” (MSCs), which are non-hematopoietic bone marrow stromal cells with fibroblast colony-forming unit (CFU-F) and multi-differentiation capacity [68]^2. Typically, the human bone-marrow derived MSCs (BM-MSCs) are isolated with a combination of non-specific cell-surface markers such as high level of CD271, CD44, CD105, CD73, CD90, and low level/absence of CD45, CD34, CD14 or CD11b, CD79a or CD19, and human leukocyte antigen HLA-DR [69]^3^,[70]^4. Among these markers, CD271 shows great efficiency to sort MSCs either alone or in combination with negative selection of markers such as CD45 [71]^5^,[72]^6. Additionally, LEPR (leptin receptor, or CD295) is used for isolating BM-MSCs in transgenic labeling mice [73]^7^,[74]^8. Although these cell markers are candidates for isolating BM-MSCs, recent evidence suggests that the BM-MSCs are a heterogeneous group of cells for some cell markers. For instance, Akiyama et al. [75]^9 demonstrated that a small portion of BM-MSCs express CD45 and CD34, which are traditionally regarded as negative markers. Meanwhile, some studies also suggested that only around 50% of MSCs are positive for CD105 [76]^10^,[77]^11, a cell marker previously considered universally expressed by MSCs derived from different tissue [78]^12. The extent of the cellular heterogeneity among the BM-MSCs is not well-defined, although a few studies have proposed some novel subtypes. One study reported a subset of cultured mouse BM-MSCs that are distinct from regular BM-MSCs based upon differential attachment to plastic culture dishes, proliferation, and self-renewal patterns [79]^9. Another study examining cultured human BM-MSCs demonstrated that CD264 marks a subpopulation of aging human BM-MSCs with differential fibroblast colony forming efficiency [80]^13. Several other efforts have attempted to deconvolute the heterogeneity of BM-MSCs through the identification of the differentiation trajectory associated with a given subpopulation. For example, one study found that effective chondrocyte differentiation could only be induced in human MSCA-1^+CD56^+ BM-MSCs, while adipocytes are derived only from MSCA-1^+CD56^- BM-MSCs in vitro [81]^14. Another study identified “skeletal stem cells” in both mice and humans, which give rise to bone, stroma, and cartilage cells in vivo in mice, but not adipocytes or myocytes [82]^15^,[83]^16. Single-cell RNA sequencing (scRNA-seq) has recently emerged as a powerful approach to study cell heterogeneity in complex tissues. scRNA-seq measures transcriptional profiles of many cells at single-cell resolution, which can be clustered to distinguish and classify cell subtypes, infer developmental trajectories, and identify novel regulatory mechanisms [84]^17^,[85]^18. scRNA-seq technology represents a major advancement beyond the conventional bulk RNA-seq transcriptomics approach which attempts to infer biological mechanisms from average gene expression, weighted by the unknown proportions of unknown cell subtypes, across a heterogeneous cell population. Several studies have applied scRNA-seq to bone marrow stroma cells. However, these studies were either conducted in mice [86]^7^,[87]^19 or cultured cells from human subjects [88]^20^,[89]^21, which may not accurately represent the transcriptional profile of human primary BM-MSCs in vivo [90]^22^,[91]^23. Our current work is the first systematic scRNA-seq analysis of freshly isolated human CD271^+ bone marrow mononuclear cells (BM-MNCs). We have successfully identified LEPR^hiCD45^low BM-MSCs in the CD271^+ BM-MNC population and further revealed distinct subpopulations in LEPR^hiCD45^low BM-MSCs along with their differentiation relationships and functional characteristics. By comparing the expression pattern of LEPR^hiCD45^low BM-MSCs with CD45^hi hematopoietic cells, we have also identified several potential novel markers for human BM-MSC purification. Our findings provide significant insight into the identities and complexities of human BM-MSCs in vivo. Methods Study population The clinical study was approved by the Medical Ethics Committee of Central South University, and written informed consents were obtained from each participant. The study population consists of two Chinese subjects who underwent hip replacement surgery at the Xiangya Hospital of Central South University in 2019, including one 84-year-old male with osteoarthritis and normal bone mineral density (BMD; BMD T-score: -0.9 at lumbar vertebrae, 2.7 at total hip) and one 67-year-old female with osteoporosis (BMD T-score: -3.3 at lumbar vertebrae, -3.7 at total hip). Study participants were screened prior to surgery based on a detailed questionnaire, medical history, and a physical examination. Subjects were excluded from the study if they had preexisting chronic conditions which may influence bone metabolism including diabetes mellitus, renal failure, liver failure, hematologic diseases, disorders of the thyroid/parathyroid, malabsorption syndrome, malignant tumors, and previous pathologic fractures [92]^24. During hip replacement surgery, surgeons collected the bone marrow from the femoral shafts from each subject and transferred the samples to our laboratory immediately following the procedure. The samples were stored at 4 °C and processed within 24 hours after collection. Experimental animals Female C57BL/6J mice were purchased from Jackson Laboratory (Bar Harbor, ME, USA). All mice were housed in pathogen-free conditions and fed with autoclaved food, and all experimental procedures were approved by the Ethics Committee of Xiangya Hospital of Central South University. BMD measurement BMD (g/cm^2) at the lumbar spine (L1-L4) and the total hip (femoral neck and trochanter) were measured with a duel energy x-ray absorptiometry (DXA) fan-beam bone densitometer (Hologic QDR 4500A, Hologic, Inc., Bedford, MA, USA). According to the World Health Organization definition [93]^25 and the BMD reference established for Chinese populations [94]^26, subjects with a BMD of 2.5 SDs lower than the peak mean of the same gender (T-score ≤ -2.5) were determined to be osteoporotic, while subjects with -2.5 < T-score < -1 are classified as having osteopenia and subjects with T-score > -1.0 are considered healthy. Human bone marrow cell dissociation Bone marrow derived mononuclear cells (BM-MNCs) were extracted from the marrow cavity of femoral shafts using a widely applied dissociation protocol 5,6]. Briefly, the bone marrow was attenuated with PBS (1:2) and mixed gently. The mixture was then equally layered onto equal volume of Ficoll (GE health care, Chicago, IL, USA), and the buffy coat was isolated by centrifugation (440 g, 35 min, 4 °C). The separated buffy coat was transferred into a new 15 ml centrifuge tube and washed with PBS. After discarding the supernatant, red blood cells were lysed with RBC Lysis Buffer (Thermo Fisher, Waltham, MA, USA). After washing twice with PBS, the remaining MNCs were further purified with CD271 magnetic MicroBeads (Miltenyi Biotec, Bergisch Gladbach, Germany) for positive selection [95]^6. Positive selection of human CD271^+ BM-MNC BM-MNCs were incubated for 10 min at 4-8 °C with monoclonal antibody (mAb) against CD271. After washing, the cells were incubated with anti-IgG1 immunomagnetic beads for 15 min at 4 °C. The cell suspension was placed on a column in a cell separator (Miltenyi Biotec), and the positive fraction was subjected to a second separation step. The cells were then counted and assessed for viability with a Countstar® Rigel S3 fluorescence cell analyzer (ALIT Life Science Co., Ltd, Shanghai, China). Isolation of murine BM-MSCs Cells were isolated from flushed bone marrow from female C57BL/6 mice (8 weeks) and dissociated using 21G needle. Cells were then plated in 75-cm^2 cell culture flasks containing 10 mL of MesenCult^TM basal expansion medium with 10× Supplement (Stemcell, Vancouver, Canada), 100 U/mL penicillin-streptomycin, L-glutamine 2 mM, and incubate at 37 °C 5% CO[2] for one week. 0.1% MesenPure^TM (Stemcell) was added for the depletion of CD45^+ cells. Stromal cells were allowed to reach 80%-90% confluency before passage or planting. Flow cytometry Cells were resuspended in 100 μL of staining medium, followed by staining with fluorochrome-conjugated antibodies on ice for 20 minutes. The antibodies used in this study to identify MSCs were anti-CD45-FITC (eBioscience, clone 30-F11, 0.5 µg/test), anti-Ter119-FITC (eBioscience, clone TER-119, 0.25 µg/test), anti-CD31-FITC (eBioscience, clone 390, 1 µg/test), and anti-CD56-PE (R&D Systems, clone # 809220, 0.5 µg/test). Cells were analyzed on a Sony MA900 Cell Sorter, where CD45/Ter119/CD31^- cells were identified as BM-MSCs, and CD56 was used to separate CD56^+ and CD56^- cell subtypes. Bone sectioning, immunostaining, and confocal imaging Freshly dissected bones were fixed in 4% paraformaldehyde overnight, followed by decalcification in 10% EDTA for 1 week, and then dehydrated using a series of graded ethanol and embedded in paraffin. Samples were then cut into 5-µm-thick longitudinally oriented sections, deparaffinized in xylene, and rehydrated in decreasing concentrations of ethanol followed by distilled water. After deparaffinization and antigen retrieval, sections were blocked in PBS with 5% bovine serum albumin (BSA) for 1 hour and then stained overnight with the following primary antibodies: goat-anti-LepR (R&D: AF497, 10 µg/mL) and rabbit-anti-CD56 (Proteintech: 14255-1-AP, 1:2000). Next, samples were incubated with appropriate secondary antibodies, including donkey anti-goat Alexa Fluor 555 and donkey anti-rabbit Alexa Fluor 647 (all from Invitrogen, 1:400). Slides were mounted with anti-fade prolong gold (Invitrogen) and images were acquired with a Zeiss LSM780 confocal microscope. Cell capture and cDNA synthesis After isolation of human CD271+ BM-MNCs, we applied the Chromium single cell gene expression platform (10x Genomics, Pleasanton, CA, USA) for scRNA-seq experiments. Cell suspensions were loaded on a Chromium Single Cell Controller (10x Genomics) to generate single-cell gel beads in emulsion (GEMs) by using Single Cell 3' Library and Gel Bead Kit V3 (10x Genomics, Cat# 1000092) and Chromium Single Cell A Chip Kit (10x Genomics, Cat#120236) according to the manufacturer's protocol. Briefly, single cells were suspended in 0.04% BSA-PBS. Cells were added to each channel, captured cells were lysed, and the released RNA were barcoded through reverse transcription in individual GEMs^27. GEMs were reverse transcribed in a C1000 Touch Thermal Cycler (Bio Rad, Hercules, CA, USA) programmed at 53 °C for 45 min, 85 °C for 5 min, and held at 4 °C. After reverse transcription, single-cell droplets were broken, and the single-strand cDNAs were isolated and cleaned with Cleanup Mix containing DynaBeads (Thermo Fisher Scientific). cDNAs were generated and amplified, and the quality was assessed using the Agilent 4200. Single cell RNA-Seq library preparation Single-cell RNA-seq libraries were prepared using Single Cell 3' Library Gel Bead Kit V3 following the manufacturer's guide ([96]https://support.10xgenomics.com/single-cell-gene-expression/librar y-prep/doc/user-guide-chromium-single-cell-3-reagent-kits-user-guide-v3 -chemistry). Single Cell 3' Libraries contain the P5 and P7 primers used in Illumina bridge amplification PCR. The 10x Barcode and Read 1 (primer site for sequencing read 1) were added to the molecules during the GEM-RT incubation. The P5 primer, Read 2 (primer site for sequencing read 2), Sample Index and P7 primer were added during library construction. The protocol was designed to support library construction from a wide range of cDNA amplification yields spanning from 2 ng to >2 μg without modification. Finally, sequencing was performed on an Illumina Novaseq6000 with a sequencing depth of at least 100,000 reads per cell for a 150 bp paired end (PE150) run. Pre-processing of scRNA-seq data Raw FASTQ files were mapped to the Reference genome (GRCh38/hg38) using Cell Ranger 3.0 ([97]https://support.10xgenomics.com/single-cell-gene-expression/softwa re/pipelines/latest/what-is-cell-ranger). To create Cell Ranger-compatible reference genomes, the references were rebuilt