Additional file 2 of Single-cell profiling reveals periosteal signatures of impaired periosteal cells proliferation in a drill-hole model of type 2 diabetes

Supplementary Material 2: Figure S1. Reduced bone mass in 12-week-old db/db mice. A. Body weight (left) and fasting blood glucose (right) of db/db and db/+ mice at different age. One-way ANOVA, Sidak’s multiple comparison test. Data are shown as mean ± SD. n = 6 (db/+ and db/db, 8- and 12-week-old); n = 3 (db/+ and db/db, 16-week-old). B. Representative stereomicroscopy (left) and 3D μCT images (right) of femurs from db/db and db/+ mice at 12-week-old. Scale bar: 2 mm. C. Femur length of the db/db and db/+ mice at 12- week-old age. Two-tailed Student’s t-test. Data are shown as mean ± SD. n = 6 (db/+); n = 8 (db/db). D. 3D in vivo μCT images of 1 mm cortical bone segments in the mid-diaphysis of the db/db and db/+ femurs at different ages. Scale bar: 2 mm. E. cortical bone area (Ct.ar), total tissue area (Tt.ar) and cortical thickness (Ct.th) of 1 mm cortical bone segments in the mid-diaphysis of the db/db and db/+ femurs at different ages. Two-way ANOVA, Sidak’s multiple comparison test. Data are shown as mean ± SD. n = 3. F. Hematoxylin and eosin staining of the distal femurs from db/db and db/+ mice. Scale bar: 1 mm. *P < 0.05, **P < 0.01, ***P < 0.001. Figure S2. Flow Cytometry and Gene Expression Analysis of Periosteal Cells. A. The flow cytometry gating strategy involved the use of an anti-CD45 antibody to exclude hematopoietic lineage cells. B-I. Vlnplot displaying the expression of key marker genes across different periosteal cell subpopulations identified in the analysis. Subpopulations include proliferating cells, endothelial cells (ECs), Schwann cells, chondrocytes, muscle satellite cells (MuSCs), pericytes, myocytes, IECs and tenocytes. J. UMAP for periosteum from db/+ and db/db. Figure S3. Time-Dependent Changes in the Periosteum Following Drill Hole Surgery. A-F. Panels A to F show the sequential changes in tibias at day 0, 3, 5, 7, 14, and 21 post-drill hole surgery, including: Representative 2D μCT images of horizontal sections of the tibias. Scale bar: 1 mm; Corresponding 3D μCT images of the regions of interest. Scale bar: 500 μm; Hematoxylin and eosin staining; Masson staining; Safranin O staining; Tartrate-resistant acid phosphatase staining. Scale bar: 500 μm and 200 μm G. Quantification of newly forming bone mass at the fracture sites as shown in the 3D μCT images, at 0, 3, 5, 7, 14, and 21 days post-fracture. Statistical analysis was performed using one-way ANOVA with Sidak’s multiple comparison test. Data are presented as mean ± SD, with n = 6. H. Representative 3D μCT images of fractured tibias at different time points post-surgery. Scale bar: 1 mm. Figure S4. Impaired callus remodeling and decreased osteogenic activity in db/db mice during fracture healing. A. Representative micro-CT images of fracture callus at day 5 and day 7 post-fracture in db/+ and db/db mice. Scale bars: 500 µm. B. H&E-stained histological sections of fracture callus at day 14 and day 21 post-fracture. Scale bars: 100 µm. C. Representative dynamic bone formation images using double labeling with alizarin red (red) and calcein (green) in callus regions at day 7 post-fracture. D. Quantification of mineral apposition rate (MAR, µm/day) Data are shown as mean ± SD. ***p < 0.001. E. TRAP-stained sections at day 7 post-fracture showing osteoclasts in callus area. F. Quantification of osteoclast surface per bone surface (Oc.S/BS, %). Data are shown as mean ± SD. ***p < 0.001. Figure S5. Cell Type Composition and Marker Gene Expression in Periosteal Cells. A-G. Vlnplot displaying the expression levels of selected marker genes across various periosteal cell subpopulations. Subpopulations include muscle satellite cells (MuSCs), Pericytes, Schwann cells, Proliferating cells, tenocytes, Endothelial cells (ECs), Chondrocytes. Figure S6. Integrated Single-Cell RNA-Seq Analysis of Periosteal Cells Across Multiple Experimental Conditions. A. UMAP visualizations of unsupervised clustering of periosteal cells from various conditions, including db/+, db/+ injured, db/db, db/db injured, Col2-Cre, Col2-Cre fracture Day 5, Col2-Cre fracture Day 10. Clusters are color-coded and labeled, showing distinct subpopulations across different experimental conditions. B. Heatmap displaying differential gene expression across the identified periosteal cell clusters from the UMAP analysis. Key marker genes associated with specific cell types or functions are highlighted on the right. Figure S7. scRNA-Seq Analysis of Distinct Cellular Dynamics in Bone Repair: Comparing Intramembranous and Endochondral Ossification. A-G. Top: UMAP visualizations of unsupervised clustering of periosteal cells from various conditions, including db/+, db/+ injured, db/db, db/db injured, Col2-Cre, Col2-Cre fracture Day 5, Col2-Cre fracture Day 10, Clusters are color-coded and labeled. H. Dot plot showing the expression levels of key marker genes across different cell types, including FAPs, injury induced fibroblasts (IIFCs), chondrocytes (CHs), PCs, and proliferating cell cluster (Mki67+). I-K. Feature plots showing the expression of representative marker genes (Acan, Postn, and Acta2) across the UMAP space, highlighting distinct gene expression patterns. Figure S8. Effect of FGL2 on Osteoblast Proliferation Under Normal and High Glucose Conditions. A. Osteoblasts were treated with vehicle or varying concentrations of FGL2 (50, 200 ng/mL) under normal (vehicle) and high glucose (4.5 mg/mL) conditions, and relative cell proliferation was assessed after 48 h to compare the effects under these different conditions. Statistical significance was assessed using one-way ANOVA with Sidak’s multiple comparison test. Data are shown as mean ± SD. **P < 0.01, ***P < 0.001. B. RT-qPCR results showing the mRNA levels of Raptor following knockout using CRISPR-Cas9 lentivirus. Statistical significance was determined using a two-tailed Student’s t-test. Data are shown as mean ± SD. **P < 0.01.

补充材料2：图S1 12周龄db/db小鼠骨量减少。A. 不同周龄db/db与db/+小鼠的体重（左）及空腹血糖（右）。采用单因素方差分析（one-way ANOVA）结合西达克多重比较检验（Sidak’s multiple comparison test）。数据以均值±标准差（mean±SD）表示，样本量n=6（db/+与db/db小鼠，8、12周龄组）；n=3（db/+与db/db小鼠，16周龄组）。B. 12周龄db/db与db/+小鼠股骨的体视显微镜成像图（左）及三维显微计算机断层扫描（micro computed tomography, μCT）成像图（右）。比例尺：2 mm。C. 12周龄db/db与db/+小鼠的股骨长度。采用双尾Student t检验。数据以均值±标准差表示，n=6（db/+组）；n=8（db/db组）。D. 不同周龄db/db与db/+小鼠股骨干中段1 mm皮质骨节段的体内三维μCT成像图。比例尺：2 mm。E. 不同周龄db/db与db/+小鼠股骨干中段1 mm皮质骨节段的皮质骨面积（cortical bone area, Ct.ar）、总组织面积（total tissue area, Tt.ar）及皮质骨厚度（cortical thickness, Ct.th）。采用双因素方差分析（two-way ANOVA）结合西达克多重比较检验。数据以均值±标准差表示，n=3。F. db/db与db/+小鼠远端股骨的苏木精-伊红（hematoxylin and eosin, H&E）染色图。*P<0.05，**P<0.01，***P<0.001。

图S2 骨膜细胞的流式细胞术（flow cytometry）与基因表达分析。A. 流式细胞术设门策略：使用抗CD45抗体排除造血系细胞。B-I. 小提琴图（Vlnplot）展示本分析鉴定的不同骨膜细胞亚群中关键标记基因的表达情况，亚群包括增殖细胞、内皮细胞（endothelial cells, ECs）、施万细胞（Schwann cells）、软骨细胞、肌肉卫星细胞（muscle satellite cells, MuSCs）、周细胞、肌细胞、肠上皮细胞（intestinal epithelial cells, IECs）及腱细胞。J. db/+与db/db小鼠骨膜的均匀流形近似与投影（Uniform Manifold Approximation and Projection, UMAP）可视化图。

图S3 钻孔术后骨膜的时间依赖性变化。A-F. 面板A至F分别展示钻孔术后第0、3、5、7、14及21天胫骨的序列变化，包括：胫骨水平切片的代表性二维μCT成像图（比例尺：1 mm）；对应感兴趣区域的三维μCT成像图（比例尺：500 μm）；苏木精-伊红染色图；马松三色染色（Masson staining）图；番红O染色（Safranin O staining）图；抗酒石酸酸性磷酸酶（tartrate-resistant acid phosphatase, TRAP）染色图（比例尺：500 μm及200 μm）。G. 骨折术后0、3、5、7、14及21天，基于三维μCT成像图量化骨折部位新形成骨量。统计分析采用单因素方差分析结合西达克多重比较检验。数据以均值±标准差表示，n=6。H. 术后不同时间点骨折胫骨的代表性三维μCT成像图。比例尺：1 mm。

图S4 db/db小鼠骨折愈合过程中骨痂重塑受损及成骨活性降低。A. db/+与db/db小鼠骨折术后第5天及第7天骨折骨痂的代表性显微CT成像图。比例尺：500 μm。B. 骨折术后第14天及第21天骨折骨痂的苏木精-伊红染色组织学切片。比例尺：100 μm。C. 骨折术后第7天骨痂区域使用茜素红（红色）与钙黄绿素（绿色）双标记的代表性动态骨形成成像图。D. 矿化沉积速率（mineral apposition rate, MAR, µm/day）的量化结果。数据以均值±标准差表示，***P<0.001。E. 骨折术后第7天抗酒石酸酸性磷酸酶染色切片，显示骨痂区域的破骨细胞。F. 每骨表面破骨细胞表面占比（osteoclast surface per bone surface, Oc.S/BS, %）的量化结果。数据以均值±标准差表示，***P<0.001。

图S5 骨膜细胞的细胞类型组成与标记基因表达。A-G. 小提琴图展示不同骨膜细胞亚群中选定标记基因的表达水平，亚群包括肌肉卫星细胞（MuSCs）、周细胞、施万细胞、增殖细胞、腱细胞、内皮细胞（ECs）及软骨细胞。

图S6 多实验条件下骨膜细胞的整合单细胞RNA测序分析。A. 来自不同实验条件（包括db/+、db/+损伤组、db/db、db/db损伤组、Col2-Cre、Col2-Cre骨折术后第5天、Col2-Cre骨折术后第10天）的骨膜细胞无监督聚类的UMAP可视化图。聚类以颜色编码并标注，显示不同实验条件下的不同细胞亚群。B. 热图展示本研究UMAP分析鉴定的不同骨膜细胞聚类间的差异基因表达，右侧高亮标注与特定细胞类型或功能相关的关键标记基因。

图S7 骨修复中不同细胞动态的单细胞RNA测序分析：比较膜内成骨与软骨内成骨。A-G. 上图：来自不同实验条件（包括db/+、db/+损伤组、db/db、db/db损伤组、Col2-Cre、Col2-Cre骨折术后第5天、Col2-Cre骨折术后第10天）的骨膜细胞无监督聚类的UMAP可视化图，聚类以颜色编码并标注。H. 点图展示不同细胞类型中关键标记基因的表达水平，包括脂肪祖细胞（fibro-adipogenic progenitors, FAPs）、损伤诱导成纤维细胞（injury-induced fibroblasts, IIFCs）、软骨细胞（chondrocytes, CHs）、浆细胞（plasma cells, PCs）及增殖细胞簇（Mki67+）。I-K. 特征图展示代表性标记基因（Acan、Postn及Acta2）在UMAP空间中的表达情况，凸显不同的基因表达模式。

图S8 纤维蛋白原样蛋白2（fibrinogen-like protein 2, FGL2）对正常及高糖条件下成骨细胞增殖的影响。A. 成骨细胞在正常（溶剂对照）及高糖（4.5 mg/mL）条件下分别用溶剂或不同浓度FGL2（50、200 ng/mL）处理，48小时后评估相对细胞增殖率，以比较不同条件下的效应。统计显著性采用单因素方差分析结合西达克多重比较检验。数据以均值±标准差表示，**P<0.01，***P<0.001。B. 实时定量聚合酶链反应（real-time quantitative polymerase chain reaction, RT-qPCR）结果显示成簇规律间隔短回文重复序列相关蛋白9（clustered regularly interspaced short palindromic repeats-associated protein 9, CRISPR-Cas9）慢病毒敲除Raptor后，其mRNA水平的变化。统计显著性采用双尾Student t检验。数据以均值±标准差表示，**P<0.01。