Fig 1: Knockout of monocytic CTSD attenuated the ischemic‐induced brain damage in diabetic mice and reduced TBI‐induced neutrophil infiltration.The CTSDmKO mice were fed with high‐fat diet for 6 weeks and intraperitoneally injected with 3 doses of streptozotocin (50 mg/kg body weight), with littermate CTSDloxP/+ mice as control mice. The mice fed with normal diet and injected with vehicle solution were served as controls. A, The concentration of CTSD in the plasma of mice were examined by ELISA. Data were expressed as mean±SEM (n=6 for each group), 2‐way ANOVA with Tukey’s multiple comparisons test. B–D, The mice were subjected to a 45‐minute occlusion of the middle cerebral artery followed by reperfusion. The brain slices were obtained at 72 hours reperfusion for TTC staining (B). Scale bar, 7.5 mm (B, left). The infarct volume was measured and plotted. Data were expressed as mean±SEM (n=6 for each group), 2‐way ANOVA with Tukey’s multiple comparisons test (B, right). The neurological deficit score was assessed using the Longa’s method at 72 hours reperfusion. Data were expressed as mean±SEM (n=10 for vehicle groups, n=18 for streptozotocin groups), 2‐way ANOVA with Tukey’s multiple comparisons test (C). Kaplan–Meier survival curves of the mice after MCAO were plotted (n=12 for Vehicle groups, n=18 for streptozotocingroups) (D). E, F, The mice were subjected to a 45‐minute occlusion of the middle cerebral artery followed by reperfusion. The mice treated with the surgical procedures except the blockage of the artery were used as sham‐operated controls. The time taken for mice to reach the floor in pole test were recorded before and after MCAO. Data were presented as mean±SEM. (n=6 for each group), RM 2‐way ANOVA (E). The mice were subjected to rotarod test to record their latency times to fall from the rotating rod before and after MCAO. Data were presented as mean±SEM (n=6 for each group), RM 2‐way ANOVA (F). G, The WT mice were exposed to TBI with sham‐operated mice as controls. Then the blood samples were harvested at D1, D3, and D7 post TBI. The concentration of CTSD in the plasma of mice were examined by ELISA. Data were expressed as mean±SEM (n=6 for each group), 1‐way ANOVA with Tukey’s multiple comparisons test. H, The CTSDmKO and CTSDloxp/+ mice were exposed to TBI with sham‐operated mice as controls. The concentration of CTSD in the plasma of mice were examined by ELISA. Data were expressed as mean±SEM (n=6 for each group), 2‐way ANOVA with Tukey’s multiple comparisons test. I, The CTSDmKO and CTSDloxp/+ mice were exposed to TBI with sham‐operated mice as controls. Then the brains were harvested at D3 post TBI and the slices were prepared. The brain slices were immunostained with CD31 antibody (red) together with antibody against MPO (green). The representative images from the cortex surrounding the injured region were presented. Arrows indicate the extravascular leukocytes. Arrowheads indicates the leukocytes within the vascular wall. Scale bar, 50 μm (left). The extravascular and intravascular MPO+ neutrophils were counted for quantification analysis (right). Data were presented as mean±SEM (n=6 per group, 2‐way ANOVA followed by Bonferroni’s multiple comparisons test). J, K, The CTSDmKO and CTSDloxp/+ mice were exposed to TBI with sham‐operated mice as controls. Then the mice were subjected to the Pole test and the time for each mouse to descend a vertical pole (50 cm in length) were recorded (J) or subjected to Rotarod test and the time for each mouse to remain on a rotating rod (20 rpm) were recorded (K) at indicated time points. Data were presented as mean±SEM (n=6 per group, RM 2‐way ANOVA with Tukey’s multiple comparisons test). CTSD indicates cathepsin D; MCAO, middle cerebral artery occlusion; mKO, MYPN knockout; MPO, myeloperoxidase; RM, repeated measures; STZ, streptozotocin; and TBI, traumatic brain injury.
Fig 2: Pro‐CTSD upregulated VCAM‐1 in brain endothelial cells for neutrophil transmigration. A, The plasma of mice was collected and examined by Western blot using antibodies of human CTSD and mouse CTSD, respectively, with the CBB gel as loading control. The recombinant human pro‐CTSD protein (rh pro‐CTSD) was loaded as a positive control. B, The upregulated genes in the brain endothelial cells of hCTSDhi mice compared with littermate WT control mice were subjected to module gene network and hub genes analysis. The size more than 5 genes and fold change >0.3 were presented. C, Violin plot showing the transcriptional levels of VCAM‐1 in brain endothelial cells in the scRNA‐seq data from WT control mice (4003 cells from 3 mice) and hCTSDhi mice (5157 cells from 3 mice). (Student’s t test). (D) The tSNE plots were presented to show the transcriptional expression of VCAM‐1 in the endothelial cell subset of the scRNA‐seq data in WT (4003 cells) and hCTSDhi (5157 cells) mice. The expression levels were indicated by color code according to the scale. E, The scRNA‐seq data of endothelial cells were further processed for pseudo‐time analysis to show the expression of VCAM‐1 and PECAM1 in pseudo‐time plot. F, The mice brain slices were immunostained with CD31 antibody (yellow), ACTA2 antibody (red), and VCAM‐1 antibody (green). DAPI (blue) was used for counterstaining. The representative images from 6 independent experiments were presented. Scale bar, 50 μm (left). The fluorescence intensity of VCAM‐1 was measured for quantification. Data were presented as mean±SEM (n=6 per group, 2‐way ANOVA with Tukey’s multiple comparisons test). G, Schematic drawings of the experimental timeline in H–L. H, The 4‐week‐old hCTSDhi mice were injected with GFP‐tagged AAV‐BR1 virus containing VCAM‐1 shRNA, with empty vector as control. Then the brains were harvested 8 weeks later and the brain slices were prepared for immunostaining with antibody recognizing VCAM‐1 (red) and CD31 (yellow), respectively. DAPI (blue) was used for counterstaining. Representative images from 6 independent experiments were presented. Scale bar, 10 μm (left). The relative fluorescence intensity of VCAM‐1 within the CD31‐positive vascular region was measured for quantification. Data were presented as mean±SEM (n=6 per group, 2‐way ANOVA with Tukey’s multiple comparisons) (right). I, J, The anesthetized mice were exposed to TBI injury. Then the brains were harvested at D3 and the slices were prepared. The brain slices were immunostained with CD31 antibody (yellow) and antibody against MPO (red). The slices were mounted and visualized by confocal microscopy. The representative images at the peri‐injury region were presented. Arrows and arrowheads indicate the extravascular and intracellular MPO+ neutrophils, respectively. Scale bar, 50 μm. The extravascular and intravascular MPO+ neutrophils were counted for quantification. Data were presented as mean±SEM (n=6 per group, 2‐way ANOVA with Tukey’s multiple comparisons) (I). The brain slices were immunostained with NeuN antibody. Representative images from the cortex surrounding the injured region were presented. Scale bar, 50 μm (up, J). The NeuN+ cells were counted for quantification. Data were presented as mean±SEM (n=6 per group, 2‐way ANOVA with Tukey’s multiple comparisons) (down, J). K, L, The 4‐week‐old hCTSDhi mice were injected with GFP‐tagged AAV‐BR1 virus containing VCAM‐1 shRNA, with empty vector as control. After TBI, the motor performance of the mice at indicated time points was assessed by the Pole test (K) and Rotarod test (L), respectively. Data were presented as mean±SEM (n=12 per group, RM 2‐way ANOVA with Tukey’s multiple comparisons test). A indicates artery; AAV, adenovirus; ACTA2, actin alpha 2; Cap, capillary; CBB, Coomassie blue; CTSD, cathepsin D; EC, endothelial cell; GFP, green fluorescent protein; GO, Gene Ontology; MPO, myeloperoxidase; NeuN, neuronal nuclear protein; PECAM‐1, platelet endothelial cell adhesion molecule‐1; RM, repeated measures; TBI, traumatic brain injury; tSNE, t‐distributed stochastic neighbor embedding; V, vein; VCAM‐1, vascular cell adhesion molecule 1; and WT, wild type.
Fig 3: Analysis of leukocytes infiltration in the brain of hCTSDhi mice upon traumatic injury and after ischemia/reperfusion injury. A–D, The anesthetized mice were exposed to TBI using controlled cortical impact, with sham‐operated mice as controls. Then the brains were harvested at indicated time points and the slices were prepared. The brain slices were immunostained with CD31 antibody (red) together with antibody against MPO (green) (A) or Ly6c (green) (C). DAPI (blue) was used for counterstaining. The slices were mounted and visualized by confocal microscopy. The representative images from the cortex surrounding the injured region were presented. Arrows indicate the extravascular leukocytes. Arrowheads indicates the leukocytes within the vascular wall. Scale bar, 50 μm. The extravascular and intravascular MPO+ neutrophils (B) and Ly6c+ monocytes (D) were counted for quantification analysis. Data were presented as mean±SEM. (n=6 per group, 2‐way ANOVA followed by Bonferroni’s multiple comparisons test). E, F, The anesthetized mice were exposed to ischemia/reperfusion injury. Then the brains were harvested 3 days later (D3) and the slices were prepared. The brain slices were immunostained with CD31 antibody (red) together with antibody against MPO (green) E. DAPI was used for counterstaining. Representative confocal images at the peri‐infarct region from 6 independent experiments were presented. Arrows indicate the extravascular leukocytes. Arrowheads indicates the leukocytes within the vascular wall. Scale bar, 50 μm. The extravascular and intravascular MPO+ neutrophils were counted for quantification (F). Data were presented as mean±SEM. (n=6 per group, 2‐way ANOVA followed by Bonferroni’s multiple comparisons test). CTSD indicates cathepsin D; MCAO, middle cerebral artery occlusion; MPO, myeloperoxidase; TBI, traumatic brain injury; and WT, wild type.
Fig 4: hCTSDhi mice exhibited worse functional recovery after TBI injury and ischemic‐ reperfusion injury. A, B, The anesthetized mice were exposed to TBI by controlled cortical impact, with sham‐operated mice as controls. Then the mice were subjected to Pole test and the time for each mouse to descend a vertical pole (50 cm in length) were recorded (A) or subjected to Rotarod test and the time for each mouse to remain on a rotating rod (20 rpm) were recorded (B) at indicated time points. Data were presented as mean±SEM (n=12 per group, RM 2‐way ANOVA with Tukey’s multiple comparisons test). C–E, The mice were subjected to 45 minutes of middle cerebral artery occlusion followed by reperfusion, with sham‐operated mice as controls. The neurological deficit score was assessed using the Longa’s method at 72‐hour reperfusion. Data were expressed as mean±SEM (n=15 for each group), Student’s t test (C). Then the mice were subjected to the Pole test and the time for each mouse to descend a vertical pole were recorded (D) or subjected to Rotarod test and the time for each mouse to remain on a rotating rod were recorded (E) at indicated time points. Data were presented as mean±SEM (n=8 per group, RM 2‐way ANOVA with Tukey’s multiple comparisons test). CTSD indicates cathepsin D; MCAO, middle cerebral artery occlusion; RM, repeated measures; TBI, traumatic brain injury; and WT, wild type.
Fig 5: Single‐cell RNA sequencing analysis of brain endothelial cells of hCTSDhi mice. A, After anesthesia, 3‐month‐old hCTSDhi mice were perfused with India ink buffer, with littermate WT mice as controls. The extracted brains were carefully removed from the skull, and the cortical vessels was imaged with stereomicroscope. Scale bar, 1 mm (left). Vascular density was calculated and plotted (right). Data were presented as mean±SEM. ns, no statistical significance (n=6 per group, 2‐tailed Student’s t test). B, The brain slices from 3‐month‐old hCTSDhi mice were prepared, with littermate WT mice as controls. The slices were stained with CD31 (red) and examined with confocal microscopy. Scale bar, 500 μm (left). The vascular density was calculated and plotted (right). Data were presented as mean±SEM (n=6 per group, 2‐tailed Student’s t test). C, D, The freshly isolated mice brain was processed and subjected to scRNA‐seq. The tSNE of the single‐cell expression profiles were presented (34 282 cells from 3 WT mice; 31 866 cells from 3 hCTSDhi mice) to show the identified cell clusters (C). The tSNE plot of endothelial cell subpopulations from the mice brain was presented (4003 cells from 3 WT mice; 5157 cells from 3 hCTSDhi mice) (D). E, F, The differentially expressed genes of endothelial cells in hCTSDhi mice compared with WT mice were analyzed. The upregulated genes were subjected to GO enrichment analysis and the GO terms showing statistical significance were presented (E). The differentially expressed genes were subjected to gene set enrichment analysis analysis and the GO terms showing statistical significance were shown (F). ASC indicates astrocytes; CPC, choroid plexus cells; CTSD, cathepsin D; DC, dendritic cells; EC, endothelial cells; EPC, ependymal cells; FDR, false‐discovery rate; GO, Gene Oncology; Hb_VC, hemoglobin‐expressing vascular cells; ImmN, immature neurons; MAG, macrophages; MG, microglial cells; NendC, neuroendocrine cells; NES, normalized enrichment score; NRP, neuronal‐restricted precursors; OLG, oligodendrocytes; OPC, oligodendrocyte progenitor cells; PC, pericytes; tSNE, t‐distributed stochastic neighbor embedding; VSMC, vascular smooth muscle cells; and WT, wild type.
Supplier Page from Abcam for Mouse Cathepsin D ELISA Kit