Increased Ccl5 RNA expression and decreased miR-324-5p expression in the peri-infarct cortex in MCAO mice. (A) Illustration of animal experimental protocol. (B) Schematic of the brain showing CC, CP, IP, IC region of the cerebral cortex. The shadow part indicated the ischemic region. (C and D) QRT-PCR analysis of Ccl5 mRNA expression (C) and miR-324-5p expression (D) in the cortex on D1, D3 and D7 after MCAO (n=5). The data were normalized by the sham mice; *p<0.05, **p<0.01 and ***p<0.001 by unpaired two-tailed Student’s t test.

Intracerebral CCL5 antibody delivery ameliorated ischemic stroke injury. (A) ELISA analysis of CCL5 concentration in the cerebral cortex of BSA, CCL5 antibody, rCCL5, and MVC treated mice on D3 and D7 after MCAO (n=3). (B and C) Representative TTC stained serial sections (B) and infarct volume (C) in brains from BSA, CCL5 antibody, rCCL5, and MVC treated mice on D3 after surgery (n=6). Scale bar: 1cm. (D) Longa scoring of BSA, CCL5 antibody, rCCL5 treated mice, and sham control from D0 to D7 after surgery (n=12). (E) Rotarod test of BSA, CCL5 antibody and rCCL5 treated mice, and sham control mice on D0, D3 and D7 from surgery (n=7). (F) Representative images of immunofluorescent double labeling of GFAP and IBA1 in the IP cortical region of BSA, CCL5 antibody and rCCL5 treated mice, and in the CP cortical region of BSA treated mice on D3 and D7 after MCAO. Scale bar: 50 μm. (G and H) The surface area of GFAP+ activated astrocytes in the IP region of BSA, CCL5 antibody and rCCL5 treated mice on D3 (G) and D7 (H) after MCAO (n=10). (I and J) The surface area of IBA1+ microglia cells in the IP region of BSA, CCL5 antibody and rCCL5 treated mice on D3 (I) and D7 (J) after MCAO (n=10). (K) Representative schematic showing dendritic branch structure in the IP cortical region of BSA, CCL5 antibody and rCCL5 treated mice, and in the CP cortical region of BSA treated mice on D7 after MCAO. Scale bar: 20 μm. (L) Sholl analysis of the dendritic branch structure in the IP region of BSA, CCL5 antibody and rCCL5 treated mice on D7 after MCAO (n=10). (M) Representative images of basal dendritic spines in the IP region of BSA, CCL5 antibody and rCCL5 treated mice, and in the CP cortical region of BSA treated mice on D7 after MCAO. Mushroom spines marked by arrows, stubby spines marked by black arrowheads and filopodial-like thin spines marked by white arrowheads. Scale bar: 10 μm. (N and O) Quantification of the total spine density (N) and mushroom spine density (O) in the IP and CP region of BSA, CCL5 antibody and rCCL5 treated mice on D7 after MCAO (n=10). The composition of mushroom spines, stubby spines and thin spines within the total dendritic spine is indicated; *p<0.05, **p<0.01 and ***p<0.001 by one-way ANOVA with Tukey’s post-hoc test. In A, *p<0.05, **p<0.01 and ***p<0.001 by unpaired two-tailed Student’s t test. In D, E and L, *p<0.05, **p<0.01 and ***p<0.001 vs. BSA group, ##p<0.01 and ###p<0.001 vs. rCCL5 group by two-way ANOVA with Tukey’s post-hoc test.

Intracerebral miR-324-5p agomir delivery ameliorated ischemic stroke injury. (A) ELISA analysis of CCL5 concentration in the cerebral cortex of NC agomir, miR-324-5p agomir and miR-324- 5p antagomir treated mice on D3 and D7 after MCAO (n=4). (B and C) Representative TTC stained serial sections (B) and infarct volume (C) in brains from sham control mice, NC agomir, miR-324-5p agomir and miR-324-5p antagomir treated mice on D3 after surgery (n=6). Scale bar: 1cm. (D) Longa scoring of NC agomir, miR-324-5p agomir and miR-324-5p antagomir treated mice, and sham control from D0 to D7 after surgery (n=15). (E) Rotarod test of NC agomir, miR-324-5p agomir and miR-324- 5p antagomir treated mice, and sham control on D0, D3 and D7 from surgery (n=7). (F) Representative images of immunofluorescent double labeling of GFAP and IBA1 in the IP cortical region of NC agomir, miR-324-5p agomir and miR-324-5p antagomir treated mice, and in the CP cortical region of NC agomir treated mice on D3 and D7 after MCAO. Scale bar: 20 μm. (G and H) The number of GFAP+ activated astrocytes per 105 μm2 in the IP and CP region of NC agomir, miR-324-5p agomir and miR-324-5p antagomir treated mice on D3 (G) and D7 (H) after MCAO (n=6). (I and J) The surface area of GFAP+ activated astrocytes in the IP region of NC agomir, miR-324-5p agomir and miR-324-5p antagomir treated mice on D3 (I) and D7 (J) after MCAO (n=10). (K and L) The surface area of IBA1+ microglia cells in the IP and CP region of NC agomir, miR-324-5p agomir and miR-324-5p antagomir treated mice on D3 (K) and D7 (L) after MCAO (n=10). (M) Representative schematic showing dendritic branch structure in the IP region of NC agomir, miR-324-5p agomir and miR-324-5p antagomir treated mice, and in the CP cortical region of NC agomir treated mice on D7 after MCAO. Scale bar: 50 μm. (N) Sholl analysis of the dendritic branch structure in the IP region of NC agomir, miR-324-5p agomir and miR- 324-5p antagomir treated mice on D7 after MCAO (n=8). (O) Representative images of basal dendritic spines in the IP region of NC agomir, miR-324-5p agomir and miR-324-5p antagomir treated mice, and in the CP cortical region of NC agomir treated mice on D7 after MCAO. Mushroom spines marked by arrows, stubby spines marked by black arrowheads and filopodial-like thin spines marked by white arrowheads. Scale bar: 10 μm. (P and Q) Quantification of the total spine density (P) and mushroom spine density (Q) in the IP and CP region of NC agomir, miR-324-5p agomir and miR-324-5p antagomir treated mice on D7 after MCAO (n=10). The composition of mushroom spines, stubby spines and thin spines within the total dendritic spine is indicated; *p<0.05, **p<0.01 and ***p<0.001 by one-way ANOVA with Tukey’s post-hoc test. In A, *p<0.05, **p<0.01 and ***p<0.001 by unpaired two-tailed Student’s t test. In D, E and N, *p<0.05, **p<0.01 and ***p<0.001 vs. NC agomir group, # p<0.05, ##p<0.01 and ###p<0.001 vs. miR-324-5p antagomir group by two-way ANOVA with Tukey’s post-hoc test.

Administration of CCL5 antibody alleviated OGD injury in cortical neurons co-cultured with astrocytes. (A) Illustration of astrocyte-neuron co-culture experimental protocol. Primary cortical astrocytes and neurons were isolated and co-cultured. After OGD treatment, BSA, CCL5 antibody, rCCL5, or MVC+rCCL5 was added to the culture medium. Alternatively, NC agomir, miR-324-5p agomir, or miR-324-5p antagomir was added to the culture medium post-OGD treatment. Samples were collected at D3 and/or D6 after OGD. (B) ELISA analysis of CCL5 concentration in the culture medium of BSA, CCL5 antibody, rCCL5, and MVC+rCCL5 group co-cultured cells on D3 after OGD, and in the culture medium of un-OGD-treated control group (n=5). (C and D) Representative images of NeuN immunofluorescence and TUNEL staining (C), and the ratio of apoptotic neurons (D) in BSA, CCL5 antibody, rCCL5, and MVC+rCCL5 group co-cultured cells on D3 after OGD (n=10). (E and F) Representative schematic (E) and Sholl analysis (F) of dendritic branch structure in BSA, CCL5 antibody, rCCL5, and MVC+rCCL5 group co-cultured cells on D6 after OGD (n=10). (G and H) Representative images of immunofluorescent double labeling of MAP2 and SYN1 (G) and quantification of the synapse puncta density (H) in BSA, CCL5 antibody, rCCL5, and MVC+rCCL5 group co-cultured neurons on D6 after OGD (n=10); *p<0.05, **p<0.01 and ***p<0.001 by one-way ANOVA with Tukey’s post-hoc test. In F, *p<0.05, **p<0.01 and ***p<0.001 vs. BSA group, # p<0.05, ##p<0.01 and ###p<0.001 vs. rCCL5 group by two-way ANOVA with Tukey’s post-hoc test.

Administration of miR-324-5p agomir alleviated OGD injury in cortical neurons co-cultured with astrocytes. (A) ELISA analysis of CCL5 concentration in the culture medium of NC agomir, miR- 324-5p agomir and miR-324-5p antagomir group co-cultured cells on D3 after OGD, and in the culture medium of un-OGD-treated control group (n=5). (B and C) Representative images of NeuN immunofluorescence and TUNEL staining (B), and the ratio of apoptotic neurons (C) in NC agomir, miR-324-5p agomir and miR-324-5p antagomir group co-cultured cells on D3 after OGD (n=10). (D and E) Representative schematic (D) and Sholl analysis (E) of dendritic branch structure in NC agomir, miR- 324-5p agomir and miR-324-5p antagomir group on D6 after OGD (n=10). (F and G) Representative images of immunofluorescent double labeling of MAP2 and SYN1 (F) and quantification of the synapse puncta density (G) in NC agomir, miR-324-5p agomir and miR-324-5p antagomir group neurons on D6 after OGD (n=10); *p<0.05, **p<0.01 and ***p<0.001 by one-way ANOVA with Tukey’s post-hoc test. In E, *p<0.05 and **p<0.01 vs. NC agomir group, # p<0.05, ##p<0.01 and ###p<0.001 vs. miR-324-5p antagomir group by two-way ANOVA with Tukey’s post-hoc test.

Intracerebral CCL5 antibody delivery enhanced the ERK/CREB pathway in the ipsilateral cortex of MCAO mice. (A) Representative Western blot of p-ERK, ERK, p-CREB, and CREB expression in the IC, IP, CC, and CP cortical proteins from BSA, CCL5 antibody, rCCL5, and MVC treated mice on D7 after MCAO (n=5). (B and C) The expression ratio of p-ERK to ERK (B), and p- CREB to CREB (C) in the IC, IP, CC, and CP cortical proteins from BSA, CCL5 antibody, rCCL5, and MVC treated mice on D7 after MCAO (n=5). All protein expression levels were normalized to the expression in the CP region; *p<0.05, **p<0.01 and ***p<0.001 by two-way ANOVA with Tukey’s post-hoc test.

Administration of miR-324-5p, CCL5 antibody and MVC enhanced neuronal ERK/CREB pathway after OGD injury. (A) Illustration of ACM-neuron co-culture experimental protocol for Figure 7B-D. After OGD treatment, primary cortical neurons were co-cultured with OGD-ACM for 3 days, then BSA, CCL5 antibody, rCCL5, or MVC+rCCL5 was supplemented in the culture medium before neuronal protein was collected. (B) Representative Western blot of p-ERK, ERK, p-CREB, and CREB expression in the BSA, CCL5 antibody, rCCL5, and MVC+rCCL5 treated neuronal proteins. (C and D) The expression ratio of p-ERK to ERK (C), and p-CREB to CREB (D) in the BSA, CCL5 antibody, rCCL5, and MVC+rCCL5 treated neuronal proteins (n=6). (E) Illustration of ACM-neuron co-culture experimental protocol for Figure 7F-K. After OGD treatment, primary cortical neurons were co-cultured with BSA OGD-ACM, anti-CCL5 OGD-ACM, or rCCL5 OGD-ACM for 3 days, with MVC or DMSO added to the medium. (F) Representative Western blot of p-ERK, ERK, p-CREB, and CREB expression in neurons co-cultured with BSA OGD-ACM, anti-CCL5 OGD-ACM, and rCCL5 OGD-ACM. (G and H) The protein expression ratio of p-ERK to ERK (G), and p-CREB to CREB (H) in neurons co-cultured with BSA OGD-ACM, anti-CCL5 OGD-ACM, and rCCL5 OGD-ACM (n=5). (I) Representative Western blot of p-ERK, ERK, p-CREB, and CREB expression in neurons co-cultured with NC agomir OGD-ACM, miR-324-5p agomir OGD-ACM, and miR-324-5p antagomir OGD-ACM. (J and K) The protein expression ratio of p-ERK to ERK (J), and p-CREB to CREB (K) in neurons co-cultured with NC agomir OGD-ACM, miR-324-5p agomir OGD-ACM, and miR-324-5p antagomir OGD-ACM (n=5). Protein expression levels were normalized to the BSA group in C and D, to the BSA group added with DMSO in G and H, to the NC agomir group added with DMSO in J and K. *p<0.05, **p<0.01 and ***p<0.001 by one-way ANOVA with Tukey’s post-hoc test. In C and D, *p<0.05, **p<0.01 and ***p<0.001 by two-way ANOVA with Tukey’s post-hoc test.

Proposed model of the neuroprotective role of astrocytic miR-324-5p by inhibiting CCL5 expression and upregulating neuronal ERK/CREB pathway after ischemic stroke.