Cerebellar Fastigial Nucleus Stimulation in a Chronic Unpredictable Mild Stress Rat Model Reduces Post-Stroke Depression by Suppressing Brain Inflammation via the microRNA-29c/TNFRSF1A Signaling Pathway

Abstract

BACKGROUND We previously reported that cerebellar fastigial nucleus stimulation reduced post-stroke depression in a rat model by reducing inflammation. This study aimed to investigate the molecular inflammatory signaling pathways associated with cerebellar fastigial nucleus stimulation in an established rat model of post-stroke depression. MATERIAL AND METHODS Twenty-four Sprague-Dawley rats included a sham group (N=6), an untreated stroke group (N=6), an untreated post-stroke depression model group (PSD) (N=6), and the model group treated with cerebellar fastigial nucleus stimulation (FNS) (N=6). The rat stroke model involved occlusion of the middle cerebral artery occlusion (MCAO). Post-stroke depression model was established using chronic unpredictable mild stress treatment and was verified using an open field test. Real-time polymerase chain reaction (PCR) and Western blot compared expression levels of microRNA-29c (miR-29c), miR-676, TNFRSF1A, tumor necrosis factor-alpha (TNF-alpha), interleukin (IL)-6 and IL-1ß in cerebellar tissue. U251 human glioblastoma cells and SH-SY5Y human neuroblastoma cells were studied in vitro. RESULTS Cerebellar fastigial nucleus stimulation reduced behaviors associated with depression in the rat model, upregulated the expression of miR-29c, and reduced the expression of TNFRSF1A and inflammatory cytokines, and mildly reduced neuronal apoptosis. Bioinformatics data analysis identified a regulatory relationship between miR-29c and TNFRSF1A. SH-SY5Y cells treated with a miR-29c mimic, or TNFRSF1A short interfering RNA (siRNA), identified a negative regulatory relationship between TNFRSF1A and miR-29c. CONCLUSIONS In a rat model, cerebellar fastigial nucleus stimulation reduced the expression of TNFRSF1A by upregulating miR-29c expression, which suppressed the expression of inflammatory cytokines, resulting in reduced severity of post-stroke depression.

Figure 1

Dynamic changes of body weight, sucrose preference, locomotor activity, and rearing activity in the sham group, the stroke group, the post-stroke depression (PSD) group, and the fastigial nucleus stimulation (FNS) group showed that fastigial nucleus stimulation treatment could reduce behavior associated with post-stroke depression. (A) Dynamic changes in rat body weight. Compared with the sham group, the rat body weights in the stroke and fastigial nucleus stimulation (FNS) group began to increase after day 11. The rat body weights in the post-stroke depression (PSD) group continuously decreased and reached their lowest level on day 30. ^# P<0.01 vs. the stroke group; * P<0.05 vs. the PSD group. The bars represent the standard deviation (SD). N=6. (B) Dynamic changes of sucrose preference. Compared with the sham and stroke groups, the consumption of sucrose water in the fastigial nucleus stimulation (FNS) group started to increase slightly after day 14 and remained constant thereafter. The consumption of sucrose water continued to decrease in the PSD group and reached the lowest level on day 25. ^# P<0.01 vs. the stroke group; * P<0.01 vs. the PSD group. The bars represent the SD. N=6. (C). Dynamic changes of locomotor activity in the open field test. Compared with that in the sham group and stroke group, the locomotor activity in the PSD group continued to decrease and reached its lowest point on day 30. The locomotor activity in the fastigial nucleus stimulation group started to increase slightly after day 17. A significant difference was found between the PSD group and the FNS group after day 21. ^# P<0.01 vs. the stroke group. * P<0.01 vs. the PSD group. The bars represent the SD. N=6. (D) Dynamic changes of rearing activity in the open field test. Compared with that in the sham group and stroke group, the rearing activity in the PSD group continued to decrease and reached its lowest point on day 30. The rearing activity in the FNS group started to increase slightly after day 17. A significant difference was found between the PSD group and the FNS group after day 21. ^# P<0.01 vs. the stroke group. * P<0.01 vs. the PSD group. The bars represent the SD. N=6.

Figure 2

The relative expression of miR-29c and miR-676 in the sham group, the stroke group, the post-stroke depression (PSD) group, and the fastigial nucleus stimulation (FNS) group showed that cerebellar fastigial nucleus stimulation treatment increased the expression of miR-29c in the rat model of post-stroke depression (PSD). (A) The relative expression of miR-29c was reduced in the post-stroke depression (PSD) group compared with the sham and stroke groups, while the cerebellar fastigial nucleus stimulation (FNS) increased the expression of miR-29c. (B) The relative expression of microRNA-676 (miR-676) showed no significant difference among different model groups.

Figure 3

The relative level of TNFRSF1A mRNA and the inflammatory cytokines, tumor necrosis factor-α (TNF-α), interleukin (IL)-6, and IL-1β, in the sham group, the stroke group, the post-stroke depression (PSD) group, and the fastigial nucleus stimulation (FNS) group showed that cerebellar fastigial nucleus stimulation reduced the expression of TNFRSF1A mRNA and inflammatory cytokines. (A) The relative expression of TNFRSF1A mRNA was increased in the rat model of post-stroke depression (PSD) compared with that in the sham-operated rats and stroke rats, whereas the fastigial nucleus stimulation treatment reduced the level of TNFRSF1A mRNA. (B) Compared with the sham group and the stroke group, the level of tumor necrosis factor-α (TNF-α) was upregulated in the PSD group, and the fastigial nucleus stimulation treatment reduced the level of TNF-α. (C) Compared with that in the sham group and stroke group, the level of IL-6 was evidently upregulated in the PSD group, and the fastigial nucleus stimulation treatment reduced the level of IL-6. (D) Compared with that in the sham group and stroke group, the level of IL-1β was upregulated in the PSD group, and the fastigial nucleus stimulation treatment reduced the level of IL-1β.

Figure 4

Expression of the TNFRSF1A protein, total caspase-3, and active caspase-3 in the sham group, the stroke group, the post-stroke depression (PSD) group, and the fastigial nucleus stimulation (FNS) group showed that the cerebellar fastigial nucleus stimulation reduced the protein expression of TNFRSF1A and caspase-3. (A) Western blot analysis showed that the expression of TNFRSF1A protein, total caspase-3, and active caspase-3 in the post-stroke depression (PSD) group were significantly increased compared with the sham group and stroke group, and fastigial nucleus stimulation treatment reduced the expression of these proteins. (B) The relative density of TNFRSF1A protein was slightly decreased in the fastigial nucleus stimulation (FNS) treated group compared with that in the untreated PSD group; the relative density of TNFRSF1A protein in the sham group and the stroke group were the lowest. (C) The relative density of total caspase-3 was slightly decreased in the FNS group compared with that in the PSD group; the relative density of total caspase-3 in the sham group and the stroke group were the lowest. (D) The relative density of active caspase-3 was slightly decreased in the FNS group compared with that in the PSD group; the relative density of active caspase-3 in the sham group and stroke group were the lowest.

Figure 5

The TUNEL assay was used to measure neuron apoptosis in the sham group, the stroke group, the untreated post-stroke depression (PSD) group, and the fastigial nucleus stimulation (FNS) group. The post-stroke depression (PSD) group was associated with the highest apoptosis index. The sham group and the stroke group were associated with the lowest apoptosis index. The apoptosis index in the fastigial nucleus stimulation (FNS) group was slightly decreased, indicating that cerebellar fastigial nucleus stimulation had a mild inhibitory effect on neuronal apoptosis.

Figure 6

Computational analysis identified the TNFRSF1A gene as a target of miR-29c, with a binding site of miR-29c located in the 3′UTR of TNFRSF1A and the regulatory relationship between TNFRSF1A and miR-29c was confirmed using luciferase assays and cells co-transfected with miR-29c mimics and plasmids carrying wild-type or mutant 3′UTR of TNFRSF1A. (A) Online bioinformatics analysis tools were used to detect a putative binding site of miR-29c in the 3′UTR of TNFRSF1A. (B) Luciferase activity was decreased when SH-SY5Y human neuroblastoma cells, which were co-transfected with miR-29c mimics and plasmids carrying wild-type 3′UTR of TNFRF1A. (C) Luciferase activity was decreased when the U251 human glioblastoma cells, which were co-transfected with miR-29c mimics and plasmids carrying wild-type 3′UTR of TNFRF1A.

Figure 7

SH-SY5Y human neuroblastoma cells treated with miR-29c mimic, TNFRSF1A short interfering RNA (siRNA), or a negative control resulted in a negative regulatory relationship between the expression of TNFRSF1A and miR-29c. (A) The relative expression of miR-29c was increased in SH-SY5Y human neuroblastoma cells treated with miR-29c mimic compared with cells treated with TNFRSF1A short interfering RNA (siRNA) or a negative control. (B) The relative expression of TNFRSF1A mRNA was reduced in SH-SY5Y human neuroblastoma cells treated with miR-29c mimic or TNFRSF1A siRNA. (C) Western blot showed that the relative expression of TNFRSF1A protein was markedly reduced in SH-SY5Y human neuroblastoma cells treated with miR-29c mimic or TNFRSF1A siRNA. (D) Relative levels of TNFRSF1A in SH-SY5Y human neuroblastoma cells treated with miR-29c mimic or TNFRSF1A siRNA were reduced.

Figure 8

U251 human glioblastoma cells were treated with miR-29c mimic, TNFRSF1A short interfering RNA (siRNA), or a negative control resulted in a negative regulatory relationship between the expression of TNFRSF1A and miR-29c. (A) The relative expression of miR-29c was increased in U251 human glioblastoma cells treated with miR-29c mimic compared with that in the cells treated with TNFRSF1A short interfering RNA (siRNA) or negative control. (B) The relative expression of TNFRSF1A mRNA was markedly reduced in U251 cells treated with miR-29c mimic or TNFRSF1A siRNA. (C) Western blot showed that the relative expression of TNFRSF1A protein was reduced in U251 cells treated with miR-29c mimic or TNFRSF1A siRNA. (D) The relative density of TNFRSF1A in U251 cells treated with miR-29c mimic or TNFRSF1A siRNA was reduced.