Role of the TLR4 pathway in blood-spinal cord barrier dysfunction during the bimodal stage after ischemia/reperfusion injury in rats

Abstract

Background: Spinal cord ischemia-reperfusion (I/R) involves two-phase injury, including an initial acute ischemic insult and subsequent inflammatory reperfusion injury, resulting in blood-spinal cord barrier (BSCB) dysfunction involving the TLR₄ pathway. However, the correlation between TLR₄/MyD₈₈-dependent and TLR₄/TRIF-dependent pathways in BSCB dysfunction is not fully understood. The aim of this study is to characterize inflammatory responses in spinal cord I/R and the events that define its clinical progression with delayed neurological deficits, supporting a bimodal mechanism of injury.

Methods: Rats were intrathecally pretreated with TAK-242, MyD₈₈ inhibitory peptide, or Resveratrol at a 12 h interval for 3 days before undergoing 14-minute occlusion of aortic arch. Evan's Blue (EB) extravasation and water content were detected at 6, 12, 18, 24, 36, 48, and 72 h after reperfusion. EB extravasation, water content, and NF-κB activation were increased with time after reperfusion, suggesting a bimodal distribution, as maximal increasing were detected at both 12 and 48 h after reperfusion. The changes were directly proportional to TLR₄ levels determined by Western blot. Double-labeled immunohistochemical analysis was also used to detect the relationship between different cell types of BSCB with TLR₄. Furthermore, NF-κB and IL-1β were analyzed at 12 and 48 h to identify the correlation between MyD₈₈-dependent and TRIF-dependent pathways.

Results: Rats without functional TLR₄ and MyD₈₈ attenuated BSCB leakage and inflammatory responses at 12 h, suggesting the ischemic event was largely mediated by MyD₈₈-dependent pathway. Similar protective effects observed in rats with depleted TLR₄, MyD₈₈, and TRIF receptor at 48 h infer that the ongoing inflammation which occurred in late phase was mainly initiated by TRIF-dependent pathway and such inflammatory response could be further amplified by MyD₈₈-dependent pathway. Additionally, microglia appeared to play a major role in early phase of inflammation after I/R injury, while in late responding phase both microglia and astrocytes were necessary.

Conclusions: These findings indicate the relevance of TLR4/MyD₈₈-dependent and TLR₄/TRIF-dependent pathways in bimodal phases of inflammatory responses after I/R injury, corresponding with the clinical progression of injury and delayed onset of symptoms. The clinical usage of TLR₄ signaling inhibitors at different phases may be a therapeutic option for the prevention of delayed injury.

Figure 1

Time dependent progression of BSCB permeability changes after spinal cord ischemia/reperfusion (I/R) injury. (A) Effects on Evans blue (EB) extravasation in the Sham and I/R groups along the entire time course from 6 to 72 h after surgery. I/R-induced a significant increase in EB extravasations at each observed time point, with the greatest intensity at 12 and 48 h, especially in the gray matter. Scale bars = 200 μm. (B) Quantification of EB content of the spinal cord (μg/g). (C)  Quantification of the water content of the spinal cord. All data are represented as means ± SEM (n = 8 per group). *P < 0.01 vs. Sham group.

Figure 2

Time course of TLR₄ activation in the spine after I/R injury. (A) Representative immunofluorescence of TLR_4. Prominent TLR₄ immunoreactivation was observed in both dorsal horns of the spinal cord in operated rats. Spinal cord I/R-induced TLR₄-up regulation increased with time and peaked at 12 and 48 h after surgery. Scale bars = 100 μm for the 100× and 50 μm for the 400× images. (B) Representative Western blot analysis and integrated density values (IDVs) of TLR₄ activation. The IDVs of TLR₄ in the I/R group with different reperfusion time points were calculated after normalizing against the Sham group and presented as relative protein expression units. The maximal of IDVs were observed at 12 and 48 h after surgery. All data are represented as means ± SEM (n = 8 per group). *P < 0.05 vs. Sham group.

Figure 3

Blocking effects of specific binding protein on TLR₄, MyD₈₈, and TRIF protein. (A-C) Representative Western blot analysis of TLR₄, MyD₈₈, and TRIF protein in L_4–6 segments of spinal cord at 12 and 48 h after I/R injury. (D-F) The integrated density values (IDVs) in each group with different reperfusion time points were calculated after normalizing against β-actin and presented as relative protein expression units. Data are presented as means ± SEM (n = 8 per group). **P < 0.01 vs. Sham group; ## P < 0.01 vs. I/R group.

Figure 4

Double-labeled immunofluorescence TLR₄ receptor and specific cell population in blood spinal cord barrier after spinal cord I/R injury. (A) Representative immunohistochemical localization of capillary endothelial cells (CD31; green), pericytes (CD13; green), microglia (Iba-1; green), astrocytes (GFAP; green), and Toll-like receptor (TLR₄; red) in spinal cords of sham-operated or operated rats at 12 and 48 h after the surgical procedure, respectively. Scale bars = 100 μm for the 200× images. (B) Effects of TLR₄ signaling on TLR4 co-localization cells. Representative double-labeling micrographs show that intrathecal pretreatment with TAK-242, MIP, and Resveratrol before I/R injury significantly prevented microglia and astrocytes upregulated expression of TLR₄ at 48 h after the surgery_. Scale bars = 50 μm for the 400× images. (C) Quantification of TLR₄-positive microglia and TLR₄-positive astrocytes in the spinal cords. Data are presented as mean ± SEM (n = 6). **P < 0.01 vs. Sham group; ## P < 0.01 vs. I/R group; &&P < 0.05 vs. TAK group.

Figure 5

Effects of TLR4 signaling on BSCB dysfunction after I/R injury. (A) Representative EB dye after intrathecal injection with TAK-242, MIP, and Resveratrol. Almost no red fluorescence was seen in spinal cord parenchyma in the Sham group at 12 and 48 h after injury. Much more red fluorescence could be seen in the I/R and Resveratrol groups at 12 h. These increased in intensity at 48 h in all groups after injury, especially in the gray matter of the I/R and MIP groups. Minimal EB red fluorescence was seen in the Sham and TAK groups at the above two time points. (B) Quantification data of EB content of spinal cord (μg/g). (C) Quantification of the water content of the spinal cord. All data are represented as mean ± SEM (n = 8 per group). Scale bars = 50 μm for 100× images. **P < 0.01 vs. Sham group; ## P < 0.01 vs. I/R group; &&P < 0.05 vs. TAK group.

Figure 6

Effects of TLR₄ signaling on NF-κB activation and inflammatory cytokines after I/R. (A) Representative Western blot and quantitative protein analysis of NF-κB in the spinal cord at 12 and 48 h after intrathecal pretreatment with TAK-242, MIP, and Resveratrol for 3 days. (B) The relative integral density values were calculated after normalizing against β-actin in each sample. (C) Quantification of IL-1β production in spinal cord at 12 and 48 h after I/R injury, as assessed by ELISA. **P < 0.01 vs. Sham group; ## P < 0.01 vs. I/R group; &&P < 0.05 vs. MIP group.