Multifaceted Mechanisms of WY-14643 to Stabilize the Blood-Brain Barrier in a Model of Traumatic Brain Injury

Abstract

The blood-brain barrier (BBB) is damaged during ischemic insults such as traumatic brain injury or stroke. This contributes to vasogenic edema formation and deteriorate disease outcomes. Enormous efforts are pursued to understand underlying mechanisms of ischemic insults and develop novel therapeutic strategies. In the present study the effects of PPARα agonist WY-14643 were investigated to prevent BBB breakdown and reduce edema formation. WY-14643 inhibited barrier damage in a mouse BBB in vitro model of traumatic brain injury based on oxygen/glucose deprivation in a concentration dependent manner. This was linked to changes of the localization of tight junction proteins. Furthermore, WY-14643 altered phosphorylation of kinases ERK1/2, p38, and SAPK/JNK and was able to inhibit proteosomal activity. Moreover, addition of WY-14643 upregulated PAI-1 leading to decreased t-PA activity. Mouse in vivo experiments showed significantly decreased edema formation in a controlled cortical impact model of traumatic brain injury after WY-14643 application, which was not found in PAI-1 knockout mice. Generally, data suggested that WY-14643 induced cellular responses which were dependent as well as independent from PPARα mediated transcription. In conclusion, novel mechanisms of a PPARα agonist were elucidated to attenuate BBB breakdown during traumatic brain injury in vitro.

Keywords: PPARα; blood-brain barrier; ischemia; pirinixic acid; stroke; traumatic brain injury.

FIGURE 1

Attenuation of barrier breakdown of cerebEND/C6 co-culture on Transwell inserts after 4 h oxygen/glucose deprivation (OGD) treatment by PPARα agonist WY-14463 and PPARα antagonist GW6471 – influence on transendothelial electrical resistance (TEER) (A) or fluorescein permeability (B). Statistical significance was labeled with ^∗ vs. NC6 and ^# vs. OC6 (p < 0.05). Data are presented as means ± SD (n = 6–36).

FIGURE 2

Influence of WY-14643 and PPARα antagonist GW6471 on effects of OC6 treatment on the protein expression of tight junction proteins ZO-1, occludin and claudin-5 of cerebEND cells. OC6 treatment of cerebENDs accords to 4 h OGD treatment with medium supernatants derived from 4 h OGD-treated C6 cells, whereas NC6 treatment means cerebEND cells incubated for 4 h under normoxic conditions with medium supernatants of 4 h normoxic treated C6 cells. Results for ZO-1 (A), occludin (B) and claudin-5 (C) were displayed. Densitometric analyzed protein expression of total cell lysates were compared to RIPA and triton-X 100 fractions. Moreover, protein distributions [%] between RIPA and triton-X 100 fractions were presented. Values were related to β-actin and according representative western blot images are depicted. Dotted lines indicated cuts of images of the same blot due to presentation reasons of selected bands. Statistical significance was labeled with ^∗ vs. NC6 and ^# vs. OC6 (p < 0.05). Data are presented as means ± SEM (n = 3–8).

FIGURE 3

Influence of WY-14643 and PPARα antagonist GW6471 on effects of OC6 treatment on phosphorylation ratios of MAP kinases ERK1/2, Akt, p38, and SAPK/JNK of cerebEND cells. OC6 treatment of cerebENDs accords to 4 h OGD treatment with medium supernatants derived from 4 h OGD-treated C6 cells, whereas NC6 treatment means cerebEND cells incubated for 4 h under normoxic conditions with medium supernatants of 4 h normoxic treated C6 cells. Results for ERK1/2 (A), Akt (B), p38 (C), and SAPK/JNK (D) were displayed. Densitometric analyzed protein expression of total cell lysates were compared to RIPA and triton-X 100 fractions. Moreover, protein distributions [%] between cytoskeleton (RIPA) and membrane enriched fractions (triton-X 100) were presented. Values were related to β-actin and according representative western blot images are depicted. Dotted lines indicated cuts of images of the same blot due to presentation reasons of selected bands. Statistical significance was labeled with ^∗ vs. NC6, ^# vs. OC6 and ^§ vs. OC6+WY (p < 0.05). Data are presented as means ± SEM (n = 3–6).

FIGURE 4

Influence of WY-14643 on proteasomal chymotrypsin-like (A), trypsin-like (B), and peptidylglutamyl-peptide hydrolyzing activity (PGPH) (C) activity after isolation of the enzymes from cerebEND cells. Statistical significance was labeled with ^∗ vs. NC6 and ^# vs. OC6 (p < 0.05). Data are presented as means ± SEM (n = 3–4).

FIGURE 5

Influence of WY-14643 on t-PA and PAI-1 expression and activity in the blood-brain barrier (BBB) in vitro ischemia model. OC6 treatment of cerebENDs accords to 4 h OGD treatment with medium supernatants derived from 4 h OGD-treated C6 cells, whereas NC6 treatment means cerebEND cells incubated for 4 h under normoxic conditions with medium supernatants of 4 h normoxic treated C6 cells. Results on mRNA as well as protein expression of t-PA (A) and PAI-1 (B) of cerebEND cells after OC6 treatment were displayed. Effects of WY-14643 on t-PA activity and of rhPAI-1 on barrier functionality were shown in (C). Densitometric analyzed protein expression of total cell lysates were compared to RIPA and triton-X 100 fractions. Protein distributions [%] between cytoskeleton (RIPA) and membrane enriched fractions (triton-X 100) were presented. Values were related to β-actin and according representative western blot images are depicted. Dotted lines indicated cuts of images of the same blot due to presentation reasons of selected bands. Statistical significance was labeled with ^∗ vs. NC6, ^# vs. OC6 and ^§ vs. OC6+WY (p < 0.05). Data are presented as means ± SEM (mRNA: n = 4–8, protein expression: n = 3–6, t-PA activity: n = 8–10, and barrier functionality: n = 12–36).

FIGURE 6

Influence of WY-14643 on brain water content [%] 24 h after controlled cortical impact in a mouse traumatic brain injury model. 60 mg/kg body weight of WY-14643 reduced the increase of brain water content in the ipsilateral brain hemisphere of wild type C57Bl/6 mice significantly. Addition of PPARα antagonist GW6471 did not reverse the beneficial effects of WY-14643 (A). Administration of 60 mg/kg body weight WY-14643 did not decrease brain water content of PAI-1 knockout mice (PAI-1^-/-) after TBI. (B). Statistical significance was labeled with ^∗ (p < 0.05). Data are presented as means ± SEM (wild type: n = 6–12; PAI-1^-/-: n = 6–7).