MFGE8 promotes adult hippocampal neurogenesis in rats following experimental subarachnoid hemorrhage via modifying the integrin β3/Akt signaling pathway

Abstract

Subarachnoid hemorrhage (SAH) is one of the most severe type of cerebral strokes, which can cause multiple cellular changes in the brain leading to neuronal injury and neurological deficits. Specifically, SAH can impair adult neurogenesis in the hippocampal dentate gyrus, thus may affecting poststroke neurological and cognitive recovery. Here, we identified a non-canonical role of milk fat globule epidermal growth factor 8 (MFGE8) in rat brain after experimental SAH, involving a stimulation on adult hippocampal neurogenesis(AHN). Experimental SAH was induced in Sprague-Dawley rats via endovascular perforation, with the in vivo effect of MFGE8 evaluated via the application of recombinant human MFGE8 (rhMFGE8) along with pharmacological interventions, as determined by hemorrhagic grading, neurobehavioral test, and histological and biochemical analyses of neurogenesis related markers. Results: Levels of the endogenous hippocampal MFGE8 protein, integrin-β3 and protein kinase B (p-Akt) were elevated in the SAH relative to control groups, while that of hippocalcin (HPCA) and cyclin D1 showed the opposite change. Intraventricular rhMGFE8 infusion reversed the decrease in doublecortin (DCX) immature neurons in the DG after SAH, along with improved the short/long term neurobehavioral scores. rhMGFE8 treatment elevated the levels of phosphatidylinositol 3-kinase (PI3K), p-Akt, mammalian target of rapamycin (mTOR), CyclinD1, HPCA and DCX in hippocampal lysates, but not that of integrin β3 and Akt, at 24 hr after SAH. Treatment of integrin β3 siRNA, the PI3K selective inhibitor ly294002 or Akt selective inhibitor MK2206 abolished the effects of rhMGFE8 after SAH. In conclusion, MFGE8 is upregulated in the hippocampus in adult rats with reduced granule cell genesis. rhMFGE8 administration can rescue this impaired adult neurogenesis and improve neurobehavioral recovery. Mechanistically, the effect of MFGE8 on hippocampal adult neurogenesis is mediated by the activation of integrin β3/Akt pathway. These findings suggest that exogenous MFGE8 may be of potential therapeutic value in SAH management. Graphical abstract and proposed pathway of rhMFGE8 administration attenuate hippocampal injury by improving neurogenesis in SAH models. SAH caused hippocampal injury and neurogenesis interruption. Administered exogenous MFGE8, recombinant human MFGE8(rhMFGE8), could ameliorate hippocampal injury and improve neurological functions after SAH. Mechanistically, MFGE8 bind to the receptor integrin β3, which activated the PI3K/Akt pathway to increase the mTOR expression, and further promote the expression of cyclin D1, HPCA and DCX. rhMFGE8 could attenuated hippocampal injury by improving neurogenesis after SAH, however, know down integrin β3 or pharmacological inhibited PI3K/Akt by ly294002 or MK2206 reversed the neuro-protective effect of rhMFGE8.

Graphical abstract and proposed pathway of rhMFGE8 administration attenuate hippocampal injury by improving neurogenesis in SAH models. SAH caused hippocampal injury and neurogenesis interruption. Administered exogenous MFGE8, recombinant human MFGE8(rhMFGE8), could ameliorate hippocampal injury and improve neurological functions after SAH. Mechanistically, MFGE8 bind to the receptor integrin β3, which activated the PI3K/Akt pathway to increase the mTOR expression, and further promote the expression of cyclin D1, HPCA and DCX. rhMFGE8 could attenuated hippocampal injury by improving neurogenesis after SAH, however, know down integrin β3 or pharmacological inhibited PI3K/Akt by ly294002 or MK2206 reversed the neuro-protective effect of rhMFGE8.

Fig. 1. Animal mortality, endovascular perforation and general observation for SAH.

A The animal usage and mortality rate in each group. B The procedure of SAH modeling by endovascular perforation method. C Representative brain sample images for sham and SAH. The blood clots mainly distributed around the Circle of Wills and brain stem after SAH. D SAH grade between groups in the experiment.

Fig. 2. Temporal expression of endogenous MFGE8 and HPCA, and cellular localization of MFGE8 after SAH.

A Experimental design, animal groups and time frame of the analyses in experiment 1. B Representative bands and (C–G). quantitative analyses of endogenous MFGE8 integrin β3, p-Akt, Akt, cyclin D1 and HPCA in left hippocampus after SAH. The MGFE8, integrin β3 and p-Akt are increased and peaked at 24 h after SAH, while cyclin D1 and HPCA show the opposite trend. **P < 0.01 vs sham group. Error bars were represented as mean ± SD. One-way ANOVA, Tukey’s test, n = 6 per group. H Double immunofluorescence staining of MFGE8 (green) with DCX (immature neuron marker, red), Iba-1(microglia marker, red) and GFAP (astrocytes marker, red) in hippocampus at 24 hr after SAH. Red box indicated the hippocampus in the brain slice. n = 2 per group. Scale bar, 50 μm. DCX doublecortin, Iba-1 ionized calcium binding adapter molecule-1, GFAP glial fibrillary acidic protein, DAPI diamidino phenylindole.

Fig. 3. Effect of rhMFGE8 application or MGFE8 siRNA on hippocampal neurogenesis after SAH.

A Experimental design, animal groups and time frame of the analyses in experiment 2. B Double immunofluorescence staining with DCX (red) and BrdU (green) in hippocampus between sham, SAH + vehicle and SAH + rhMFGE8 group as well as SAH + Scr siRNA, SAH + MFGE8 siRNA group. Arrows indicated the DCX positive and BrdU positive cells. C Quantitative analyses of DCX positive BrdU positive cells in three groups. The DCX positive BrdU positive cells are decreased in SAH+vehicle group compared to sham, which are reversed by rhMFGE8 treatment. The cells are also decreased in SAH + MFGE8 siRNA group compared to SAH + Scr siRNA group. The Red box indicated the hippocampus in the brain slice. **P < 0.01 vs sham group, ^##P < 0.01 vs SAH + vehicle group, ^$$P < 0.01 vs SAH + Scr siRNA group. Error bars were represented as mean ± SD. One-way ANOVA, Tukey’s test, n = 3 per group. Scale bar, 100 μm. BrdU bromodeoxyuridine.

Fig. 4. Effect of rhMFGE8 application on hippocampal neural stem/progenitor cells after SAH.

A Double immunofluorescence staining with Nestin(green) and MCM2(red) in hippocampus between sham, SAH + vehicle and SAH + rhMFGE8 group. Triangles indicated the nestin positive and MCM2 positive cells. B Double immunofluorescence staining with BrdU(red) and MCM2(green) in hippocampus between sham, SAH + vehicle and SAH+rhMFGE8 group. Arrowheads indicated the MCM2 positive and BrdU positive cells. C, D Quantitative analyses of Nestin positive MCM2 positive cells and BrdU positive MCM2 positive cells in three groups. **P < 0.01 vs sham group, ^##P < 0.01 vs SAH + vehicle group, Error bars were represented as mean ± SD. One-way ANOVA, Tukey’s test, n = 3 per group. Scale bar, 100 μm. MCM2, minichromosome maintenance complex component 2.

Fig. 5. Long-term neurobehavioral effects of rhMFGE8 administration after SAH.

A Experimental design, animal groups and time frame of the analyses in experiment 3. B Foot fault test at 1 to 3 week show that rhMFGE8 treated SAH animals have less foot fault times. C Rotarod test of 5 rpm and 10 rpm at 1–3 week show that rhMFGE8 treated SAH groups have longer falling latency. D Representative trace map in probe quadrant test showed rhMFGE8 treated SAH spent more time in target area. E Escape latency and (F). swim distance of water maze test from day 1 day 5 show that rhMFGE8 treated SAH animals have less escape latency and swimming distance to find the platform. G Probe quadrant duration in targeted area. H swim velocity of probe trial test. *P < 0.05, vs. sham group; ^#P < 0.05, vs^. SAH + vehicle group. Data was represented as mean ± SD. Two-way ANOVA, n = 8 per group.

Fig. 6. Knockdown integrin β3 or blocked PI3K/Akt signaling pathway abolished the rhMFGE8 treatment effects on neurological function.

A Experimental design, animal groups and time frame of the analyses in experiment 4. B Modified Garcia score and beam balance score show that rhMFGE8 treatment improves the scores, with the effect reversed by int-β3 siRNA and PI3K/Akt inhibitor. C Representative western blot bands and (D–K). Quantitative analyses of MFGE8, int β3, PI3K, p-Akt, Akt, mTOR, cyclin D1, HPCA and DCX at 24 hr after SAH. **P < 0.01 vs sham group; ^##P < 0.01 vs^. SAH+vehicle group; ^$$ P < 0.01 vs. SAH + rhMFGE8+ scr siRNA; ^&& P < 0.01 vs. SAH+rhMFGE8 + vehicle^. Data were represented as mean ± SD. One-way ANOVA, Tukey’s test, n = 6 per group.

Fig. 7. Knockdown integrin β3 or blocked PI3K/Akt signaling pathway abolished the effect of rhMFGE8 improving neurogenesis after SAH.

A Immunohistochemical staining of DCX and (B). quantitative analyses of DCX relative density and DCX positive cell in the hippocampus. Red box indicated the hippocampus in the brain slice. **P < 0.01 vs sham group; ^##P < 0.01 vs^. SAH + vehicle group; ^$$ P < 0.01 vs. SAH + rhMFGE8+ scr siRNA; ^&& P < 0.01 vs. SAH + rhMFGE8 + vehicle^. Data were represented as mean ± SD. One-way ANOVA, Tukey’s test, n = 3 per group.

Fig. 8. Administration of rhMFGE8 attenuated hippocampal injury after SAH.

Nissl staining of hippocampus and quantitative analyses of Nissl staining in CA1, CA3 and dentate gyrus (DG). Red box indicated the hippocampus in the brain slice. *P < 0.05, vs. sham group; #P < 0.05, vs. SAH + vehicle group. Data was represented as mean ± SD. n = 3 per group.