Ginkgo Flavonol Glycosides or Ginkgolides Tend to Differentially Protect Myocardial or Cerebral Ischemia-Reperfusion Injury via Regulation of TWEAK-Fn14 Signaling in Heart and Brain

Abstract

Shuxuening injection (SXNI), one of the pharmaceutical preparations of Ginkgo biloba extract, has significant effects on both ischemic stroke and heart diseases from bench to bedside. Its major active ingredients are ginkgo flavonol glycosides (GFGs) and ginkgolides (GGs). We have previously reported that SXNI as a whole protected ischemic brain and heart, but the active ingredients and their contribution to the therapeutic effects remain unclear. Therefore, we combined experimental and network analysis approach to further explore the specific effects and underlying mechanisms of GFGs and GGs of SXNI on ischemia-reperfusion injury in mouse brain and heart. In the myocardial ischemia-reperfusion injury (MIRI) model, pretreatment with GFGs at 2.5 ml/kg was superior to the same dose of GGs in improving cardiac function and coronary blood flow and reducing the levels of lactate dehydrogenase and aspartate aminotransferase in serum, with an effect similar to that achieved by SXNI. In contrast, pretreatment with GGs at 2.5 ml/kg reduced cerebral infarction area and cerebral edema similarly to that of SXNI but more significantly compared with GFGs in cerebral ischemia-reperfusion injury (CIRI) model. Network pharmacology analysis of GFGs and GGs revealed that tumor necrosis factor-related weak inducer of apoptosis (TWEAK)-fibroblast growth factor-inducible 14 (Fn14) signaling pathway as an important common mechanism but with differential targets in MIRI and CIRI. In addition, immunohistochemistry and enzyme linked immunosorbent assay (ELISA) assays were performed to evaluate the regulatory roles of GFGs and GGs on the common TWEAK-Fn14 signaling pathway to protect the heart and brain. Experimental results confirmed that TWEAK ligand and Fn14 receptor were downregulated by GFGs to mitigate MIRI in the heart while upregulated by GGs to improve CIRI in the brain. In conclusion, our study showed that GFGs and GGs of SXNI tend to differentially protect brain and heart from ischemia-reperfusion injuries at least in part by regulating a common TWEAK-Fn14 signaling pathway.

Keywords: Shuxuening injection; TWEAK–Fn14 signaling; cerebral ischemia—reperfusion injury; ginkgo flavonol glycosides; ginkgolides; myocardial ischemia—reperfusion injury.

Figure 1

Changes of echocardiographic characterization of cardiac function in myocardial ischemia–reperfusion injury (MIRI) mice. After 30 min of ischemia and 18 h of reperfusion, the effect of ginkgo flavonol glycosides (GFGs, 2.5 ml/kg), ginkgolides (GGs, 2.5 ml/kg), and Shuxuening injection (SXNI, 2.5 ml/kg) on cardiac function was quantitatively ▲ evaluated. (A) Representative echocardiography images of different groups. Bar graph quantitation of echocardiographic changes in cardiac function in different groups detected in M-mode: (B) LVEF %, (C) LVFS %, (D) CO, (E) stroke volume, (F) LVIDd, (G) LVIDs, (H) LVPWd, (I) LVPWs, (J) LV Vold, (K) LV Vols, (L) heart rate, and (M) LV mass. Values were expressed as mean ± SEM (n = 5). ^## P < 0.01 vs. sham group, *P < 0.05, **P < 0.01 vs. I/R group, ^▲ P < 0.05, ^▲▲ P < 0.01 vs. GFGs group (LV, left ventricular; EF, ejection fraction; FS, LV fractional shortening; CO, cardiac output; LVIDd, LV internal dimensions at diastole; LVIDs, LV internal diameter systole; LVPWd, LV posterior wall diastole; LVPWs, LV posterior wall systole; LV Vols, LV systole volume; LV Vold, LV diastole volume).

Figure 2

GFGs and GGs improved the coronary blood flow and left ventricular function after myocardial ischemia and reperfusion in MIRI mice. After 30 min of ischemia and 18 h of reperfusion, the effect of GFGs (2.5 ml/kg), GGs (2.5 ml/kg), and SXNI (2.5 mg/kg) on cardiac function was quantitatively evaluated. (A) Representative echocardiography images of coronary blood flow were determined with different groups. (B) AoV VTI, (C) AV peak velocity, and (D) AV peak pressure were detected in color Doppler mode. Quantitative assessment of hemodynamic on left ventricular function based on (E) +dp/dt _max and (F) −dp/dt _max. Values were expressed as mean ± SEM (n = 5). ^## P < 0.01 vs. sham group, *P < 0.05, **P < 0.01 vs. I/R group, ^▲▲ P < 0.01 vs. GFGs group (AV Peak Vel, aortic valve peak velocity; AoV VTI, aorta velocity–time integral mean velocity; +dp/dt_max, left ventricular maximum upstroke velocity; −dp/dt_max, left ventricular maximum descent velocity).

Figure 3

GFGs and GGs alleviated the myocardial I/R injury in MIRI mice. After 30 min of ischemia and 24 h of reperfusion, the protective effects of GFGs and GGs on MIRI were assessed by hematoxylin and eosin staining and measurement of biochemical parameters. (A) Hematoxylin and eosin staining results for the histopathological of the myocardial tissue indicated the histopathological changes in I/R model as well as that caused by GFGs, GGs, and SXNI. Myocardium damage sections were observed and indicated with yellow arrows in representative pictures (12.5×, 100×, and 400× magnification). (B–E) Effects of GFGs and GGs on the alterations of lactate dehydrogenase (LDH), aspartate aminotransferase (AST), creatine kinase-MB (CK-MB), and creatine kinase (CK) in serum following reperfusion (n = 6). Values were expressed as mean ± SEM. ^# P < 0.05, ^## P < 0.01 vs. sham group, *P < 0.05, **P < 0.01 vs. I/R group, ^▲ P < 0.05 vs. GFGs group.

Figure 4

GGs reduced the cerebral infarct size, edema, and neurological deficit score in cerebral ischemia–reperfusion injury (CIRI) mice. After 60 min of ischemia and 24 h of reperfusion, brain tissue was imaged and transected for detection of cerebral damage by micro-CT and 2, 3, 5-Triphenyl-2H-tetrazolium chloride (TTC) staining, separately. Normal brain tissues were represented in red, whereas the white marked areas of infarct. (A) Representative images of TTC staining in different groups, including sham, I/R model, I/R+GGs, I/R+GFGs, and I/R+SNXI groups. TTC staining was performed at 24 h after stroke. (C) Quantitation of TTC stain of brain slices as the percentage of infarct volumes of each group (n = 4–5). The occurrence of cerebral edema causes the midline of the coronal image of the brain to shift (the red arrow in Figure B was the offset midline). Therefore, the more the midline offset distance, the more severe the brain edema. (B) Representative images of micro-CT imaging in different groups, including sham, I/R model, I/R+GGs, I/R+GFGs, and I/R+SNXI groups. (D) Quantitation of midline offset distance of each group (n = 4–5). In addition, the mice were scored for neurological deficits. (E) Neurological deficit score. Values were expressed as mean ± SEM. ^### P < 0.001 vs. sham group, *P < 0.05, **P < 0.05 vs. I/R group, ^▲ P < 0.05, ^▲▲ P < 0.01 vs. GFGs group.

Figure 5

Expression levels of tumor necrosis factor-related weak inducer of apoptosis (TWEAK) and Fn14 in MIRI and CIRI mice. After 24 h of reperfusion, heart and brain were removed from the mice, and their supernatant and tissue block were prepared for immunohistochemical and ELISA analyses of the soluble TWEAK ligand and in situ Fn14 receptor protein, respectively. (A) Effects of GFGs, GGs, and SXNI on the alteration of TWEAK in heart tissue supernatant (n = 3). (B) Effects of GFGs, GGs, and SXNI on the alteration of TWEAK in brain tissue supernatant (n = 3). (C) Immunohistochemical staining for the heart tissue section of MIRI and control mice indicating the effects of GFGs, GGs, and SXNI on the expression of Fn14 protein (in brown). Representative images (100× magnification) and quantification (n = 3) are shown. (D) Immunohistochemical staining for the brain tissue section of CIRI and control mice indicating the effects of GFGs, GGs, and SXNI on the expression of Fn14 protein (in brown). Representative images (100× magnification) and quantification (n = 3) are shown. Values were expressed as mean ± SD B. ^## P < 0.01 vs. sham group, *P < 0.05, **P < 0.01, vs. model group, ^▲ P < 0.05 vs. GFGs group.

Figure 6

GFGs and GGs regulated common and different genes in MIRI and CIRI mice. Previously identified differentially expressed genes, databases of GFGs and GGs of SXNI and their corresponding targets were used to reveal the interaction network by Ingenuity^® Pathway Analysis (IPA). (A) The relationship between GFGs and GGs of SXNI (in blue) and their crucially targeted genes in CIRI (in light blue) and MIRI (in light purple), as well as the shared target gene (in green). (B) Effects of GFGs and GGs on the expression of interleukin 6 and annexin A1 in mice subjected to MIRI or CIRI (n = 3). Values were expressed as mean ± SD. ^# P < 0.05, ^## P < 0.01 vs. sham group, **P < 0.01, vs. model group.