The Role Of G-CSF Mediated "CXCR4 And Non-CXCR4 Effects" In The Healing Process After Experimental Myocardial Infarction.

Background:Despite the fact that stem cell mobilization via granulocytecolony-stimulating factor (G-CSF) after myocardial infarction (MI) appears toimprove ventricular remodeling in some experimental and clinical reports,the effectsof G-CSF on post-infarct healing and related mechanisms remain controversial.G-CSF can promote mobilization of bone marrow stem cells into peripheral blood,thus facilitating the recruitment of CRCR4⁺ cell into ischemic tissue by formingCXCR4/SDF-1 complex,which could be categorized CXCR4 mediated effects.Inaddition,other effects,such as direct action with G-CSF receptor on cardiomyocytesand subsequent activation of Jak2-Stat3 signal to inhibit apoptosis,could becategorized as non-CXCR4 mediated effects.In this study,we aimed to investigatethe effects of G-CSF on cardiac remodeling after experimental myocardial ischemiareperfusion with treatment of G-CSF and AMD3100 (CXCR4 antagonist) tosystematically evaluate the influence of G-CSF mediated CXCR4 effects andnon-CXCR4 effects on cardiac remodeling in the early and late phase of infarcthealing.Materials and Methods:Rat model of myocardial ischemia reperfusion (I/R) wasestablished in 129 male Wistar rats.Survival animals were randomly divided intoimmediate treatment group (i.e.1 hour after repeffusion,I) or delayed treatmentgroup (i.e.24 hours after reperfusion,D).Each group was further divided into G-CSF(I-G and D-G) and G-CSF plus AMD3100 (CXCR-4 antagonist) (I-GA and D-GA)groups according different regiments of drug administration.7 days and 3 monthsafter I/R,rats were sacrificed and hearts were prepared separately for histologicalanalysis and protein extraction.Before scarification,left ventrioular (LV) functionwas assessed by LV catheterization via right carotid artery by Millar 2F pressurecatheter (SPR320) in all 3 months groups.Collagen deposition and maturation weredetermined by picrosirius-red stained heart tissue sections,under circularly polarizedlight.Arteriogenesis in the infarct and remote area was evaluated by FITC-labeled α-SMA immunofluorecent staining.Wheat germ agglutinin (WGA) staining wasperformed for determination of cross-sectional area (CSA) of viable cardiomyocytesin the septum.Gelatin zymography was used for determination of matrixmetalloproteinase-2 (MMP-2) and MMP-9 activity in the infarct area.The proteinabundance of MMP-2,MMP-9,TIMP-1,TIMP-2,transforming growth factorβ₁(TGF-β₁),stromal cell derived factor-1 (SDF-1) and vascular endothelial growthfactor (VEGF) was determined by western blot.Results:7 days after I/R,rats treated with G-CSF plus AMD3100 co-administrationshowed higher mononuclear cells counts in peripheral blood compared with controlgroup (P＜0.05).There was a borderline difference in infarct size between immediateand delayed G-CSF treatment group (P=0.050).Arteriolar density within infarctregion in I/R groups increased significantly than that in sham group (P＜0.05).3 months after I/R,â‘ Immediate G-CSF treatment was associated with impairedLV remodeling compared with control group,as show by decreased LV free wall(P=0.056) and septum thickness (P=0.031),increased expansion index (P=0.056),increased infarct size (P=0.301,vs 3m-saline;P＜0.05,vs 7d-I-G),increased crosssectional area (CSA) of vital myocardiocytes (P＜0.05),and aggregated LV functionwith incareased LV diastolic pressure (LVDP) (P＜0.05,) and decreased±dP/dt_max(P=0.053 and 0.059,respectively).However,delayed G-CSF treatment resulted inimproved LV remodeling compared with immediate G-CSF treatment,as illustratedby decreased LV free wall thickness (P=0.143),decreased expansion index (P=0.079),decreased CSA of vital myocardioeytes (P＜0.05),and improved LV function withdecreased LVDP (P＜0.05) and increased±dP/dt_max (P＜0.05).Co-treatment withAMD3100 also exhibited better LV function than immediate G-CSF treatment withdecreased LVDP and (P＜0.05) and increased±dP/dt_max (P＜0.05).â‘¡Mobilization ofstern cells by G-CSF,and G-CSF plus AMD3100 could give rise to decreasedcollagen mutation (P＜0.05),but did not significantly influence the collagendeposition in the infarct area.â‘¢G-CSF,and G-CSF plus AMD3100 treatments wereassociated with increased arteriolar density.In addition,G-CSF,and G-CSF plus AMD3100 administration associated with down-regulation of MMP-2/TIMP-2 and MMP-9/TIMP-1,and up-regulation ofTGF-β₁,SDF-1,and VEGF expression levels.Conclusion:In the present long-term (3 months) experimental study,G-CSFtreatment is not associated beneficial effects on LV histopathological and mechanicalchanges after myocardial ischemia/reperfusion injury.Immediate G-CSF treatment (1hour after reperfusion) could result in more significantly worsened post-infarct LVremodeling,compared with delayed treatment (24 hours after reperfusion),whichindicates that the timing of G-CSF administration has a great impact on LV woundrepair.On the contrary,co-administration of AMD3100 may mitigate G-CSF inducedadverse post-infarct LV remodeling,implying a potential role of non-CXCR4 effectsin post-infarct healing process.