|Home | About | Journals | Submit | Contact Us | Français|
We determined if increasing adherence of multipotent stromal cells (MSCs) would amplify their effects on coronary collateral growth (CCG).
Adhesion was established in cultured coronary endothelials cells (CECS) or MSCs treated with epidermal growth factor (EGF). EGF increased MSCs adhesion to CECs, and increased intercellular adhesion molecule (ICAM-1) or vascular cell adhesion molecule (VCAM-1) expression. Increased adherence was blocked by EGF receptor antagonism or antibodies to the adhesion molecules. To determine if adherent MSCs, treated with EGF, would augment CCG, repetitive episodes of myocardial ischemia (RI) were introduced and CCG was measured from the ratio of collateral-dependent (CZ) and normal zone (NZ) flows. CZ/NZ was increased by MSCs without treatment vs RI-control and was further increased by EGF-treated MSCs. EGF-treated MSCs significantly improved myocardial function vs RI or RI+ MSCs demonstrating that the increase in collateral flow was functionally significant. Engraftment of MSCs into myocardium was also increased by EGF treatment.
These results reveal the importance of EGF in MSCs adhesion to endothelium and suggest that MSCs may be effective therapies for the stimulation of coronary collateral growth when interventions are employed to increase their adhesion and homing (in vitro EGF treatment) to the jeopardized myocardium.
Ischemic heart disease (IHD) remains the leading cause of death in the United States.1 A well developed coronary collateral vasculature ameliorates manifestations of IHD, myocardial infarction and sudden death.2,3 Despite the importance of the coronary collateral circulation, the majority of patients fail to develop sufficiently large coronary collateral vessels and thus, are vulnerable to coronary occlusion.
Bone marrow-derived cells were used to repair infarcted myocardium4-7 or to stimulate angiogenesis in peri-infarct regions. Although there is controversy regarding whether multipotent stromal cells (MSCs) from bone marrow exert a beneficial role by transdifferentiation8 or by their paracrine role,9-12 there is no doubt that for any benefit, MSCs must home to injured tissues. Homing of MSCs involves adhesion to the endothelium and then migration through the wall;13,14 however, the underlying mechanisms for homing are not understood. Expression of specific ligands facilitating recruitment of MSCs to an injured site represents the first step in the homing of stem cells. Therefore, amplification of homing and adhesion of MSCs may augment their therapeutic role.
Epidermal growth factor (EGF) or the EGF family member, HB-EGF (heparin-binding epidermal growth factor-like growth factor)is involved in a variety of physiological events: differentiation of human derived MSCs into endothelial cells15, adhesion of fibroblast16, and angiogenesis.17,18,19 Interestingly, EGF-R phosphorylation was involved in cell adhesion,20 induction of VCAM-121 and MMP-922 expression in human smooth muscle cells. Based on these findings we reasoned that EGF may play a role in facilitating homing of MSCs to “injured” areas of the heart, and further the effects of the cells. We sought to determine if MSCs amplify coronary collateral growth, and if enhancing adherence of MSCs would further stimulate CCG. Our results support the concept that MSCs augment coronary collateral growth, and that if they are treated by EGF to enhance expression of adhesion molecules, both their homing to the myocardium and their ability to increase collateral growth are increased.
CECs were purchased from VEC technologies Inc and were cultured in MCDB-131 complete medium (VEC technologies) medium. The cells were used from passage 3-7.
MSCs were harvested from the bone marrow of the femurs and tibias of 6- to 12-month-old Lewis rats by inserting a 21-gauge needle into the shaft of the bone and flushing it with 30 ml of complete α-modified Eagle's medium (MEM) containing 20% fetal bovine serum (FBS), 2 mM L-glutamine, 100 U/ml penicillin, 100 μg/ml streptomycin, and 25 ng/ml Amphotericin B. Cells were filtered through a 70-μm nylon filter and were plated into one 75 cm2 flask. The cells were grown in complete MEM at 37°C and 5% CO2 for 3 days, the medium was replaced with fresh medium, and the adherent cells were grown to 90% confluency to obtain samples defined as passage zero (P0) cells.23
Adhesion assays were performed using eight-well glass chamber slides (Falcon) on which CECs were plated, reaching confluence and labeled with DiO, 3,3′-dioctadecyloxacarbocyanine perchlorate. All experiments were performed with 1000 cells, MSCs labeled with 1,1′-dioctadecyl-3,3,3′,3′ tetramethylindocarbocyanine perchlorate (DiI; Molecular Probes). After incubation at 37°C (24 hours) cells were washed three times with PBS, and adherent cells were counted in 25 fields/well using epifluorescence microscopy (100×). The number of adherent cells was normalized to the total number of added cells. In some experiments, CECs and MSCs were pretreated for adhesion studies with one of the following substances EGF (10 ng, 50 ng, 100 ng, 1 mg per ml), EGF-R inhibitor AG1478 (1μM), cyclohexamide, an inhibitor of protein synthesis (CHX, 100 ng), anti-VCAM-1 or anti-ICAM-1 (10 μg/ml) hour before and during the adhesion assay.
After 90% of confluence for CECs and 60% for MSCs, the cells were starved for 48 hrs in MCDB-131 medium with 1% serum for CECs and α-MEM-media without serum for MSCs. Cultured cells were then stimulated with EGF (Sigma) for 16 hrs with and without AG1478 (Sigma Aldrich). Cells were harvested and subjected to western blot analysis using ICAM-1 and VCAM-1 antibodies (Santa Cruz).
Three to 4 month old, 300-350g, male Sprague Dawley (SD) rats were used for chronic (10 days) implantation of a pneumatic snare over the left anterior descending coronary artery (LAD), as described by Toyota et al24. MSCs were harvested, labeled with DiI and resuspended in PBS and 8 × 105 cells (in 500 μl over 20 sec) were injected into the LV lumen via (27 G needle) after the implantation of the occluder. As a control for the effect of cell infusion, PBS alone was delivered.
Rats were allowed to recover for 2 to 3 hours before the start of a 10 day repetitive ischemia protocol (RI) as described by Toyota et al24. Rats under RI protocol were divided into 3 groups: RI-control, RI+MSCs, RI+MSCs/EGF-activated. Experimental protocols were approved by the Institutional Animal Care and Use Committee of Northeastern Ohio Universities College of Medicine.
At the end of RI protocol (after 10 day of RI), radioactive microspheres (Perkin Elmer, 15.5 μm, 103Ru ~5 × 105) were injected into the left ventricle (LV) over 20s. Briefly the microspheres were agitated for 15 min, suspended in saline (total volume 150 μl) then injected directly into the left ventricle (LV) apex during LAD occlusion using a 27 G insulin syringe over a 10 s period. The normal and collateral-dependent zones were divided and total weight of each was measured. Radioactivity was measured using a gamma counter and collateral blood flow was measured as a ratio of the cpm/g to the two regions, i.e., CZ/NZ flow ratio.
Echocardiographic function was performed after 10 days of repetitive occlusion using a Toshiba Aplio (SSH770 with a fundamental frequency of 8.5, a 1 cm focus, and a frame rate of 300-500 frames/sec. Acquistion and analysis of the images were obtained blinded. For each measurement, three images were analyzed and averaged. LV wall thickness and cavity dimensions were measured by M-mode echocardiography, using American Society of Echocardiography guidelines. Measurements were made from a parasternal long axis view, or from a parasternal short-axis view (papillary level). From these M-mode measurements, LV mass, LV fractional shortening and EF were calculated. Per cent FS was calculated from (LVEDD−LVESD)/LVEDD × 100.
CECs or MSCs were stained with EGF, Phospho EGF-R (Cell signaling) antibody and secondary rabbit anti mouse coupled to Alexa Fluor (Molecular Probes). Cellular nuclei were counterstained with DAPI (Molecular Probes). Ten days after IR protocol the rats were anesthetized with pentobarbital and perfused through the abdominal aorta with ice-cold heparanized PBS followed by 4% paraformaldehyde. The hearts were embedded in Tissue-Tek OCT compound and sectioned at 5 μm using a cryostat and mounted onto slides for IHC. Using primary antibodies: rabbit polyclonal anti-human factor VIII (DAKO) to detect endothelial cells; mouse monoclonal anti-smooth muscle alpha-actin (clone 1A4; Sigma) to detect smooth muscle cells. Goat anti-mouse and chicken anti-rabbit AlexaFluor 594 were used as secondary antibodies. To detect the MSCs labeled with DiI engrafted into the myocardium, epifluorescence microscopy was used and numbers of cells were counted in microscopic fields and expressed as numbers of cells per field.
Data (mean ± SEM) were analyzed using one-way analysis of variance (ANOVA) followed by Student Newman-Keuls posthoc analysis for statistical comparisons between groups. P<0.05 was considered as significance.
Adhesion of MSCs to CECs was dependent on the period that the MSCs interacted with CECs reaching a maximum at 4 h of incubation (Figure 1A). Incubation of CECs with EGF at concentrations between 10 to1000 ng/ml for 16 h dose-dependently enhanced MSCs adhesion with a maximum at 100 ng/ml (Figure 1B).
Treatment of MSCs with 100 ng/ml EGF for 16 h (Figure 2A), increased adhesion to CECs (n=4, 70±7%) versus untreated controls (25±3%). When both MSCs and CECs were treated the adhesion was not different than treatment of either cell type. The effect of EGF on MSCs adhesion was EGF receptor-dependent, because AG1478 blocked adhesion. Immunofluorescence revealed EGF-R on endothelial cells and phosphorylation of EGF-R following EGF (Figure 2B). AG1478 inhibited phosphorylation of EGF-R. Figure 2C illustrates that the adhesion of MSCs to CECs induced by EGF was significantly reduced by CHX treatment (29% versus 63%) suggesting that EGF-induced adhesion requires new protein synthesis.
VCAM-1 and ICAM-1 antibodies attenuated EGF-induced adhesion of MSCs following treatment of either the MSCs or CECs (Figure 3A). Moreover, treatment of CECs or MSCs with EGF 100 ng induced the expression of ICAM-1 and VCAM-1 suggesting the importance of these molecules in MSCs adhesion (Figure 3B). AG1478 decreased EGF-induced expression of ICAM-1 and VCAM-1 suggesting that ICAM-1 and VCAM-1 induction is mediated by EGF-R signaling.
Ten days after RI, the untreated control-RI group exhibited decreased contractility and LV dilation (Figure 4 A). FS was higher in the group treated with MSCs compared to controls (Figure 4 B), and was further increased in the MSCs+EGF group. Administration of MSCs reduced LV cavity dilation as judged by LVEDD and LVESD compared to the controls. However, posterior wall thickness did not differ among the groups (Figure 4B).
Coronary collateral growth (indicated by CZ/NZ flow ratio) was increased by MSCs compared to controls. Importantly, CCG was higher in RI group injected with MSCs + EGF compared to untreated MSCs (Figure 5A).
To identify homing of MSCs in the myocardium, cells were visualized using fluorescence microscopy (for Dil). MSCs were found in the heart after 10 days of RI (Figure 5B-E). In non-ischemic tissue we observed few stem cells regardless of treatment: Control, 12±4 vs EGF-treated, 13±3 cells/microscopic field. In ischemic tissue many more of the cells were observed 57±18 cells/microscopic field (P<0.05 vs non-ischemic samples), and the number of cells in ischemic tissue was further augmented by EGF treatment (162±9 cells/field, P<0.05 vs non-ischemic samples and vs ischemic without treatment). Moreover, many of the engrafted MSCs were positive for α-actin SMC or Factor VIII, suggesting their integration in the growing blood vessels as smooth muscle and endothelial cells (Figure 6A-L). It was more difficult to ascertain the proportion of stem cells expressing markers for α-actin or Factor VIII, but generally speaking, the cells expressing these markers appeared to be located in coronary blood vessels. Tissue sections from hearts (hematoxylin and eosin staining) subjected to these treatments are shown in the online supplement.
In the recent past, clinical trials have tested therapies involving VEGF or FGF to generate functional coronary collataral vessels,25 but unfortunately these trials failed. The use of stem cells to develop coronary collaterals has potential to provide cells, which can integrate in the new blood vessel, and serve as “paracrine factories” for angiogenic factors.26-29 Multipotent stromal cells represent an attractive candidate for cell-based therapies due to their self renewal capacity and multiple paths of differentiation.30 Moreover MSCs can be harvested from bone marrow, expanded in culture without losing their differentiation potential, and administered autologously to eliminate concerns with rejection.
In the current study we demonstrated that rats treated with MSCs, activated in vitro with EGF, demonstrated amplified homing, increased coronary collateral development and better cardiac function than those without treatment or with untreated MSCs. To our knowledge, this is the first study, reporting the importance of adhesion of MSCs in coronary arteriogenesis.
EGF enhanced adhesion of MSCs to CECs via EGF-R phosphorylation in vitro and the homing of stem cells to the heart. Inhibition of EGF receptor phosphorylation in MSCs and CECs blocked the adhesion of MSCs to CECs demonstrating the importance of EGF-R activation. Interestingly, EGF-R phosphorylation is increased in ischemic conditions.31 EGF treatment induced the expression of ICAM-1 and VCAM-1 on MSCs, amplifying adhesion of MSCs to endothelial cells and increasing homing and migration of MSCs to the ischemic territory. We believe the increased adhesion of MSCs produced the salubrious effect on cardiac function and arteriogenesis through increased numbers of cells homing to the ischemic/injured areas of the heart. Although our results support this concept there are some other possibilities. First, it is possible that EGF treatment increases survival of the MSCs, and if more cells survive in vivo, then we would observe more in the myocardium. We do not favor this explanation because the numbers of untreated and EGF treated MSCs were the same in the non-ischemic myocardium. If EGF enhanced survival then we would expect more cells in all locations, but this was not the observation. Second, treatment of MSCs with EGF could enhance their “crosstalk” with resident cells, e.g., smooth muscle, endothelium, and facilitate their transdifferentiation into, or fusion with, another cell type. We have no way of unequivocally addressing this possibility, and enhanced crosstalk may contribute to the mechanims of EGF facilitation of the MSCs therapeutic.
HB-EGF (heparin-binding epidermal growth Factor) was reported to participate in angiogenesis and recruitment of MSCs during embryonic development.17,32,33 However, angiogenesis and arteriogenesis stimulated by ischemia were normal in HB-EGF-/- null mice.31 In this report, HB-EGF-/- mice showed a residual level of EGF-R activation suggesting that other EGF-R ligands could compensate for EGF, which could have masked the contribution of HB-EGF in ischemia-induced angiogenesis or arteriogenesis.
MSCs were reported to regenerate infarcted myocardium;6,34 however, mechanisms underlying therapeutic effects have not been defined with an intense debate over differentiation, paracrine function, and fusion.6,34-38 There is even further debate about the time of administration, as treatment during the time of any surgical procedure (as in our study) could decrease homing to the heart because the cells are being attracted to sites of the surgical wounds. Nonetheless, we still observed significant number of cells in the myocardium subjected to repetitive ischemia. Importantly, we observed co-localization of MSCs with α-actin and Von Willebrand factor suggesting that MSCs participated in the growth of the coronary vasculature, although admittedly we cannot determine if the vessels with MSCs are collaterals.
One final aspect of our results warrants discussion. In previous reports24,39 we have observed that following RI, there is robust growth of coronary collaterals in response to RI without any treatment. In our present study the growth of coronary collaterals in response to RI without any treatment was less than what we have reported. We believe the reason for this discrepancy is related to strain differences among rats. Along this line, different strains of mice show extreme variations in innate collateral development,40 which also then impacts on the eventual development of the collateral circulation.41
Stem cell therapies hold promise for the treatment of ischemic cardiovascular disease. Our study is the first to show that EGF-R activation of MSCs enabled better homing and increased coronary collateral growth in a model of repetitive myocardial ischemia. The betterment of collateral development was also associated with an inhibition of LV remodeling and improved cardiac function. These findings provide a new paradigm for amplifying the effects of administered MSCs (increasing their homing) and this combination of MSCs with EGF may be useful for the treatment of ischemic heart disease.
The authors would like to thank Jean Carnal and Elizabeth McIlwain for their excellent technical assistance. This work was supported by NIH grants HL32788, HL34286, HL 073755 and P40-RR17447.