Intracoronary delivery of extracellular vesicles from human cardiac progenitor cells reduces infarct size in porcine acute myocardial infarction

Abstract

Small extracellular vesicles (sEVs) are non-replicating, membrane-derived nanoparticles secreted by cells into the extracellular space.1,2 They carry bioactive molecules that can modulate recipient cell functions. When derived from human cardiac progenitor cells (CPCs) or cardiosphere-derived cells (CDCs), sEVs are hypoimmunogenic and cardioprotective and exhibit pro-angiogenic capacity, making them attractive candidates for the treatment of acute myocardial infarction (AMI). We and others have shown that sEVs secreted by human CPCs or CDCs reduce infarct size and improve cardiac function in AMI models3–6; however, benefits have only been demonstrated after their intramyocardial (IM) injection in open-chest AMI models. Since most AMI patients undergo emergency percutaneous coronary intervention (PCI), clinical translation of sEV-based therapies will require percutaneous delivery modalities. While intracoronary (IC) delivery of sEVs during PCI would be suitable for AMI patients, a previous study in porcine AMI showed a lack of efficacy.5 In contrast, percutaneous, catheter-based IM delivery of CDC-sEVs using the NOGA cardiac navigation system showed promise in a porcine model of convalescent myocardial infarction.5 We here evaluated the feasibility, safety, and efficacy of percutaneous, catheter-based IM and IC delivery of CPC-sEVs in a translational AMI porcine model. A CPC master cell bank was established from right atrial appendage tissue of a male patient (83 years old) undergoing aortic valve surgery. The protocol was approved by the Institutional Ethics Committee (IEC). The sEVs were isolated from CPC-conditioned medium by tangential flow filtration using a Good Manufacturing Practice (GMP) protocol7,8, and stored at −80°C. Quality control included nanoparticle tracking analysis and sEV marker analysis. The sEV markers CD63, CD81, tumor susceptibility gene-101 (TSG101), and syntenin-1 were investigated by Western blotting; glucose-regulated protein-94 (GRP94) was used as a marker for residual cytosolic components in sEV preparations.1 The animal studies were approved by the IEC (University of Kaposvar, Hungary). Closed-chest, 90 min mid-left anterior descending artery (LAD) balloon occlusion, ischaemia-reperfusion AMI was induced in 17 pigs (body weight: 29.6 ± 0.8 kg), as described previously.9 Within 15 min of reperfusion, animals received either IM sEVs [n = 6, 11 mg (≈10 × 1011 particles) in 3 mL Plasma-Lyte A, 10 injections of 300 µL each] using the NOGA system or IC sEVs [n = 6; 22 mg (≈20 × 1011 particles) in 20 mL Plasma-Lyte A] or vehicle (n = 5, control) selectively instilled into the LAD at 80 mL/h over 15 min. Cardiac magnetic resonance imaging (cMRI) was performed at baseline, 2–3 days, and 1 month after treatment (Figure 1D). Intracoronary (IC) delivery of clinical-grade human cardiac progenitor cell (CPC)–derived small extracellular vesicles (sEVs) reduces infarct size and improves left ventricular ejection fraction (LVEF) in porcine acute myocardial infarction (AMI). (A) CPC–derived sEVs were isolated using a large-scale, Good Manufacturing Practice compliant protocol. Three lots of sEVs (TFF11–13) manufactured from the same master cell bank were used for porcine studies. The mean particle diameter in the three lots was 121 ± 6.2 nm, 143 ± 7.4 nm, and 131 ± 8.7 nm, respectively. (B) sEV concentrations (measured as TSG101 levels) and contaminants (free total protein) in CPC conditioned medium (CM) and final sEV formulations. The latter were depleted of free proteins. (C) Western blot analysis of sEVs from the three lots showed comparable levels of the sEV surface–expressed tetraspanins CD63 and CD81, as well as of the luminal sEV markers TSG101 and syntenin-1. GRP94 (used as a marker for cytosolic components) was not detectable. These data show homogeneous EV composition among the different sEV formulations used in the porcine AMI experiments. (D) In the first study, AMI was induced in 19 pigs, of which 2 pigs died before sEV treatment. The remaining 17 pigs randomly received either intramyocardial (IM) sEVs by NOGA-guided injection (IM; n = 6), IC sEVs (IC; n = 6), or vehicle (Ctrl; n = 5) selectively instilled into the left anterior descending artery (LAD). Cardiac magnetic resonance imaging (cMRI) was performed before treatment (baseline), at 2–3 days, and at 1 month after treatment. One pig died due to cardiac arrest during IM sEV delivery (§), and two IC sEV-treated pigs died during anaesthesia induction for cMRI at 2–3 days ($). The remaining 14 pigs completed the study. (E–G) At 1 month after treatment, LVEF (%) tended to be higher after IC sEV delivery (P = .137), compared with IM sEV delivery and control. LVEF increased from 2–3 days to 1 month after IM or IC sEV delivery, but not in controls. Infarct size (IS, %) tended to be lower after IC sEV treatment (P = .10) compared to IM sEV treatment and control. Infarct volume (IV, mL) was significantly lower after IC sEV treatment (P = .044) when compared with IM sEV treatment and controls (*P < .05). (H) In the second study, AMI was induced in 11 pigs which were randomly assigned to either receive IC sEVs (n = 6) or vehicle (Ctrl; n = 5) and were followed-up for 3 months with serial cMRI. One pig (control) died during anaesthesia induction for cMRI at 2–3 days ($). Ten animals completed the study. (I–K) LVEF was significantly higher after IC sEV treatment, compared to control (P = .019). Infarct size (IS, %) and infarct volume (IV, mL) were significantly reduced after IC sEV treatment (P = .005 and P = .046, respectively). Infact size (IS, %) significantly decreased from 2–3 days to 3 months after IC sEV therapy (P = .002), but not in controls (*P < .05 and **P < .01). (L) Representative cMRI images of long-axis late gadolinium enhancement in a control and in a sEV-treated pig at 2–3 days and 3 months after treatment. (M and N) Representative images of myocardial fibrosis by Elastica van Gieson staining and blood vessels by CD31 (PECAM-1, 1:400, Novus Biological; detection: Dako RB Envision) and alpha-smooth muscle actin staining (αSMA, 1:400, Dako; detection: universal HRP-Polymer Kit) of paraffin-embedded samples from the infarct border zone of a control and an IC sEV-treated pig at 3 months, respectively. The sEV therapy significantly reduced myocardial fibrosis and increased vessel density in the infarct border zone. The degree of myocardial fibrosis and CD31-positive vascular structures were randomly counted on five high-power fields per section in controls (n = 4) and sEV-treated pigs (n = 6) after tissue harvest at 3 months. Images were acquired using digital slide scanners (Zeiss Axio Scan.Z1, Carl Zeiss/Switzerland, NanoZoomer 2.0 HAT, Hamamatsu Photonics/France) and analyzed using QuPath software (v0.3.2, Queen’s University, Northern Ireland). Results are expressed as collagen content (% area) and vessel number/mm². Variable distribution was assessed using Kolmogorov–Smirnov and Shapiro–Wilk tests. Normally distributed variables (infarct size, IS, %; infarct volume, IV, mL) were expressed as mean ± standard deviation and analyzed by Student’s t-test (Control vs. IC) or analysis of variance with post-hoc Bonferroni test (Control vs. IC vs. IM) or analysis of variance for repeated measures (paired longitudinal analyses). Not normally distributed variables [LVEF (%), collagen content (%), and vessel density (mm2)] were expressed as median and interquartile range and analyzed by Mann–Whitney test (Control vs. IC) or Kruskal–Wallis test (Control vs. IC vs. IM) or Friedman test (paired longitudinal analyses). Significance was considered for P-value <.05 (*), < .01 (**), or <.001 (***). LVEF, left ventricular ejection fraction; AMI, acute myocardial infarction; IS, infarct size (% of LV mass); IV, infarct volume (mL); Ctrl, vehicle treatment; IM, IM sEV treatment; and IC, IC sEV treatment AMI was induced in 11 pigs, which were then randomized to receive either IC sEVs [n = 6; 22 mg (≈20 × 1011 particles)] or vehicle (n = 5). The cMRI was performed at 2–3 days, 1 month, and 3 months after treatment (Figure 1H). Infarct border zone samples were paraffin-embedded and stained with hematoxylin-eosin or Elastica van Gieson in order to assess the degree of myocardial fibrosis. Blood vessels were analyzed by CD31 and alpha-smooth muscle actin (αSMA) immunostaining.9 Statistical analysis was performed using GraphPad Prism (v9.0.0) and IBM SPSS (v25). A detailed description (variable distribution and statistical tests) is provided in the legend to Figure 1. The full report is available at https://github.com/CardiovascularTheranostics/EXOpig. Figure 1A–C summarizes sEV characteristics. Baseline cMRI data were comparable between groups. At 2–3 days post-treatment, cardiac output [CO (L/min); 1.9 ± 0.5L/min; 2.8 ± 0.7L/min; 2.9 ± 0.4L/min; P = .036] and right ventricular ejection fraction (RVEF, %; 26.2 ± 8.1%; 48.5 ± 3.6%; 46.4 ± 11.2%; P = .003) were decreased in IM sEV-treated pigs compared to IC sEV-treated pigs and controls, respectively. Left ventricular stroke volume [LVSV (mL); 26.2 ± 3.3mL; 33.5 ± 10.4mL; 37.8 ± 9.0mL; P = .107] also showed a trend towards a decrease in IM sEV-treated pigs. These results may reflect procedure-related needle injury after IM sEV injection. At 1 month post-treatment, there was a trend towards an increase in left ventricular ejection fraction (LVEF; %) in IC sEV-treated pigs compared to IM sEV-treated pigs and controls [47.7% (39.4%; 53.4%); 39.6% (34.0%; 44.5%); 43.6% (33.0%; 45.3%), respectively; P = .137] (Figure 1E–G). LVSV increased concomitantly from 2–3 days to 1 month post-treatment in IC sEV-treated pigs and IM sEV-treated pigs, but not in controls (+21.7 ± 4.7%; + 34.2 ± 13.7%; −14.9 ± 10.1%, respectively; P < .001). At the endpoint, there was a trend towards a decrease in infarct size (IS, % of left ventricular mass) in IC sEV-treated pigs and IM sEV-treated pigs compared with controls (15.3 ± 3.3%; 20.0 ± 8.4%; 29.3 ± 12.1%, respectively; P = .100). Infarct volume (IV, mL) was reduced in IC sEV-treated pigs compared to IM sEV-treated pigs and controls (13.5 ± 2.5mL; 17.6 ± 8.5mL; 29.1 ± 10.7mL, respectively; P = .044; Figure 1E–G). Using pairwise comparisons, IC sEV significantly reduced infarct volume at 1 month compared to controls (13.5mL vs. 29.1mL; P = .040), whereas IM EV did not (17.6mL vs. 29.1mL; P = .153). These results suggested a favorable efficacy profile for IC sEV delivery, which was therefore evaluated in a randomized study. At the 3-months endpoint of the randomized, placebo-controlled study, the LVEF was significantly higher in IC sEV-treated pigs compared to controls [36.2% (35.1%; 37.1%); 33.4% (32.3%; 35.0%); P = .019]. Infarct size was reduced in IC sEV-treated pigs compared to controls (18.0 ± 2.4%; 23.9 ± 2.4%; P = .005). Similar results were found for infarct volume (18.0 ± 4.4mL vs. 24.3 ± 3.7mL; P = .046). Infarct size decreased from 2–3 days to 3 months post-treatment in IC sEV-treated pigs, but not in controls (−26.5 ± 8.8%; +2.5 ± 11.4%; P = .002). During the same time interval, infarct volume decreased in IC sEV-treated pigs (−8.7 ± 21.5%), whereas it increased in controls (+16.8 ± 24.1; P = .119) (Figure 1H–K). These results suggest an encouraging efficacy profile for IC sEV administration. No adverse events related to the sEV therapy were observed during follow-up in either study, suggesting an encouraging safety profile. In addition, IC sEV delivery significantly reduced myocardial fibrosis while increasing blood vessel density at 3 months (Figure 1L–N). We evaluated percutaneous, catheter-based IM and IC delivery of GMP-grade CPC-sEVs in a translational AMI porcine model. IC sEV administration demonstrated beneficial effects including scar reduction, functional improvement, neoangiogenesis, and anti-fibrotic effects in the heart. Conversely, IM sEV delivery was associated with early functional impairment, likely due to needle injury. Gallet et al. previously evaluated CDC-derived sEV delivery in porcine AMI.5 They reported a lack of effects on infarct size and LVEF at 48 hours after IC sEV delivery, whereas IM sEV delivery was effective.5 While their and our studies used similar IC delivery protocols and sEV dosages, they are methodologically not comparable because Gallet et al.5 addressed very early effects of IC sEV delivery (48 hours), whereas we evaluated late effects (up to 3 months). Importantly, we also show that sEVs can be manufactured as allogeneic, off-the-shelf, ready-to-use, GMP-grade products, making them suitable for AMI patients undergoing PCI. Our study has several limitations. First, the sample size was small and confirmatory studies are needed. Second, although we used a translational AMI model, the effects of sEVs may differ in elderly patients with comorbidities. Third, given the translational focus of our study, evaluation of cardiac sEV uptake, biodistribution, and mechanisms of action were beyond the scope of the present study and need to be addressed in future studies. Finally, systemic effects of IC sEV delivery may have contributed to the observed benefit. Pending validation by further studies, catheter-based IC delivery of clinical-grade CPC-sEVs may warrant phase 1 clinical evaluation in AMI patients. Viviana Lo Cicero, Andrea Brambilla, and Sabrina Soncin provided technical assistance in the preparation and characterization of the sEVs. All authors declare no conflict of interest for this contribution. Outside this contribution V.F. has relevant (institutional) financial activities with following commercial entities: Medtronic GmbH, Biotronik SE & Co., Abbott GmbH & Co. KG, Boston Scientific, Edwards Lifesciences, Berlin Heart, Novartis Pharma GmbH, JOTEC GmbH, and Zurich Heart in relation to educational grants (including travel support), fees for lectures and speeches, fees for professional consultation and research and study funds. Data are available on reasonable request. G.V. and M.Y.E. were supported by two research grants from the Swiss Heart Foundation, Berne, Switzerland. L.B. was supported by a research grant from the Swiss National Foundation (EXaCT, ID:182948). G.V. was supported by the Cecilia Augusta Foundation, Lugano, Switzerland. All pig experiments were approved by an IEC (University of Kaposvar, Hungary) and conducted in accordance with the Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines. None supplied.

Topics & Concepts