Consistent Long-Term Therapeutic Efficacy of Human Umbilical Cord Matrix-Derived Mesenchymal Stromal Cells After Myocardial Infarction Despite Individual Differences and Transient Engraftment

Abstract

Human mesenchymal stem cells gather special interest as a universal and feasible add-on therapy for myocardial infarction (MI). In particular, human umbilical cord matrix-derived mesenchymal stromal cells (UCM-MSC) are advantageous since can be easily obtained and display high expansion potential. Using isolation protocols compliant with cell therapy, we previously showed UCM-MSC preserved cardiac function and attenuated remodeling 2 weeks after MI. In this study, UCM-MSC from two umbilical cords, UC-A and UC-B, were transplanted in a murine MI model to investigate consistency and durability of the therapeutic benefits. Both cellular products improved cardiac function and limited adverse cardiac remodeling 12 weeks post-ischemic injury, supporting sustained and long-term beneficial therapeutic effect. Donor associated variability was found in the modulation of cardiac remodeling and activation of the Akt-mTOR-GSK3β survival pathway. In vitro, the two cell products displayed similar ability to induce the formation of vessel-like structures and comparable transcriptome in normoxia and hypoxia, apart from UCM-MSCs proliferation and expression differences in a small subset of genes associated with MHC Class I. These findings support that UCM-MSC are strong candidates to assist the treatment of MI whilst calling for the discussion on methodologies to characterize and select best performing UCM-MSC before clinical application.

Keywords: Wharton's jelly; cardiac fibrosis; cell therapy; donor variability; mesenchymal stromal (or stem) cells; myocardial infarction; regeneration/repair; umbilical cord matrix derived mesenchymal stromal cells (hUCM-MSCs).

Figure 1

Long term improvement in left ventricle systolic function 12 weeks after MI and UCM-MSC delivery from the 2 donors. (A) Representative PSLAX views in B- and M-mode. (B) Ejection fraction and (C) fractional shortening calculated using the Simpson's method (PSLAX B-mode). (D) Cardiac output determined through the quantification of the left ventricular outflow tract stroke volume (LVOT SV) in PW-doppler mode, aortic root diameter and Heart rate. (E) Myocardial Performance Index was assessed computing exclusively PW-doppler data (Mitral Valve Closure to Opening Time (MVCOT) and LV Ejection time). Mean ± SD.

Figure 2

UC-B and, at less extent UC-A, decrease cardiac remodeling 12 weeks after MI. (A) Representative transverse histological sections of hearts stained with Masson's Trichrome of each group. Collagen deposition (scarring) is identified in blue. (B) Left ventricular (LV) infarcted area, (C) infarcted midline, and (D) LV lumen dilation were quantified using the MIQuant Software. (E) Infarcted wall thickness (mm). (F) Representative immunofluorescence images of CD31 detection in the infarcted tissue (autofluorescence in green). (G) Percentage of CD31-expressing cells in the infarcted area and (H) borderzone. Each point represents the mean of the quantification for multiple 300 μm apart sections per heart.

Figure 3

UCM-MSC delivered into the LV show similar engraftment following myocardial infarction (MI) but differentially activate signaling pathways. (A) Representative images of time-course bioluminescence detection of luciferase expressing UC-A and UC-B delivered into the LV tissue after MI. (B) Average radiance determined in the thoracic cavity area using equivalent regions of interest for all animals in the study. Data shown as Average radiance ±SEM normalized to measurements of each animal at day 1. (C) Western-blot analysis of the LV borderzone at day 2 post MI and UCM-MSC myocardial delivery (blots shown for 3 animals per group; n = 6 per study group).

Figure 4

Transcriptomic profiles of UC-A and UC-B in normoxia and in response to hypoxia. (A) Schematic overview of the experiment. (B) Unsupervised hierarchical clustering of RNA-Seq datasets obtained from 3 independent experiments of each conditionUCM-MSC line. (C) Principal component analysis (PCA) of the 12 RNA-Seq datasets showing an overlap on PCA1 and clustering on PCA2. (D) Number of total differentially expressed genes (DEG) and of genes up- and down- regulated in UC-B compared to UC-A, under normoxia and hypoxia conditions (fold-change>±2, FDR < 0.05). (E) Volcano Plot of the DEG highlighting the statistically significant matrissome-associated genes. (F) Heatmap hierarchically organized by combined ranked score of DEG in normoxia conditions alongside the corresponding GO and KEGG pathways enrichment analysis. (G) Analysis of the DEG in normoxia vs. hypoxia for each UCM-MSC donor. Heatmap hierarchically organized by combined ranked score of the common DEG altered in the two donors in response to hypoxia; enriched GO and KEGG pathways are plotted.

Figure 5

UC-A and UC-B display equivalent in vitro angiogenic potential on cardiac microvascular endothelial cells. (A) Schematic overview of the experiment. (B) In vitro tubulogenesis assay of human myocardial microvascular endothelial cells. Representative images of the formed tubular-like structures 7.5 h after incubation with the conditioned media of the 4 experimental groups. EBM (basal media) and EGM-MV (complete media) controls are shown. Scale bar: 200 μm. (C) (i) Total tube number. (C) (ii) Total Tube length. (C) (iii) Number of Junctions. Results of 3 independent experiments are depicted. (*p < 0.05 and **p < 0.01) (D) Gene expression of N-Cadherin in UCM-MSC from the two donors from 3 independent hypoxia induction experiments (p = 0.05). n.s. - non-significant.