A total of 56 rats were enrolled and underwent MI. Two of the 56 animals died after MI before group assignment. The remaining 54 rats were assigned to one of the 4 treatment groups (vehicle, MSCs, CMCs, or CPCs). Six of these animals (5 in the MSC group and 1 in the CMC group) died after cell infusion. Thus, a total of 48 rats (13 in the vehicle group, 11 in the MSC group, 11 in the CMC group, and 13 in the CPC group) completed the protocol and were included in the final analysis (Supplementary Table I).
There were no significant differences in body weight among the four groups throughout the experimental protocol (Supplementary Table II). Similarly, there were no significant differences among the groups with respect to LV weight, LV volume, or spleen weight index measured at postmortem examination (Supplementary Table II).
Effect of intravenous administration of cells on LV function measured by echocardiography. Echocardiographic measurements are summarized in Supplementary Table III; representative echocardiographic images are illustrated in Supplementary Fig. I. Before treatment (30 days after MI, Pre-Rx), LV ejection fraction (EF) dropped ≥ 20 points from the baseline value (before MI) in all rats (Supplementary Fig. II); the average drop was similar among groups (-23.0 ± 0.6, -22.3 ± 0.4, -24.2 ± 1.1, and − 24.2 ± 1.0, respectively, in the vehicle-, MSC-, CMC-, and CPC-treated group)(Supplementary Fig. II B). At this time point, LV end-diastolic diameter (LVEDD), area (LVAd), and volume (LVEDV), and end-systolic diameter (LVESD), area (LVAs), and volume (LVESV) were markedly increased from baseline (Supplementary Table III), whereas infarcted LV wall thickness at end-diastole (IWTd) and end-systole (IWTs), fractional shortening (FS), and fractional area change (FAC) were markedly decreased (Supplementary Table III). There were no significant differences among the vehicle, MSC, CMC, and CPC groups in any of these variables, indicating that, before treatment, the severity of post-MI LV remodeling and dysfunction was comparable in all groups.
At 35 days after treatment (Post-Rx), however, the four groups exhibited a different course. As expected [6, 9, 10], the vehicle-treated group showed no improvement and/or further deterioration of LV function (Fig. 2). Specifically, there was a significant further decline in LV EF (-2.1 ± 0.5 %, P < 0.05, Fig. 2B) and FAC (-1.46 ± 0.63%, P < 0.05, Fig. 2D); there was also a nominal increase in LVESV (+ 8.73 ± 4.67 µL, P = NS, Fig. 2F) and decrease in stroke volume (SV, -2.86 ± 4.12 µL, P = NS, Fig. 2H) compared with the pretreatment (Pre-Rx) values at 30 days after MI (Fig. 2).
In contrast to the further deterioration of LV function in the vehicle group, intravenous administration of the three types of cells produced a variable degree of improvement in global LV function (Fig. 2). Administration of MSCs improved LV EF (+ 0.50 ± 0.51% vs. -2.10 ± 0.48% in the vehicle group, P < 0.01, Fig. 2A&B) but failed to significantly alter FAC (-0.43 ± 0.36 vs. -1.46 ± 0.63%, P = NS, Fig. 2D), SV (+ 9.81 ± 6.26 vs. -2.86 ± 4.12 µL, P = NS, Fig. 2H), and LV ESV (+ 6.74 ± 5.26 vs. +8.73 ± 4.67 µL, P = NS, Fig. 2F) compared with the vehicle group.
Administration of CMCs improved both LV EF (+ 2.74 ± 0.43%, P < 0.01, Fig. 2A&B) and FAC (+ 2.58 ± 0.32%, P < 0.01, Fig. 2C&D) compared with vehicle; there was also a non-significant decrease in LV ESV (+ 0.28 ± 6.99 µL, P = NS, Fig. 2F) and increase SV (+ 14.58 ± 8.19 µL, P = NS, Fig. 2H) compared with the vehicle group. In addition, compared with intravenous infusion of MSCs, intravenous administration of CMCs produced a slightly, but significantly, greater improvement in EF (+ 2.74 ± 0.43 % vs. +0.50 ± 0.51%, respectively, P < 0.01, Fig. 2B) and FAC (+ 2.58 ± 0.32 % vs. -0.43 ± 0.36%, P < 0.05, Fig. 2D).
Intravenous administration of CPCs resulted in superior beneficial effect on LV function (Fig. 2); compared with the vehicle group, there was a marked increase in LV EF (+ 6.51 ± 0.36%, P < 0.01, Fig. 2A&B), FAC (+ 4.98 ׅ± 0.9%, P < 0.01, Fig. 2C&D), and SV (+ 27.54 ± 7.14 µL, P < 0.01, Fig. 2G&H) and decrease in LV ESV (-7.76 ± 5.95 µL, P < 0.05, Fig. 2F). Furthermore, compared with the MSC- and CMC-treated groups, intravenous administration of CPCs produced a greater improvement in LV EF (P < 0.01 for both the MSC- and CMC-treated group, Fig. 2A&B), FAC (P < 0.01 and P < 0.05, respectively, for the MSC- and CMC-treated group, Fig. 2C&D), and SV (P < 0.05 for the MSC-treated group, Fig. 2G).
Effect of intravenous administration of cells on LV function measured by hemodynamic studies. As an additional assessment of LV function, hemodynamic studies were performed at 35 days after treatment (prior to euthanasia). The hemodynamic parameters are summarized in Fig. 3 and Supplementary Table IV. As shown in Fig. 3, intravenous administration of MSCs did not significantly alter hemodynamic variables, although there was a slight trend toward improvement (Fig. 3 and Supplementary Table IV). Intravenous infusion of CMCs improved significantly LV dP/dtmax (8337 ± 150 mmHg/s in the CMC-treated group vs. 7543 ± 95 in the vehicle group, P < 0.01, Fig. 3D), dP/dtmin (-7867 ± 104 vs. -7066 ± 221, P < 0.01, Fig. 3D), and dP/dtmax-EDV (33.5 ± 1.8 vs. 27.3 ± 1.0 mmHg/s/µL, P < 0.05, Fig. 3F), but other hemodynamic parameters were not affected compared with the vehicle group (Fig. 3). Similar to the echocardiographic data, hemodynamic studies also revealed that intravenous administration of CPCs produced superior improvement in LV function; thus, compared with the vehicle group, hemodynamic parameters (SV [136.2 ± 4.8 µL in the CPC-treated group vs. 117.6 ± 4.0 µL in the vehicle group, P < 0.05, Fig. 3B], stroke work [SW, 13.5 ± 0.4 vs. 10.7 ± 0.3 mmHg*µL, P < 0.01, Fig. 3C], dP/dtmin [-7934 ± 111 vs. -7066 ± 221 mmHg/s, P < 0.01, Fig. 3D], EF [54.1 ± 0.5 vs. 49.8 ± 0.7%, P < 0.01, Fig. 3E], dP/dtmax-EDV [32.0 ± 1.2 mmHg/s/µL, P < 0.05, Fig. 3F], end-systolic elastance [0.86 ± 0.02 vs. 0.73 ± 0.03 mmHg/µL, P < 0.05, Fig. 3G], and Tau_w [10.7 ± 0.2 vs. 11.7 ± 0.2 ms, P < 0.05, Fig. 3H]) were significantly improved, although no significant difference was found among the three cell-treated groups.
In summary, two independent methods of functional assessment (echocardiography and hemodynamic studies with a conductance catheter) demonstrated that intravenous administration of three types of cells at a same dose (12 x 106 cells) in rats with chronic MI yielded variable degrees of improvement in LV function. The extent of LV functional improvement (measured by an array of echocardiographic parameters [Fig. 2] - EF, FAC, ESV, and SV - and hemodynamic variables [Fig. 3] - LV EDP, LV dP/dt, dP/dtmax-EDV, SV, SW, Tau_w, EF, and elastance) showed a pattern of a gradient effect, being least with MSCs, intermediate with CMCs, and maximal with CPCs.
LV morphology. Masson’s trichrome-stained LV sections were analyzed to determine the risk region, scar size, and volume of viable myocardium. As shown in Fig. 4A, the risk region was similar among the 4 groups. None of the 3 types of cells significantly modified scar size or the volume of viable myocardium when compared with the vehicle group, although there was a trend for MSCs and CPCs to reduce scar size and increase viable myocardium (Fig. 4A).
Picrosirius red-stained LV sections were analyzed to determine the effect of the 3 types of cells on myocardial collagen deposition. As shown in Fig. 4B, the collagen content in both the risk and noninfarcted region was not significantly modified by any of the cells given intravenous 35 days prior when compared with the vehicle group, although it tended to be reduced by MSCs and CPCs.
Compared with vehicle, CD45 positive cells tended to be lower in the risk region of rats treated with CMCs and CPCs (Fig. 5A) and in the noninfarcted region of rats treated with all three types of cells (Fig. 5B) than vehicle, but the difference was not statistically significant.
Profile of inflammatory cells in blood. As shown in Fig. 6, blood levels of inflammatory cells including WBCs, neutrophils (Ne), lymphocytes (Ly) as well as the Ne/Ly ratio and the platelet/Ly ratio were similar among groups at baseline (before cell infusion). Intravenous administration of the 3 types of cells did not result in significant changes in the inflammatory cell profile at 48 h or 35 days after cell treatment.