Imaging survival and function of transplanted cardiac resident stem cells

Abstract

Objectives: The goal of this study is to characterize resident cardiac stem cells (CSCs) and investigate their therapeutic efficacy in myocardial infarction by molecular imaging methods.

Background: CSCs have been isolated and characterized in vitro. These cells offer a provocative method to regenerate the damaged myocardium. However, the survival kinetics and function of transplanted CSCs have not been fully elucidated.

Methods: CSCs were isolated from L2G85 transgenic mice (FVB strain background) that constitutively express both firefly luciferase and enhanced green fluorescence protein reporter gene. CSCs were characterized in vitro and transplanted in vivo into murine infarction models. Multimodality noninvasive imaging techniques were used to assess CSC survival and therapeutic efficacy for restoration of cardiac function.

Results: CSCs can be isolated from L2G85 mice, and fluorescence-activated cell sorting analysis showed expression of resident CSC markers (Sca-1, c-Kit) and mesenchymal stem cell markers (CD90, CD106). Afterwards, 5 x 10(5) CSCs (n = 30) or phosphate-buffered saline control (n = 15) was injected into the hearts of syngeneic FVB mice undergoing left anterior descending artery ligation. Bioluminescence imaging showed poor donor cell survival by week 8. Echocardiogram, invasive hemodynamic pressure-volume analysis, positron emission tomography imaging with fluorine-18-fluorodeoxyglucose, and cardiac magnetic resonance imaging demonstrated no significant difference in cardiac contractility and viability between the CSC and control group. Finally, postmortem analysis confirmed transplanted CSCs integrated with host cardiomyocytes by immunohistology.

Conclusions: In a mouse myocardial infarction model, Sca-1-positive CSCs provide no long-term engraftment and benefit to cardiac function as determined by multimodality imaging.

Figure 1. Isolation and culturing of cardiac stem cells

(A) Ventricular cells from L2G85 transgenic mice stained for Sca-1 (red) (I). After a period ranging from 1 to 3 weeks, phase-bright cells migrated over a layer of fibroblast-like cells (II). The phase-bright cells were collected and Giemsa stain showed the cell with large nucleus (III). Scale bar=50μm (i, ii), 2μm (III). (B) Subculture and in vitro differentiation of phase-bright cells. Morphology of CSCs (I) cultured in poly-D-lysine-coated plates and expressing GFP (II). With cardiac differentiation medium, some CSCs can form cardiac sphere (arrow) with GFP expression (III, IV). (C) Proliferation curves show linear growth of cultured CSCs over a one month period. (D) & (E) Ex vivo imaging analysis of CSCs show increasing bioluminescence signals with cell numbers (r²=0.98)

Figure 2. Characterization of surface markers in cardiac stem cells

Quantification by FACS analysis of CSC phase-bright cells and its derived cells. Bone marrow mononuclear cells derived mesenchymal stem cells (BM-MSC) were used as control. (A) CSCs express robust GFP both on phase-bright CSCs and their subcultures. After 2–3 passages, there was upregulation of Sca-1. Lower panels showed Sca-1 antibody isotype control. (B) Quantitative analysis of cell markers expression by FACS. CSCs express high level mesenchymal stem cells markers, CD29, CD90, CD44, and CD106. After subculturing, Sca-1 up-regulated but c-Kit, CD34, CD31, and CD144 markers all down-regulated. Compare to BM-MSC, CSCs express less CD54, but higher CD29 and CD90. All FACS experiments were performed in triplicates.

Figure 3. Multipotent capacity of cardiac stem cells

(A) Cardiac and smooth muscle differentiation of CSCs in vitro. Immunostaining of GFP positive CSCs with cardiac troponin T (cTnT), myocyte enhancer factor 2C (MEF-2C), connexin-43, and α-smooth muscle actin (α-SMA). (B) Endothelial differentiation of CSCs in vitro. The cells were cultured in EGM-2 medium with 10 ng/ml VEGF and showed endothelial differentiation by morphology (I) and uptake of Dil-ac-LDL (II). Endothelial tube formation by differentiated CSCs after 12 hours of plating on Matrigel (III) whereas undifferentiated CSCs cells do not form cord-like structures (IV). (C) Oil red staining and RT-PCR analysis of PPARγ and lipoprotein lipase (LPL) expression shows adipogenic differentiation of the CSCs induced for 2 weeks. (D) Alizarin red S staining of calcium and RT-PCR analysis of osteocalcin expression shows CSCs induced to differentiate into osteoblasts.

Figure 4. Reporter gene imaging of CSCs fate after transplantation

(A) A representative animal injected with 5×10⁵ CSCs shows significant bioluminescence activity at day 2, which decreases progressively over the following 8 weeks. (B) Detailed quantitative analysis of signals from all animals transplanted with CSCs. Signal activity is expressed as photons/sec/cm²/sr. (C) Estimation of percent donor cell survival plotted as % signal activity (normalized to day 2) over the 8 week period following transplantation.

Figure 5. Echocardiographic evaluation of cardiac contractility

(A) Representative M-mode echocardiographic data of infarcted hearts receiving PBS vs. CSCs at day 2, week 4, and week 8. (B,C) Comparison of fractional shortening (FS) and ejection fraction (EF) between the two groups 7 days before (baseline), 2 days, 28 days, and 56 days after LAD ligation.

Figure 6. [¹⁸F]-FDG PET imaging of cardiac viability

(A) Representative image in a normal mouse heart and at day 2, day 28, and day 56 in infarcted hearts receiving PBS vs. CSCs. *P=0.46 and P=0.38 vs. PBS group at day 28 and day 56 respectively. (B) Detailed quantitative analysis of signals from all animals transplanted with CSCs and PBS group. There is no significant difference of the FDG uptake (% ID/g) between the two groups. *P=0.31 and P=0.40 vs. PBS group at day 28 and day 56 respectively. (C–D) Representative polar map of the microPET images obtained from mice treated with PBS vs. CSC. Measurements are based on 50% thresholds.

Figure 7. Tracking of grafted CSCs by immunofluorescence

(A – B) CSCs within the recipient myocardium 3 days after injection shown at low and high magnification. (C–D) Transplanted CSCs can differentiate and integrate with host myocardium as confirmed by GFP and α-sarcomeric actin (α-SA) double staining. At day 14, CSCs could differentiate into cardiomyocytes as confirmed by α-SA and GFP double stainings (Figure 7C). However, this population became significantly decreased when the tissues were examined at day 28 (Figure 7D), which is also consistent with the decrease in bioluminescence signals over this period of time. Scale bar=100μm (A), 20μm (B, C, D).