Prolonged survival of transplanted stem cells after ischaemic injury via the slow release of pro-survival peptides from a collagen matrix

Abstract

Stem-cell-based therapies hold considerable promise for regenerative medicine. However, acute donor-cell death within several weeks after cell delivery remains a critical hurdle for clinical translation. Co-transplantation of stem cells with pro-survival factors can improve cell engraftment, but this strategy has been hampered by the typically short half-lives of the factors and by the use of Matrigel and other scaffolds that are not chemically defined. Here, we report a collagen-dendrimer biomaterial crosslinked with pro-survival peptide analogues that adheres to the extracellular matrix and slowly releases the peptides, significantly prolonging stem cell survival in mouse models of ischaemic injury. The biomaterial can serve as a generic delivery system to improve functional outcomes in cell-replacement therapy.

Fig. 1. Preparation of peptide-linked collagen

a, A schematic depicting the method of col×D×pep synthesis. b, Quantification of amine groups on collagen before and after crosslinking using the TNBSA assay. The results are normalized to collagen concentration and expressed as the number of amines per 1,000 residues of collagen (mean ± s.d., n = 3). **P < 0.01. c, Tris-borate-EDTA-PAGE to detect free dendrimers. d, A click reaction scheme showing the fluorescent labelling of acetylene-labelled peptides with an azide probe. e, Quantification of peptides crosslinked to collagen by click chemistry. The results are normalized to collagen concentration and expressed as nanomoles of peptide per milligram of collagen (mean ± s.d., n = 3). *P < 0.05. f, SDS-PAGE to detect collagen and free peptides. g, Scanning electron microscope images of crosslinked collagens. Scale bars, 25 μ m. Distinct samples were measured and the experiments were performed in triplicate.

Fig. 2. Slow release of peptides from the col×D×pep pro-survival matrix in vitro and in vivo

a–c, In vitro peptide release of BMP2 (a), EPO (b) and FGF2 (c). Peptide-crosslinked collagen was deposited in a test tube and covered by a layer of 1× PBS buffer. The amount of released peptide was determined by fluorescent labelling using click chemistry. The admixture of free peptides with collagen or dendrimerized collagen was tested as controls. The results are normalized to total peptide input and shown as accumulated release. d, For in vivo peptide release, peptide-crosslinked collagen was injected into the hind limbs of SCID mice. The gastrocnemius muscle was then explanted at the indicated time points after delivery for assessment of released peptides by extraction and fluorescent labelling using the click chemistry approach. In a–c, the differences are significant between collagen + peptide and col×D× pep at all time points. Each data point is the mean ± s.d., n = 3, P < 0.05; in d, the difference is significant between crosslinked and uncrosslinked peptides; a blank (for example, saline injected) served as a control to measure the background signal. Each data point is mean ± s.d., n = 3. *P < 0.05; **P < 0.001. Distinct samples were measured and experiments were performed in triplicate.

Fig. 3. Evaluation of cell survival and limb perfusion after implanting BMMNCs with col×D×pep in SCID and immunocompetent mice after femoral artery ligation

a, Representative BLI of SCID mice (n = 5 per group) after hind-limb injection of BMMNCS with PBS + cells, peptide + cells or col×D×pep + cells treatment groups. b, Quantification of BLI. The results are mean ± s.d. (n = 5 for all groups). #P < 0.05 col×D×pep + cells versus PBS alone; *P < 0.05 col×D×pep + cells compared to PBS and peptide + cells. Distinct samples were measured and experiments were performed in duplicate. c, Representative laser Doppler images for the col×D× pep alone, PBS + cells, peptide + cells and col×D×pep + cells treatment groups. The arrows indicate the ischaemic limb on day 14 in different groups. d, Quantification of blood flow as determined by laser Doppler imaging. Results are mean±s.d. (n = 5 for all groups). **P < 0.01 compared to all groups.

Fig. 4. Evaluation of limb perfusion after implanting BMMNCs with col×D×pep in immunocompetent mice after femoral artery ligation

a, Representative laser Doppler images for PBS + cells, collagen + cells and col×D×pep + cells treatment groups. The arrows indicate the ischaemic limb on day 14 in different groups. b, Quantification of blood flow as determined by laser Doppler imaging. The results are mean ± s.d. (n = 5 for all groups). **P < 0.01 compared to PBS and col×D×pep groups.

Fig. 5. The col×D×pep pro-survival matrix promotes long-term cell survival in vivo

a, Representative bioluminescence images of SCID mice (n = 6 per group) after intra-myocardial injection of CPCs mixed with PBS, unmodified collagen, free peptides, admixture of collagen and free peptides, or col×D× pep cocktail as indicated. b, Quantification of BLI signals. The signals for each mouse are normalized to the value on day 1 post-injection. The results are presented as percentage of day 1 (mean ± s.d., n = 6, #P < 0.05 col×D×pep + cells versus all other groups except collagen +peptide + cells, *P<0.01 col×D×pep + cells versus all other groups). c, Representative bioluminescence images of immunocompetent mice (n = 6 per group) after intra-myocardial injection of CPCs mixed with PBS, unmodified collagen or col×D× pep cocktail as indicated. d, Quantification of BLI signals. The signals for each mouse are normalized to the value on day 1 post-injection. The results are presented as percentage of day 1 (mean ± s.d., n = 6); #P < 0.05 col×D×pep + cells versus all other groups except collagen + peptide + cells. *P < 0.05 col×D×pep + cells versus all other groups.

Fig. 6. Evaluation of graft function after implanting cardiac progenitor cells with col×D×pep in a SCID model of myocardial infarction

a, GFP signals (overlaid with a bright-field image) from representative hearts harvested from mice 30 days post-injection. Scale bars, 1 mm. b, Immunofluorescence staining of heart tissues for phosphorylated ERK1/2 and AKT. Four days post-injection for cells + peptide group; 14 days post-injection for cells + col×D× pep. The effect of treatment with col×D× pep was statistically different from those for other treatment groups at all time points. c, Representative M-mode echocardiographic data for infarcted hearts receiving 1 × 106 CPCs mixed with PBS, collagen alone or col×D× pep. Mice receiving PBS served as a control. d, Comparison of fractional shortening. The results are mean ± s.d. (n = 6 per group). **P<0.01, *P< 0.05 compared to all control groups. e, Comparison of the left ventricular end-diastolic dimension (LVEDD) and the left ventricular end-systolic dimension (LVESD). The results are mean ± s.d. (n = 6 per group). **P<0.01, *P< 0.05 compared to all control groups.

Fig. 7. Evaluation of the effects of CPC delivery with col×D×pep pro-survival matrix on post-infarct ventricular function by echocardiography and MRI

a, Representative M-mode echocardiographic data for infarcted hearts in immunocompetent FVB animals receiving 1 ×10⁶ CPCs mixed with PBS, collagen alone or col×D×pep. b, Comparison of fractional shortening. The results are mean±s.d. (n = 6 for all groups). **P<0.01, *P< 0.05 compared to all control groups. c, Representative short-axis MRI images show hearts receiving PBS alone, CPCs mixed with PBS, CPCs mixed with collagen or CPCs mixed with col×D× pep in immunocompromised mice. The hearts are shown at end-diastole and end-systole. Scale bar, 1 mm. d, Quantitative MRI assessments of the left ventricular ejection fraction of infarcted mice. The results are mean±s.d. (n = 6 for all groups). **P< 0.01 compared to other groups.

Fig. 8. Evaluation of left ventricular remodelling in immunodeficient mice by MRI after delivery of col×D×pep

a, Quantitative MRI assessments of the left ventricular end-diastolic volume (LVEDV) and the left ventricular end-systolic volume (LVESV) of infarcted mice. The results are mean ± s.d. (n = 6 for all groups). **P < 0.01, *P < 0.05 compared to other groups. b, Quantification of the amount of scar and viable tissue by histology. The data are mean ± s.d., n = 3. *P < 0.05. Distinct samples were measured and experiments were performed in triplicate. c, Representative haematoxylin and eosin (top) and Masson's trichrome (bottom) staining of left ventricular tissue of mice receiving CPCs mixed with PBS, collagen alone or col×D× pep. Blue on the Masson's trichrome tissue signifies scar tissue.