Fig 2: MSC secretome enhancement via AIMS.a Schematic of the secretome enhancement via AIMS. b Heatmap presenting the secretome profiles of different groups after incubating for 4 days. The data was processed by the Z-score normalization and clustering analysis. c–j ELISA assay measuring the secretion of FGF-2, IGF-1, VEGF, HGF-1, IL-6, IL-10, IFN-γ, and TNF- α from different groups after incubating for 4 days. k The amounts of secreted exosomes from different groups measured by the EXOCET Exosome Quantitation Kit. l AchE activity of different groups after incubating for 4 days. m Secreted exosome numbers of different groups analyzed using NTA. n A representative TEM image of secreted exosomes from MSC aggregates. n = 3 tests. Scale bar: 100 nm. o NTA test measuring the diameters of secreted exosomes from different groups. The statistical analysis was performed using one-way ANOVA with Tukey’s post-hoc test. Data are graphed as the mean ± SD (n = 4, biological repeats). Source data are provided as a Source Data file.

Fig 3: Harnessing the AIMS-enhanced MSC secretome for immunomodulation applications.a Heatmap measuring the inflammatory factors secreted from IFN-γ primed MSCs from different groups. b–i ELISA assay measuring the secretion of FGF-2, IGF-1, VEGF, HGF-1, IL-6, IL-10, IFN-γ, and TNF- α from different groups after incubating for 4 days. j Flow cytometry analysis measuring the percentage of CD86 and CD163 positive human THP-1 cells after incubating with the conditional medium from MSCs for 2 days. k Flow cytometry analysis measuring the percentage of CD4 positive T cells after incubating with the conditional medium from MSCs for 3 days. l Flow cytometry analysis measuring the B cell proliferation after incubating with the conditional medium from MSCs for 7 days. m, n ELISA assay measuring the secretion of TNF- α and IL-10 from B cells. The statistical analysis was performed using one-way ANOVA with Tukey’s post-hoc test. Data are graphed as the mean ± SD (n = 4 tests, biological repeats). Source data are provided as a Source Data file.

Fig 4: The enhanced MSC secretome was attributed to increased cell-cell interactions mediated by N-cadherin.a Schematic illustrating how N-cadherin mediated cell-cell interactions contribute to the enhancement of the MSC secretome. b Heatmap measuring the secretome profiles of different groups after functional blocking of N-cadherin of MSCs and incubating for 4 days. The data was processed by the Z-score normalization and clustering analysis. c–j ELISA assay measuring the secretion of FGF-2, IGF-1, VEGF, HGF-1, IL-6, IL-10, IFN-γ, and TNF-α from different groups after functional blocking of N-cadherin of MSCs and incubating for 4 days (n = 4 tests, biological repeats). k Representative FDA staining images of different groups after incubating for 0, 12, 24 h, and 3 days, respectively. n = 3 tests with similar results. Scale bar: 100 μm. l Western blotting detecting the expression of N-cadherin of MSCs from different groups. n = 3 tests. a: control, b: 3D Microplate, c: 10 min, d: 30 min, e: 60 min. m Flow cytometry analysis measuring the expression of N-cadherin of MSCs from different groups. The statistical analysis was performed using one-way ANOVA with Tukey’s post-hoc test. Data are graphed as the mean ± SD. Source data are provided as a Source Data file.

Fig 5: Biomechanical properties and histological assessment of the rabbit humeral bone-SST tissue samples 8-weeks post-surgery. 3D-printed PHT scaffolds containing FGF-2 and TGF-β3 (PHT-GFs) facilitated at least 1-cm of tendon regeneration and exhibited native supraspinatus tendon (SST)-like mechanical properties. (a) Experimental design for in vivo characterization of 3D-printed PHT scaffold in rabbit SST injury model. (b) Rabbit SST surgery procedures. (c) Representative force-displacement curve. (d) Ultimate load was increased in PHT-GFs group compared to the intact control group. (e) Stiffness was similar between PHT and PHT-GFs compared to the intact control group. (f) Ultimate displacement was increased in PHT and PHT-GFs compared to the intact control group. (g) Total work-to-failure was increased in PHT and PHT-GFs compared to the intact control group. (h) Ashby chart simultaneously comparing ultimate load and stiffness of repaired tendon constructs normalized to their uninjured tendon control showed that this work (PHT and PHT-GFs) attained the highest normalized ultimate load and stiffness. Panels (d), (e), (f), and (g) – Control: n = 16, PHT: n = 8, PHT-GFs: n = 8. *: Statistical significance (p ≤ 0.05), **: p ≤ 0.01; ***: p ≤ 0.001 relative to intact control group. (i) Representative H&E, Masson's Trichrome and Picrosirius Red (brightfield and under polarized light microscopy) staining of PHT and PHT-GFs samples. Longitudinal sections showing humeral bone and SST were used. Black, orange and blue rectangles represent the region-of-interest (ROI) used for semiquantification (left). Dotted lines demarcate regions of 3D-printed PHT scaffold and are labeled with ‘PHT’. Scale bar = 5000 μm. Representative high magnification images of H&E, Masson's Trichrome and Picrosirius red staining (polarized light images) showing the regenerated tendon fibers in PHT and PHT-GFs group (right). Collagen expression was semi-quantified by the signal intensity in the ROI of Picrosirius Red-stained brightfield images. (j) Macroscopic histopathological score. Radar charts showed that PHT-GFs attained macroscopic histopathological scores similar to normal SST relative to PHT group. (k) Representative images of immunofluorescence staining for tendon-associated markers Scx, Col1 and Tnmd with no primary antibody control. Orange dotted lines indicate the border between PHT scaffold and adjacent tissues. Red lines indicate the enlarged region. Semiquantification of tendon markers expression. Scale bars as indicated in the images. Panels (i) and (j) – Sample number n = 6 for each group. *: Statistical significance (p ≤ 0.05); **: p ≤ 0.01.