Injectable Therapeutic Organoids Using Sacrificial Hydrogels

Abstract

Organoids are becoming widespread in drug-screening technologies but have been used sparingly for cell therapy as current approaches for producing self-organized cell clusters lack scalability or reproducibility in size and cellular organization. We introduce a method of using hydrogels as sacrificial scaffolds, which allow cells to form self-organized clusters followed by gentle release, resulting in highly reproducible multicellular structures on a large scale. We demonstrated this strategy for endothelial cells and mesenchymal stem cells to self-organize into blood-vessel units, which were injected into mice, and rapidly formed perfusing vasculature. Moreover, in a mouse model of peripheral artery disease, intramuscular injections of blood-vessel units resulted in rapid restoration of vascular perfusion within seven days. As cell therapy transforms into a new class of therapeutic modality, this simple method-by making use of the dynamic nature of hydrogels-could offer high yields of self-organized multicellular aggregates with reproducible sizes and cellular architectures.

Keywords: Bioelectronics; Biomaterials; Biotechnology.

Graphical abstract

Figure 1

Schematic Diagram of Method of Using Sacrificial Hydrogels to Produce Therapeutic Organoids (A) Schematic demonstrating the parallels between the surface micromachining method to fabricate MEMS devices such as a microcantilever (top) and the use of sacrificial alginate microwells to fabricate organoids (bottom). Both methods involve the use of a sacrificial layer (blue) to fabricate the final structure (red). (B) Time required to completely uncrosslink alginate microwells following incubation with different concentrations of a chelator (sodium citrate) by measuring the percent change in mass over time (n = 3, data are represented as mean ± standard deviation). (C) Schematic diagrams (top) and corresponding experimental images (bottom) showing the steps of organoid fabrication and in vivo perfusion. Experimental data were collected using GFP-labeled HUVECs and RFP-labeled mouse MSCs. First, a co-culture of endothelial cells (green) and mesenchymal stem cells (red) is seeded on dissolvable alginate microwells. Second, after being cultured in maintenance medium without growth factors for 3 to 4 days, cells self-organize into organoids with an endothelial core. A switch into culture medium with vasculogenic growth factors for an additional four days promoted formation of vessels within the organoids. Third, alginate microwells were dissolved with 5% sodium citrate to release organoids. Fourth, suspension of organoids could be centrifuged and assembled into a macro-tissue in vitro to study vascular formation or injected into the subdermis or ischemic hindlimb of a mouse to demonstrate engraftment in vivo. Fifth, injected organoids rapidly connected to form perfused microvasculature in vivo. Scale bar: 100 μm. (D) The liquid handling steps in the process: (1) seeding the co-culture of ECs (green) and MSCs (red) by pipetting cells onto alginate microwell construct, (2) adding maintenance media once the cells have settled to the bottom of the microwells (approx. 30 minutes), (3) switching to vasculogenic media once an endothelial core has formed, and (4) gently dissolving the alginate microwells (approx. 12 minutes) to harvest organoids (the organoids can be gently washed prior to injection).

Figure 2

Production of Organoids at Large Scale and Functionality of Organoids to Form Macrotissue (A) Pictures of three alginate microwells constructs for inserts into 24-well plates, 12-well plates, or 60-mm dishes with the capacity to produce 24 ×1,000; 12 ×3,000; or 30,000 organoids respectively. (B) Picture of 250 million cells for seeding into alginate microwells. Cells in this figure are GFP-labeled HUVECs and RFP-labeled mouse MSCs. (C) Stitched brightfield image of cells seeded in a 60-mm construct with 30,000 wells to create 30,000 organoids. Scale bar: 1 cm. (D) Picture of a 1-mm-thick macrotissues with an area of 1 cm² assembled in vitro by collecting the 30,000 mature prevascularized organoids produced with the alginate microwell (A and B) construct in a 60-mm dish. Scale bar: 1 cm. (E) Fluorescence images of the macrotissue in (D) with a close-up of the closely packed organoids with endothelial cores (green). Scale bars: 1 mm (left) and 500 μm (right).

Figure 3

Production of Vascularized Organoids with High Reproducibility in Size and Structure (A) Confocal fluorescence images of co-culture organoids of GFP-labeled HUVECs (green) and RFP-labeled mouse MSCs (red) over the first three days in maintenance medium without growth factors (top) or in vasculogenic medium with 40 ng/mL VEGF and 40 ng/mL bFGF (bottom). The cells self-organize by migration and either formed endothelial cores when cultured in media without growth factors (top) or had endothelial cells randomly distributed near the surface of the organoid and did not form endothelial cores when cultured in media with growth factors (bottom). (B) Overlay of fluorescent and transmitted images showing parallel production of organoids in arrays of different sizes of microwells (with either 100, 200, or 400 μm diameter) and different co-culture ratios (1 EC: 3 MSC, 1 EC: 1 MSC, or 3 EC: 1 MSC). Different sizes of microwells yield different sizes of organoids, either unvascularized with only MSCs or prevascularized with a co-culture of ECs and MSCs, and different co-culture ratios yield different endothelial core sizes. Scale bar: 100 μm. (C) Quantitative analysis of cell aggregation into organoids and the formation of an endothelial core over time in 200 μm microwells, as measured by the radius of the smallest circle that can contain all MSCs (red) or all ECs (green) (n> 20). Data are represented as mean ± standard deviation. (D) Barplot showing the size of fully contracted organoids (red) and the size of the endothelial cores (green) for all tested microwell sizes and co-culture ratios. Data are represented as mean ± standard deviation. (E) Reproducibility of endothelial cores; the number of organoids produced in 1 mm² (dark gray) and the number of organoids containing and endothelial core (light gray) for all tested microwell sizes and co-culture ratios. See also Figures S2–S6.

Figure 4

Production of Vascularized Organoids with Human Cells (A) Maturation of endothelial cores with dynamic culture conditions for two co-culture ratios; 1 GFP-HUVEC: 3 hAMSC (left) and 1 GFP-HUVEC: 1 hAMSC (right). The cells are seeded (day 0) and initially cultured in maintenance medium without growth factors to form endothelial cores. After three days the organoids were cultured in vasculogenic medium with 40 ng/mL VEGF and 40 ng/mL bFGF and the endothelial cores matured into vessels with discernable lumens (red arrows) and sprouts (white arrows). Scale bar: 200 μm. (B) Epifluorescence, brightfield, and overlay images showing early self-organization of prevascularized organoids over the first 20 h, with a 1 GFP-HUVEC: 1 hAMSC co-culture in 400 μm microwells. Scale bar: 200 μm. (C) Epifluorescence, brightfield, and overlay images showing fusion of prevascularized organoids (same conditions as in right A and B) into mesotissues over the first 24 h of the fusion process within a 400-μm collagen-doped alginate microwell. Scale bar: 200 μm. See also Figures S7 and S8.

Figure 5

Rapid In Vivo Vascularization in Healthy Mice upon Injection of Organoids, as Observed in Real Time via a Window Chamber (A) Schematic diagram of experimental setup for observing vascular formation and integration with host vasculature in vivo in real time via a window chamber. Organoids (from human cells formed under dynamic culture conditions in 200-μm microwells yielding organoids 71 ± 5 μm in diameter) were injected into a window chamber implant in an SCID mouse. (B) Real-time in vivo stereoscopic images of prevascularized microtissues with 1 GFP-HUVEC: 1 hAMSC (top row) and unvascularized organoids with hAMSC only (bottom row) through window chamber at different time points. In the top row, newly formed vessels are apparent within 4 days, and blood-filled vessels observed by day 7. In the bottom row, the dashed white line indicates the area of organoids implant and no neo-vascularization was observed. Scale bar: 500μm. (C) Quantification of neo-vascularization of the prevascularized organoids as the total length of vasculature within three ROIs of 800-by-800 μm containing up to 60 blood-filled vessel branches. The total length of vasculature increases substantially after day 7 for prevascularized organoids. There is no substantial difference in total length of the vasculature for the unvascularized organoids. Note that the blood-filled vessel intersections have been interpreted as branches and not as overpasses (which are two separate vessels, one over the other); however, further work is required to verify this. Data are represented as mean ± standard deviation. (D) Distributions of branching length in the newly formed microvasculature (B and C) at day 7, 9, and 11. Lines above histogram indicate the mean branch length and standard deviation for day 7, day 9, and day 11 as 93 ± 39 μm, 86 ± 29 μm, and 93 ± 44 μm, respectively. (E) Real-time in vivo images of prevascularized organoids with endothelial cells in green. The confining pressure of the intact fascia likely causes the spheroids to be organized in a two-dimensional grouping and was imaged at the plane of the endothelial cores. Red arrow heads point to luminous, blood-filled vessels (as indicated by dark lines in fluorescence images and dark areas of brightfield images). Scale bar: 250 μm. See also Figures S9–S12.

Figure 6

Rapid In Vivo Restoration of Perfusion and Muscle Fiber Regeneration upon Injection of Organoids in Ischemic Hindlimb (A) Representative images of blood perfusion in the hindlimbs measured with laser speckle contrast imaging (LSCI) for the two experimental groups. The superficial femoral artery was isolated from the femoral vein and nerve bundle along the length of the thigh, ligated and excised. The mouse then received four injections along the length of the thigh of either 25 μL PBS (control) or organoids (corresponding to 0.5 × 10⁶ cells) injected at each site. The organoids were formed in 200-μm microwells with maintenance media and a 1 mEC: 1 mMSC co-culture ratio yielding vascularized organoids 71 ± 5 μm in diameter with endothelial cores. The perfusion of each limb was measured as the average LSCI intensity of the planar surface of the paw (dashed white outline). (B) Quantification of perfusion in the hindlimbs as the perfusion ratio (R/L) between the naive left (L) hindlimb and the ischemic right (R) hindlimb (n=3 mice for each condition) with the control mice in blue and the organoid treatment mice in green. Data are represented as mean ± standard deviation. The best-fit (dashed) and 95% confidence interval (dotted) lines are shown (from day 0–9 for organoids, from day 0–14 for control). (C) Histology of the gastrocnemius muscle on day 14 with an H&E stain. White arrows indicate centralized nuclei of regenerating muscle fibers. Scale bar: 50 μm. (D) Percentage of myofibers characterized as necrotic because of hyalinization (pink) or as regenerating, viable myofibers assessed by centralized nuclei (purple) (n=3).(4 ROIs of 587 × 440 μm² for each condition). Data are represented as mean ± standard deviation; ∗ indicates significantly more regenerating fibers with p < 0.05. See also Figures S10 and S13.