Transplantation of human embryonic stem cell-derived retinal pigment epithelial cells via injectable microfluidic-templated microgels for retinal regeneration

Abstract

Retinal pigment epithelial (RPE) cells are specialized neural cells crucial for vision, while human embryonic stem cell-derived retinal pigment epithelial (hESC-RPE) cells hold great potential as a viable cell source for treating retinal degenerative diseases like retinitis pigmentosa (RP). However, the transplantation efficiency and viability of hESC-RPE cell suspensions are relatively low due to detrimental shear-force during operations and host immune-clearance. We herein develop an alternative transplantation strategy with the aid of a microfluidic-templating microgel cell carrier to achieve substantially enhanced loading and delivery efficiency of hESC-RPE cells, thereby promoting visual function recovery after subretinal injection in the RP model Royal College of Surgeons (RCS) rats. Specifically, injectable monodispersed microgels consisting of gelatin-methacryloyl/Hyaluronic acid-methacryloyl core coated with fibrin shell (denoted as Fib@GHMS) were fabricated in a high-throughput and controllable manner, facilitating the adhesion and proliferation of hESC-RPE cells. RCS rats treated with microcarriers showed significantly improved visual function, evidenced by increased B-wave amplitudes and the preservation of the inner nuclear layer at 8 weeks post-surgery. In conclusion, our innovative delivery system Fib@GHMS for hESC-RPE cell transplantation presents a potential therapeutic graft for retinal tissue engineering. It may open a new avenue for clinical transplantation of minimally invasive cell-based treatments in retinal degenerative diseases.

Keywords: Cell transplantation; Cell-laden microgel; Injectable granular gel; Microfluidics; hESC-RPE.

Graphical abstract

Fig. 1

Scheme showing the transplantation strategy using minimally invasive injection of hESC-RPE cell-loaded Fib@GHMS for the treatment of retinal degeneration diseases. Monodisperse GelMA/HAMA microgels (GHMS) were manufactured via microfluidic droplet-templating method followed by photopolymerization to solidify hydrogel precursor. Thereafter, the microgels were further coated with a thin layer of fibrin via immerging the microgels in aqueous solution containing fibronectin, which were subsequently washed with thrombin to generate fibrin-coated GHMS (Fib@GHMS). hESC-RPE cells were cultured on the microgel surface in vitro to proliferate to a sufficient cell density, they were then injected into the subretinal space of RCS rats for transplantation.

Fig. 2

Microfluidic-templating fabrication of Fib@GHMS microgels. A: FTIR spectra of HAMA, GelMA and composite GH hydrogels, showing the reduced peak at 1556 cm⁻¹ corresponding to the GH double network hydrogel. B: Screening the minimum concentrations required of the GelMA and HAMA precursors to form a stable gel. C: Time sweeps of 2 % GelMA, 0.5 % HAMA, and G₂H_0.5 hydrogels to evaluate the mechanical performance. D: AutoCAD design of the microfluidic device. E: Microscopic images of the microchannel showing the preparation of uniformly sized aqueous droplets containing GelMA and HAMA precursors. H: Light microscopic images showing the generated monodisperse droplets. SEM images showing the morphology of GHMS (F, G) and Fib@GHMS microgels (I, J). K: The droplet size control via varying the flow ratio of Q_aqu/Q_oil. L: Statistical analysis of the microgel diameter range to analyze particle size uniformity. M: In vitro degradation of microgels.

Fig. 3

Formulating shear-thinning, self-healing granular gels based on Fib@GHMS. A: Oscillatory frequency sweep (1 % strain) tests of granular gels of Fib@GHMS. B: Strain sweep (1 Hz) tests of Fib@GHMS. C: Evolution of Fib@GHMS storage and loss modulus during one destructive shearing cycle (strain from 0.5 % to 500 %, 1 Hz), showing granular gel self-healing. D: Granular gel viscosity as a function of shear rate revealed shear-thinning behaviour. E: Images of Fib@GHMS injectability and moldability.

Fig. 4

The potential of hESC-RPE cell-Loaded Fib@GHMS microgels to restore retinal function. A: Cellular spreading and viability at different timepoints (4 h, 1 d, 3 d, 5 d) on the microcarriers. B: dsDNA cell proliferation assay of hESC-RPE cells cultured on the microcarriers. C: Cell cytoskeleton staining of hESC-RPE cells cultured on the microcarriers on day 5. D: Quantitative analysis of cell numbers on the microcarriers at different timepoints. E: Volcano plot depicting DEGs between the Fib@GHMS and TCP groups. F: Heatmap of upregulated genes and G: downregulated genes in the Fib@GHMS group compared with the TCP group in the retinal function recovery microenvironment (p < 0.05). KEGG pathway enrichment analysis of DEGs. H: Twenty significantly enriched GO terms for the differential genes. I: Twenty significantly enriched KEGG pathways for the differential genes.

Fig. 5

Specific functional expression of hESC-RPE cells cultured on Fib@GHMS microgels. A: qRT–PCR comparing the expression profiles of key genes between hESC-RPE cells on Fib@GHMS and in TCP cultures (n = 3). The results are presented as 'fold expression' values to highlight the enhanced expression levels of certain genes in Fib@GHMS cultures compared with adherent cultures (dashed line). Gene expression was normalized to the endogenous reference gene (GAPDH). Wilcoxon paired signed-rank tests were used to compare ΔCt values for each gene in Fib@GHMS and adherent cultures. B, C: The levels of secreted PEDF and VEGF in conditioned culture media. The error bars represent standard deviations. D: Immunostaining of hESC-RPE cells on the surface of Fib@GHMS using antibodies against MITF, RPE65, CRALBP, Laminin, Best1, ZO-1, and LRAT. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗∗p < 0.0001.

Fig. 6

Subretinal injection of hESC-RPE cell-loaded Fib@GHMS restores visual function. A: OCT evaluation of the biosafety and degradation status of hESC-RPE cell-loaded Fib@GHMS microcarrier in the subretinal space of RCS rats at 2, 4, and 8 weeks. Scale bar, 100 μm. GCL: ganglion cell layer; INL: inner nuclear layer; RPE: retinal pigment epithelium layer, including the transplanted RPE region. B: ERG assessment of the visual function recovery status of hESC-RPE cell-loaded Fib@GHMS in the subretinal space of RCS rats at 2, 4, and 8 weeks. C: Comparison of the relative decrease in retinal thickness (μm) in different groups at 8 weeks. Comparison of retinal a-wave(D) and b-wave(E) amplitudes (μV) for different groups at 2, 4, and 8 weeks. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗∗p < 0.0001.

Fig. 7

Retinal morphological analysis and tissue immunofluorescence staining after hESC-RPE cell-loaded Fib@GHMS transplantation in RCS rats. A: Representative hematoxylin and eosin-stained images of the whole eye and the transplanted area of rats from each group at 2, 4, and 8 weeks. Observation of degradation in the transplanted area and assessment of graft biocompatibility. The red dashed area indicates the subretinal transplantation bulge region. B: Comparison of transplantation area thickness (μm) for different groups at 2, 4, and 8 weeks. C: Close up image of HE stained histological sections where the injection was administered, revealing the healthy retina. D: The horizontal thickness distributions of the ONL near the optic disc in RCS rats of each group at 8 weeks, summarized from the results of serial sections of eyes. E: The retina thickness at 8 weeks. F: Representative immunofluorescence images at 4 and 8 weeks. Evaluation of the short-term survival of transplanted cells and visual function recovery in RCS rats after the subretinal transplantation of Fib@GHMS and cells. Selected areas (orange box) were marked to further observe the grafted hESC-RPE. Red: Transplanted cells labeled with Dil; Pink: RPE65 staining to identify RPE cells; Green: Cralbp staining to identify RPE cells overlying photoreceptor cells; DAPI: Nuclear staining. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)