Mesenchymal stromal cell-derived membrane particles suppress kidney fibrosis

Abstract

Background: Kidney injury, typically accompanied by inflammation, is a driver for kidney fibrosis, which contributes to the development of kidney failure. Mesenchymal stromal cells (MSC) have been proposed to have anti-fibrotic potential, but challenges such as their short persistence after infusion and inability to cross the lung capillary system due to their large size hamper their use for treatment of kidney fibrosis. It is hypothesized that the effects of MSC are partially dependent on phagocytosis of fragments of MSC by target cells and inhibiting excessive immune activation response. To exploit this effect of MSC, we developed nanosized membrane particles (MP) from MSC and explored their anti-fibrotic activity and immunomodulation effect in mouse and human kidney fibrosis models.

Methods: MP were generated from culture-expanded MSC through extrusion of isolated membranes. Unilateral kidney ischemia reperfusion injury (IRI) in male Balb/c mice was used to induce kidney fibrosis. MP generated from 1 × 10⁶ MSC were injected in the tail vein immediately after anesthesia recovery. In a second model, human induced pluripotent stem cell-derived kidney organoids were exposed to 1% O₂ for 48 h and 100 ng/mL IL-1β for 96 h to mimic IRI in vitro for inducing fibrosis. MP generated from 0.5 × 10⁶ MSC were added to the medium for 4 consecutive days. Fibrosis and immune cell markers were subsequently measured.

Results: IRI induced the expression of transforming growth factor beta (TGFβ) and collagen type I alpha 1(COL1A1) in mouse kidneys. MP treatment significantly reduced TGF-β mRNA at day 3 while COL1A1 mRNA and protein were downregulated at day 7. We found no evidence for an immunomodulatory effect of MP, as the number and activity of infiltrating T cells and macrophages did not change. In kidney organoids, a rise in COL1A1 and TGF-β demonstrated successful fibrosis induction by hypoxia and IL-1β. MP significantly decreased these fibrosis markers. Additionally, immunohistochemistry revealed a reduction in the myofibroblast marker alpha smooth muscle actin.

Conclusions: Our results demonstrate that MP have anti-fibrotic properties in mouse kidney IRI and human kidney organoid models. These results indicate that MP have potential for the development of kidney fibrosis-inhibiting therapy.

Keywords: Human kidney organoid; Immunomodulation; Ischemia reperfusion injury; Kidney fibrosis; Membrane particle; Mesenchymal stromal cell; Regenerative medicine.

Fig. 1

Generation and characterization of MP from AT-MSC. A A schematic overview of the generation of MP. B Cryo-electron microscopy image of MP. MP (arrows) display a spherical shape and a visible lipid bilayer. C Representative profile of size distribution and concentration of MP by nanoparticle tracking analysis

Fig. 2

MP do not affect monocyte, macrophage and T cell numbers in IRI kidneys and blood. A A schematic overview of the establishment of mouse IRI model and MP administration. B–C Gene expression analysis of macrophage maker F4/80 and T cell marker CD3 in 3 days MP and 7 days MP treated experiments. Data are represented as mean ± SEM. Each symbol represents a data point from 1 kidney sample. N = 10 for sham, 12 for IRI and 9 for IRI + MP in 3 days experiment; n = 10 for sham, 8 for IRI and 10 for IRI + MP in 7 days experiment. D Percentage of classical and non-classical monocytes in blood analyzed by flow cytometry in 3 days MP treated experiments. Data are represented as mean ± SEM. Each symbol represents a data point from 1 blood sample. N = 6 for sham, 10 for IRI and 9 for IRI + MP. E Percentage of Treg cells in blood in different time points. Data are represented as mean ± SEM. Each symbol represents a data point from 1 blood sample. N = 7 for sham, 13 for IRI and 10 for IRI + MP in 3 days experiment; n = 3 for sham, 5 for IRI and 6 for IRI + MP in 7 days experiment. Treg cell, regulatory T cell. ns means non-significant. SEM, standard error of the mean. Asterisks reflect statistically significant differences (P < 0.05); ns means non-significant. SEM, standard error of the mean

Fig. 3

MP inhibit fibrosis in a mouse IRI model. A, B Gene expression analysis of TGF-β and COL1A1 3 days and 7 days after MP treatments. Data are represented as mean ± SEM. Each symbol represents a data point from 1 kidney sample. N = 8 for sham, 10 for IRI and 9 for IRI + MP in 3 days experiment; n = 10 for sham, 8 for IRI and 10 for IRI + MP in 7 days experiment. C, D Representative slices stained with COL1A1 and α-SMA in kidneys from 7 days after MP treatment and their quantitative analysis analyzed by Qupath. Zoom-in detailed views of dashed circle area are shown below the images. Scale bars: 2.5 mm; detailed view: 100 μm. Data from quantitative analysis are represented as mean ± SEM. Each symbol represents the analysis of 1 image from 1 kidney. N = 10 for sham, 8 for IRI and 10 for IRI + MP. Asterisks reflect statistically significant differences (P < 0.05); ns means non-significant. SEM, standard error of the mean; TGF-β, transforming growth factor beta; COL1A1, collagen type I alpha 1; α-SMA, alpha smooth muscle actin

Fig. 4

Generation and characterization of iPSC-derived kidney organoids. A A schematic overview of kidney organoid differentiation procedure. B Representative microscope images of different time points in kidney organoid culture. Scale bars: 750 μm. C Immunohistochemistry images for markers of glomerular cells (PODXL), proximal tubular cells (Villin1), distal tubular cells (ECAD), interstitial cells (PDGFRα) and endothelial cells (CD31). Zoom-in detailed views of dashed circle area are shown below the images. Scale bars: 500 μm; detailed view: 50 μm

Fig. 5

MP inhibit fibrosis in kidney organoid model. A, B Gene expression analysis of TGF-β and COL1A1 in the control, fibrotic (1% O₂ + IL-1β) and fibrotic + MP treatment kidney organoids at day 4. Data are represented as mean ± SEM. Each symbol represents a data point from 1 kidney organoid. N = 26 for control, 26 for 1% O₂ + IL-1β and 18 for 1% O₂ + IL-1β + MP. C, D Immunohistochemistry images and quantitative analysis analyzed by Qupath from COL1A1 and α-SMA in the above groups. Zoom-in detailed views of dashed circle area are shown below the images. Scale bars: 500 μm; detailed view: 50 μm. Data from quantitative analysis are represented as mean ± SEM. Each symbol represents the analysis of 1 image. α-SMA: n = 20 for control, 28 for 1% O₂ + IL-1β and 23 for 1% O₂ + IL-1β + MP. COL1A1: N = 18 for control, 29 for 1% O₂ + IL-1β and 21 for 1% O₂ + IL-1β + MP. Asterisks reflect statistically significant differences (P < 0.05). SEM, standard error of the mean. TGF-β, transforming growth factor beta; COL1A1, collagen type I alpha 1; α-SMA, alpha smooth muscle actin. E Immunofluorescence analysis of PDGFRα and α-SMA expression in above groups. Zoom-in detailed views of square area are shown below the images. Scale bars: 200 μm; detailed view: 50 μm