Mesenchymal stromal/stem cells promote intestinal epithelium regeneration after chemotherapy-induced damage

Abstract

Background: Allogeneic hematopoietic stem cell transplantation (HSCT) is a curative treatment for leukemia and a range of non-malignant disorders. The success of the therapy is hampered by occurrence of acute graft-versus-host disease (aGvHD); an inflammatory response damaging recipient organs, with gut, liver, and skin being the most susceptible. Intestinal GvHD injury is often a life-threatening complication in patients unresponsive to steroid treatment. Allogeneic mesenchymal stromal/stem cell (MSC) infusions are a promising potential treatment for steroid-resistant aGvHD. Data from our institution and others demonstrate rescue of approximately 40-50% of aGvHD patients with MSCs in Phase I, II studies and minor side effects. Although promising, better understanding of MSC mode of action and patient response to MSC-based therapy is essential to improve this lifesaving treatment.

Methods: Single cell human small intestine organoids were embedded in Matrigel, grown for 5 days and treated with busulfan for 48 h. Organoids damaged by treatment with busulfan or control organoids were co-cultured with 5000, 10,000, and 50,000 MSCs for 24 h, 48 h or 7 days and the analyses such as surface area determination, proliferation and apoptosis assessment, RNA sequencing and proteomics were performed.

Results: Here, we developed a 3D co-culture model of human small intestinal organoids and MSCs, which allows to study the regenerative effects of MSCs on intestinal epithelium in a more physiologically relevant setting than existing in vitro systems. Using this model we mimicked chemotherapy-mediated damage of the intestinal epithelium. The treatment with busulfan, the chemotherapeutic commonly used as conditioning regiment before the HSCT, affected pathways regulating epithelial to mesenchymal transition, proliferation, and apoptosis in small intestinal organoids, as shown by transcriptomic and proteomic analysis. The co-culture of busulfan-treated intestinal organoids with MSCs reversed the effects of busulfan on the transcriptome and proteome of intestinal epithelium, which we also confirmed by functional evaluation of proliferation and apoptosis.

Conclusions: Collectively, we demonstrate that our in vitro co-culture system is a new valuable tool to facilitate the investigation of the molecular mechanisms behind the therapeutic effects of MSCs on damaged intestinal epithelium. This could benefit further optimization of the use of MSCs in HSCT patients.

Fig. 1

Co-culture with MSCs rescues the busulfan-induced damage of small intestine organoids. A Schematic overview of the in vitro co-culture model of MSCs and small intestine organoids damaged by treatment with busulfan. Single cell small intestine organoids were embedded in matrigel and grown for 5 days (day 0–5) and treated with busulfan (35 µM) for 48 h (day 5–7). Organoids damaged by treatment with busulfan or control organoids were co-cultured with 5000, 10,000, and 50,000 MSCs for 24 h or 48 h (day 7–8/9) and the surface area of organoids was measured. B Representative images of control organoids and organoids treated with busulfan co-cultured without or with 10,000 MSCs at 48 h after co-culture are shown. C Busulfan reduced the size of the organoids in organoid donor 1 and organoid donor 2. D Co-culture with MSCs increased the size of the organoids after treatment with 5000, 10,000, and 50,000 cells in organoid donor 1 and increased the size of the organoids after treatment with 50,000 MSCs in organoid donor 2. E Co-culture with 5 out of 9 tested bone marrow derived-MSC donors increased the size of organoids treated with busulfan. The quantification of surface area of the organoids was represented as fold change as compared to control. Results are shown as means ± SEM of data from 2 different organoid donors co-cultured with at least 3 MSC donors. Due to the large biological variation in organoid size, the statistical analysis of the effect of individual MSC donors on the size of busulfan-induced damaged organoids (E) was based on all evaluable individual organoids (of at least 3 organoid/matrigel droplets cultured in different wells). Scale bars, 1000 µm. * p < 0.05, ** p < 0.01, *** p < 0.001, and **** p < 0.0001 as compared to control (Kruskal–Wallis test or a Mann Whitney test)

Fig. 2

Co-culturing MSCs with single cell organoids damaged by busulfan improves intestinal organoid development. A. Schematic overview of the in vitro co-culture model of longer-term effects of busulfan and MSCs on small intestine organoids. Single cell small intestine organoids were embedded in matrigel and grown for 5 days (day 0–5) and treated with busulfan (35 µM) for 48 h (day 5–7). After treatment with busulfan organoids were dissociated to single cells and co-cultured with 5000 and 10,000 MSCs for 7 days (day 7–14) and the surface area and total number of organoids were measured. B. Representative images of control organoids and busulfan-treated organoids co-cultured without or with 10,000 MSCs at 7 days after co-culture are shown. C Busulfan reduced the size of the organoids in organoid donor 1 and organoid donor 2. D Co-culture with 10,000 MSCs increased the size of the organoids in organoid donor 1 and in organoid donor 2 . E Busulfan reduced the number of organoids in organoid donor 1. F Co-culturing these organoids with 5000 and 10,000 MSCs increased the number of organoids at 7 days after co-culture. The quantification of surface area of the organoids and the number of organoids was represented as fold change as compared to control. Results are shown as means ± SEM of data from 2 different organoid donors co-cultured with at least 3 MSC donors. Scale bars, 1000 µm. * p < 0.05 as compared to control (Kruskal–Wallis test or a Mann Whitney test)

Fig. 3

MSC secretome contributes to the rescue of busulfan-induced damage of small intestine organoids. A Schematic overview of the in vitro co-culture model of MSCs and small intestine organoids damaged by treatment with busulfan in a Transwell insert system. Single cell small intestine organoids were embedded in matrigel, grown for 5 days (day 0–5) in the Transwell insert, and treated with busulfan for 48 h (day 5–7). Organoids damaged by busulfan were co-cultured with 17,000 MSCs grown in the lower compartment of the Transwell system for 48 h (day 7–9) and the surface area of organoids were assessed at day 9. B Representative images of control organoids and organoids treated with busulfan co-cultured in Transwell system without or with 17,000 MSCs at 48 h after co-culture are shown. C Busulfan reduced the size of the donor 1 organoids. D Co-culture with MSCs in the Transwell system increased the size of the donor 1 organoids. The quantification of surface area of the organoids was represented as fold change as compared to control. Results are shown as means ± SEM of data from organoid donor 1. Due to the large biological variation in organoid size, the statistical analysis of the effect of individual MSC donors on the size of busulfan-induced damaged organoids was based on all evaluable individual organoids (1 organoid/matrigel droplet cultured in different wells in duplicate). Scale bars, 2000 µm. * p < 0.05, and **** p < 0.0001 as compared to control (Kruskal–Wallis test or a Mann Whitney test or one-way ANOVA)

Fig. 4

Transcriptomic analysis of busulfan-treated small intestine organoids. A Busulfan-treated organoids or control organoids were co-cultured with CellTraceTM violet dye-labelled MSCs for 48 h and single cell small intestine organoid were isolated from these co-cultures by FACS sorting. B DE analysis between control and busulfan-treated small intestine organoids (red dots indicate significant genes; p value < 0.05) and the accompanying upregulated (C) and downregulated (D) pathways. E Heat map of the differentially expressed genes involved in the EMT, proliferation, and apoptosis pathway in control organoids and busulfan-treated organoids is shown (N = 3)

Fig. 5

Proteomic analysis of busulfan-treated small intestine organoids. A DE analysis between control and busulfan-treated small intestine organoids (red dots indicate significant proteins; adjusted p value ≤ 0.1 and ≥ 1.5-fold change differences) and the accompanying upregulated (B) and downregulated (C) pathways are shown. D Heat map of the differentially expressed proteins involved in the EMT, proliferation, and apoptosis pathway in control organoids and busulfan-treated organoids is shown (N = 3)

Fig. 6

Transcriptomic analysis of busulfan-treated small intestine organoids co-cultured with and without MSCs. A DE analysis between busulfan-treated organoids co-cultured with 10,000 MSCs and without MSCs (red dots indicate significant genes; p value < 0.05) and the accompanying upregulated (B) and downregulated (C) pathways are shown. D Heat map of the differentially expressed genes involved in the EMT, proliferation, and apoptosis pathway in busulfan-treated organoids co-cultured with and without MSCs is shown (N = 3)

Fig. 7

Proteomic analysis of busulfan-treated small intestine organoids co-cultured with and without MSCs. A DE analysis between busulfan-treated organoids co-cultured with 10,000 MSCs and without MSCs (red dots indicate significant proteins; adjusted p value ≤ 0.1 and ≥ 1.5-fold change differences). B Venn diagram of the DE proteins in busulfan-treated organoids as compared to control (grey circle) and of DE proteins in busulfan-treated organoids co-cultured with and without MSCs (red circle) and the accompanying heat map analysis is shown

Fig. 8

MSCs promote regeneration of busulfan-induced damage in intestinal epithelium by regulating the proliferation and apoptosis pathways. A Apoptotic cell death analysis by flow cytometry, of busulfan-treated single cell small intestine organoids stained with Alexa Fluor (AF) 647-conjugated annexin V after co-culture with or without MSCs. B Effects of co-culturing busulfan-treated organoids with 0, 5000, 10,000, and 50,000 MSCs on the percentage of apoptotic single cell organoids are shown. C Proliferation analysis by flow cytometry, of busulfan-treated single cell small intestine organoids stained with EdU antibody after co-culture with or without MSCs. D Effects of co-culturing busulfan-treated organoids with 0, 5000, 10,000, and 50,000 MSCs on the percentage of proliferating single cell organoids are shown. Results are shown as means ± SEM of organoid donor 1 and organoid donor 2 co-cultured with at least 3 MSC donors. ** p < 0.01 and *** p < 0.001 as compared to control (Kruskal–Wallis test)