Intraperitoneal but not intravenous cryopreserved mesenchymal stromal cells home to the inflamed colon and ameliorate experimental colitis

Abstract

Background and aims: Mesenchymal stromal cells (MSCs) were shown to have immunomodulatory activity and have been applied for treating immune-mediated disorders. We compared the homing and therapeutic action of cryopreserved subcutaneous adipose tissue (AT-MSCs) and bone marrow-derived mesenchymal stromal cells (BM-MSCs) in rats with trinitrobenzene sulfonic acid (TNBS)-induced colitis.

Methods: After colonoscopic detection of inflammation AT-MSCs or BM-MSCs were injected intraperitoneally. Colonoscopic and histologic scores were obtained. Density of collagen fibres and apoptotic rates were evaluated. Cytokine levels were measured in supernatants of colon explants. For cell migration studies MSCs and skin fibroblasts were labelled with Tc-99m or CM-DiI and injected intraperitonealy or intravenously.

Results: Intraperitoneal injection of AT-MSCs or BM-MSCs reduced the endoscopic and histopathologic severity of colitis, the collagen deposition, and the epithelial apoptosis. Levels of TNF-α and interleukin-1β decreased, while VEGF and TGF-β did not change following cell-therapy. Scintigraphy showed that MSCs migrated towards the inflamed colon and the uptake increased from 0.5 to 24 h. Tc-99m-MSCs injected intravenously distributed into various organs, but not the colon. Cm-DiI-positive MSCs were detected throughout the colon wall 72 h after inoculation, predominantly in the submucosa and muscular layer of inflamed areas.

Conclusions: Intraperitoneally injected cryopreserved MSCs home to and engraft into the inflamed colon and ameliorate TNBS-colitis.

Figure 1. Biological characterization of mesenchymal stromal cells.

Phenotypic analysis of MSCs was carried out by flow cytometry, which revealed that BM- and AT-MSCs expressed the cell markers CD90 (Thy-1) and CD29, but did not express lineage markers such as CD45, CD11b and CD34 (A–F). Solid black lines show AT-MSC, dotted black lines show BM-MSC, and gray lines are isotype control. Functionally, MSCs have the capacity to form different cell lineages. AT- (G–I) and BM- (J–K) MSCs were able to differentiate into adipocytes (H, K) and osteocytes (I, L). These cells were used in subsequent experiments. Oil Red O (G–H, J–K) and Von Kossa (I, L) staining. Length bars represent 50 µm.

Figure 2. Effect of MSCs on the clinical and colonoscopic parameters of the colitis model.

Survival did not differ significantly among experimental groups. Of the TNBS-colitic animals, 2 treated with vehicle died on days 4 and 7, and 1 treated with BM-MSCs died on day 9. Survival was analyzed by the Kaplan–Meier log-rank test (A). Following TNBS-induction, animals presented a progressive weight loss compared with controls (*p<0.001). After receiving AT- or BM-MSCs, animals gradually regained weight, in contrast to vehicle-treated animals (**p<0.046; ***p<0.004) (B). Colonoscopic imaging was obtained after colitis induction at day 4, and after MSCs administration at day 11. In control experiments, colonoscopy was performed following intra-rectal saline enemas, and intraperitoneal administration of vehicle in TNBS-induced animals (C). Differences before and after treatment were evaluated with the Wilcoxon matched-pair signed rank test. Values are mean±S.E.M. of 10 animals/group.

Figure 3. Effect of MSCs on histological parameters and collagen deposition in the colon.

Paraffin sections were stained with HE (A), and phosphomolibidic acid-picrosirius red dye (C) (original magnification ×100). AT- or BM-MSC were administered intraperitoneally at day 4. In control experiments, histological evaluation was performed following intraperitoneal administration of vehicle, and intra-rectal saline enemas. Colonic samples were scored according to histological parameters described in materials and methods. Density of collagen fibres was determined by a computerized image analysis system. Horizontal bars represent medians, boxes represent the 25th and 75th percentiles, and vertical bars represent ranges, of 10 animals/group. Differences were analyzed using ANOVA on ranks with Dunnett's test (B, D).

Figure 4. Effect of MSCs on apoptotic rates within the colon.

Apoptotic cells were detected by the TUNEL assay. Photomicrographs of the colon show representative samples of a control, and from TNBS-induced colitic animals treated with BM- and AT-MSCs, or vehicle, and an internal negative control without TdT enzyme (original magnification ×100) (A). Percentages of apoptotic cells in the colonic epithelium and in the lamina propria were analyzed in at least 10 different areas per tissue section. Epithelial values of control and MSCs-treated animals were lower compared to vehicle-treated colitis (p<0.008). Lamina propria values of MSCs- or vehicle-treated animals were higher compared with the control group (p<0.04). Horizontal bars represent medians, boxes represent the 25th and 75th percentiles, and vertical bars represent ranges of 10 animals/group. Differences were analyzed using ANOVA on ranks with Dunnett's test (B).

Figure 5. Effect of MSCs on cytokine production in the inflamed colon.

Colonic explants were cultured for 24 h at 37°C. After centrifugation, supernatants were used for measurement of the concentration of cytokines by an ELISA method for rat TNF-α (A), IL-1β (B), VEGF (C), TGF-β (D), and IL-10 (E), as described in Supplementary Materials. AT-, BM-MSC, and vehicle were administered to colitic animals, while control group did not receive any treatment. Values are expressed as picogram of cytokine/mL of culture supernatant, and normalized by protein contents. Horizontal bars represent medians, boxes represent the 25th and 75th percentiles, and vertical bars represent ranges of 10 animals/group. Differences were evaluated with the one-way ANOVA on ranks with Dunnett's test.

Figure 6. Migration of intraperitoneally and intravenously administered MSCs in experimental animals.

Tec-99m-labeled MSCs were administered to animals and followed for 24 hours using a gamma-camera. Tec-99m-labeled MSCs were injected intraperitoneally in the right lower abdominal quadrant, and migrated towards the inflamed colon on the contra-lateral left lower quadrant, but not to the non-inflamed colon of controls. Skin fibroblasts used as additional controls did not migrate towards the inflamed colon (A). Tec-99m-labeled MSCs injected through the jugular vein accumulated first into the lungs and then gradually migrated towards liver, spleen, kidneys, and bladder. After 24 hours, labelled cells were barely detectable (B). Images are representative of 3 independent experiments. All animals are lying on their backs. Right (R); left (L).

Figure 7. Distribution and engraftment of intraperitoneally administered MSCs in the colon.

Cm-DiI-labelled MSCs (red) injected intraperitoneally in the right lower abdominal quadrant were tracked after 72 hours. After surgical removal of the distal colon, samples were cut in a cryostat, counterstained with DAPI (blue), and analyzed under confocal microscopy. Labelled cells migrated towards the inflamed colon on the contra-lateral left lower quadrant, but not to the non-inflamed colon of normal controls. Cm-DiI-labeled MSCs were mainly observed in the muscular layer and the submucosa (SM), but also in the lamina propria of the mucosa (M), predominantly in the bottom of crypts. AT-MSCs distributed throughout the colon wall, from the peritoneum to the epithelial layer, whereas BM-MSCs concentrated at the peritoneal surface, muscular layer and the submucosa. No labelled MSCs were observed in control animals. Intestinal lumen (L). Images are representative of 3 independent experiments. Length bars represent 100 µm.