Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2016 Jan 14;11(1):e0146970.
doi: 10.1371/journal.pone.0146970. eCollection 2016.

Fibroblast Cell-Based Therapy for Experimental Autoimmune Diabetes

Free PMC article

Fibroblast Cell-Based Therapy for Experimental Autoimmune Diabetes

Reza B Jalili et al. PLoS One. .
Free PMC article


Type 1 diabetes (T1D) results from autoimmune destruction of insulin producing β cells of the pancreatic islets. Curbing autoimmunity at the initiation of T1D can result in recovery of residual β cells and consequently remission of diabetes. Here we report a cell-based therapy for autoimmune diabetes in non-obese diabetic (NOD) mice using dermal fibroblasts. This was achieved by a single injection of fibroblasts, expressing the immunoregulatory molecule indoleamine 2,3 dioxygenase (IDO), into peritoneal cavity of NOD mice shortly after the onset of overt hyperglycemia. Mice were then monitored for reversal of hyperglycemia and changes in inflammatory/regulatory T cell profiles. Blood glucose levels dropped into the normal range in 82% of NOD mice after receiving IDO-expressing fibroblasts while all control mice remained diabetic. We found significantly reduced islet inflammation, increased regulatory T cells, and decreased T helper 17 cells and β cell specific autoreactive CD8+ T cells following IDO cell therapy. We further showed that some of intraperitoneal injected fibroblasts migrated to local lymph nodes and expressed co-inhibitory molecules. These findings suggest that IDO fibroblasts therapy can reinstate self-tolerance and alleviate β cell autoreactivity in NOD mice, resulting in remission of autoimmune diabetes.

Conflict of interest statement

Competing Interests: The authors have declared that no competing interests exist.


Fig 1
Fig 1. Reversal of hyperglycemia in NOD mice after cell therapy.
IDO-expressing or control fibroblasts were injected intraperitoneally to NOD mice with overt hyperglycemia in the first two weeks following the onset of diabetes. These mice did not receive any other treatment. Blood glucose levels were checked twice a week. Panels A show blood glucose levels in IDO treated mice (n = 11 mice). Time point 0 denotes fibroblasts injection time. All control animals (B) that received no cells, control fibroblasts (Con. Fib.) or IDO expressing fibroblasts and an IDO inhibitor, 1-methyl tryptophan (IDO Fib. + 1MT) remained hyperglycemic (n = 5 mice/ group). Kaplan-Meier survival curve with log-rank analysis (C) showed significant decrease in diabetes rate of IDO fibroblast treated group. Intraperitoneal glucose tolerance tests (D) showed improvement of glycemic control in the IDO cell therapy group (red squares) compared to the control mice (blue diamonds).
Fig 2
Fig 2. Histology of pancreas following cell therapy.
Histology of representative pancreas sections of NOD mice before (at the commencement of autoimmune diabetes) and after cell therapy is shown. Panel A shows hematoxylin-eosin staining (left and center) and insulin immuno-staining (right) of pancreas sections of a NOD mouse euthanized in the first week after the onset of hyperglycemia. Panels B and C show pancreatic sections of an untreated control NOD mouse with chronic diabetes (6 weeks post-diabetes), an IDO cell treated NOD mouse (18 weeks post-treatment), respectively. Middle and right columns show higher magnification (×400) of the marked areas of the left column (×100). Panel D shows Insulitis scores calculated as described in Methods. The ratio (%) of islets with insulitis significantly decreased after IDO cell therapy animals compared to those of the early onset diabetic mice and controls (E). (*) denotes significant decrease in islet insulitis in IDO groups compared to the controls and diabetes onset (P<0.001, n = 5).
Fig 3
Fig 3. Frequency of regulatory, autoreactive CD8+, and Th17 T cells in NOD mice after cell therapy.
NOD mice received IDO-expressing fibroblasts upon initiation of spontaneous diabetes. At the endpoint of the study, spleens and pancreatic lymph nodes of these mice were harvested and frequencies of various subtypes of T cells were measured using flow cytometry. Panels A & B show the frequencies of CD 25+ FoxP3+ regulatory T cells (gated on CD4+ T cells) in spleens (SPL) and pancreatic lymph nodes (PLN) of mice, respectively. Autoreactive T cells were detected using NRP-V7 high-avidity peptide/MHC class I tetramers as described in Methods. Panels C &D show representative plots for CD8+ tetramer positive cells in spleen (SPL) and pancreatic lymph nodes (PLN) using a lymphocyte gate and excluding CD4+ and B220+ cells. The numbers in upper right corner of plots indicate the percentage of CD8+ B220- CD4- tetramer+ cells. Panels F & G show representative plots for IL-17 and CD4 flow cytometry analysis of lymphocytes in spleens (SPL) and pancreatic lymph nodes (PLN). The numbers in upper right corner of plots indicate the percentage of CD4+ IL-17+ (Th17) cells. Panels E & H show quantification of the frequencies of autorective CD8+ T cells and Th17 cells, respectively. (*) denotes significant decrease of inflammatory cells in IDO groups compared to the controls (P<0.001, n = 5).
Fig 4
Fig 4. Expression of co-inhibitory and anti-inflammatory molecules by dermal fibroblasts and tracking of fibroblasts after intraperitoneal injection.
IDO-expressing and control mouse dermal fibroblasts were examined for mRNA expression of co-inhibitory molecules Programmed Cell Death 1 and 2 ligands (PD-L1 and PD-L2), anti-inflammatory molecule interleukin-1 receptor antagonist (IL1-Ra), IDO-1, and GAPDH (as loading control) using reverse transcriptase PCR. Panel A shows high levels and almost equal expression of anti-inflammatory molecules in both control (CTL) and IDO-expressing fibroblasts. To track fibroblasts, IDO-expressing and control fibroblasts were labeled using PKH26 red fluorescent cell membrane labeling kit and injected to NOD mice intraperitoneally. Animals were euthanized at four time points with 14 days intervals (i.e. week 2 to week 8). Cells were harvested from peritoneal cavity (by lavage; Perit. Lav.), mesenteric membranous tissue (Mesentry), mesenteric lymph nodes (Mesent. Ln.), and retro-peritoneal lymph nodes (Retro-Perit. Ln.). Cells were stained for CD90.2 (a fibroblast marker) and PD-L1 and tested using flow cytometry. Panels B to E show frequency of PKH26+ cells at different time points (Week 2 to Week 8; W2to W8) in harvested tissues as indicated in the bottom of the panels. Black circles and white circles represent IDO-expressing and control fibroblasts, respectively. (*) denotes significant difference in frequency of PKH26+ in IDO fibroblast-injected mice versus control fibroblast-injected mice (p<0.001, n = 5). Panels F to I show frequencies of CD90+ PD-L1+ cells in PKH26+ gate. Black bars and white bars represent IDO-expressing and control fibroblasts, respectively. Panels J & K show frozen sections of mesenteric lymph nodes, two weeks following IP injection of control or IDO fibroblast, respectively. White arrows point to red fluorescent (PKH26+) cells among lymph node cells with blue colored nuclei (DAPI stained). Scale bar: 10 μm.
Fig 5
Fig 5. Increased frequency of CD4+ CD25+ FoxP3+ cells after co-culture with IDO-expressing fibroblast.
Lymphocytes were isolated from mesenteric lymph nodes of recently diabetic NOD mice and were cultured either alone or co-cultured with control or IDO-expressing fibroblasts as described in Methods. The frequency of CD4+ gated CD25+ FoxP3+lypmocytes were measured using flow cytometry before culturing (Pre-culture) and after 7 days mono- or co-cultures. Panel A shows representative plots for CD25+ FoxP3+ regulatory T cells (gated on CD4+ T cells) and panel B shows frequencies of these cells in different experimental groups. (*) denotes statistically significant increase in the frequency of regulatory T cell after co-culturing with IDO-expressing fibroblasts compared to pre-culture amount (P<0.001, n = 3).
Fig 6
Fig 6. Expression of CC-chemokine receptor 7 (CCR7) by dermal fibroblasts.
IDO-expressing and control mouse dermal fibroblasts were examined for expression of CCR7 before and after intraperitoneal injection. To track them, fibroblasts were labeled with PKH26 red fluorescent cell membrane labeling kit. Panel A shows representative flow cytometry plots showing frequency of CCR7+ cells gated on PKH26+ CD90+ window. Upper and bottom rows show IDO-expressing and control fibroblasts data, respectively. The plots on the left side show fibroblasts before peritoneal injection (Before inj.). The middle plots and right side plots show cells that were harvested from peritoneal cavity by lavage (Perit. Lav.) or extracted from mesenteric lymph nodes (Mesent. Ln.) two weeks post IP injection, respectively. Panel B show quantification of CCR7 expression on IDO-expressing and control fibroblasts before and after IP injection in peritoneal cavity and mesentric lymph nodes. (*) denotes statistically significant difference in CCR7 expression on IDO and control fibroblasts following IP injection mice compared to the before injection level (P<0.0001, n = 5).
Fig 7
Fig 7. Schematic illustration of the possible mechanism for IDO fibroblast therapy model.
IDO-expressing fibroblasts are injected to recently diabetic mice, intraperitoneally. A group of these fibroblasts express CCR7 and migrate to regional lymph nodes. These fibroblasts get in touch with naive T cells there and via IDO and PD-L1/2 medicated mechanisms convert them into regulatory T cells. These Tregs will then reach pancreas through blood and lymphatic streams and suppress autoimmune and inflammatory effectors immune cells. As a result, insulitis level diminishes and consequently diabetes is controlled.

Similar articles

See all similar articles

Cited by 3 articles


    1. Bluestone JA, Herold K, Eisenbarth G. Genetics, pathogenesis and clinical interventions in type 1 diabetes. Nature. 2010;464: 1293–1300. - PMC - PubMed
    1. Ablamunits V, Sherry NA, Kushner JA, Herold KC. Autoimmunity and beta cell regeneration in mouse and human type 1 diabetes: The peace is not enough. Ann N Y Acad Sci. 2007;1103:19–32. - PubMed
    1. Clemente-Casares X, Tsai S, Huang C, Santamaria P. Antigen-specific therapeutic approaches in type 1 diabetes. Cold Spring Harb Perspect Med. 2012. February;2(2):a007773 10.1101/cshperspect.a007773 - DOI - PMC - PubMed
    1. Godfrey KJ, Mathew B, Bulman JC, Shah O, Clement S, Gallicano GI. Stem cell-based treatments for Type 1 diabetes mellitus: bone marrow, embryonic, hepatic, pancreatic and induced pluripotent stem cells. Diabet Med. 2012;29(1): 14–23. 10.1111/j.1464-5491.2011.03433.x - DOI - PubMed
    1. Aguayo-Mazzucato C, Bonner-Weir S. Stem cell therapy for type 1 diabetes mellitus. Nat Rev Endocrinol. 2010;6(3): 139–48. 10.1038/nrendo.2009.274 - DOI - PubMed

Publication types

MeSH terms