Suppression of murine colitis and its associated cancer by carcinoembryonic antigen-specific regulatory T cells

Abstract

The adoptive transfer of regulatory T cells (Tregs) offers a promising strategy to combat pathologies that are characterized by aberrant immune activation, including graft rejection and autoinflammatory diseases. Expression of a chimeric antigen receptor (CAR) gene in Tregs redirects them to the site of autoimmune activity, thereby increasing their suppressive efficiency while avoiding systemic immunosuppression. Since carcinoembryonic antigen (CEA) has been shown to be overexpressed in both human colitis and colorectal cancer, we treated CEA-transgenic mice that were induced to develop colitis with CEA-specific CAR Tregs. Two disease models were employed: T-cell-transfer colitis as well as the azoxymethane-dextran sodium sulfate model for colitis-associated colorectal cancer. Systemically administered CEA-specific (but not control) CAR Tregs accumulated in the colons of diseased mice. In both model systems, CEA-specific CAR Tregs suppressed the severity of colitis compared to control Tregs. Moreover, in the azoxymethane-dextran sodium sulfate model, CEA-specific CAR Tregs significantly decreased the subsequent colorectal tumor burden. Our data demonstrate that CEA-specific CAR Tregs exhibit a promising potential in ameliorating ulcerative colitis and in hindering colorectal cancer development. Collectively, this study provides a proof of concept for the therapeutic potential of CAR Tregs in colitis patients as well as in other autoimmune inflammatory disorders.

Figure 1

Transduction efficiency of CARS into Teffs and Tregs. (a) Molecular configuration of the CARs. (b) Transduction efficiency of CD4⁺ Teffs with irrelevant CAR (left) or CEA-specific CAR (right), as measured by flow cytometry. (c) Transduction efficiency and %FoxP3 in CD4⁺CD25⁺ Tregs transduced with irrelevant CAR (left) or CEA-specific CAR (right). Expression of the CEA-specific CAR was detected by incubating the cells with soluble CEA followed by staining with an anti-CEA antibody. The irrelevant CAR was detected using biotinylated antigen followed by staining with fluorophore-labeled streptavidin. Shaded curves in (b) represent staining of each CAR with the opposite CAR's antigen. CAR, chimeric antigen receptor; CEA, carcinoembryonic antigen.

Figure 2

In vitro stimulation and cytotoxicity of CEA-specific CAR T cells. (a,b) Colorectal tumors were extracted from CEABAC-2 mice and dissociated to serve as target cells for CD4⁺ CAR Teff stimulation. (a) In vitro proliferation, as indicated by CFSE dilution. CD4⁺ CAR Teffs were labeled with CFSE and cocultured with target cells for 7 days prior to FACS analysis. (b) IFN-γ (top) and IL-2 (bottom) secretion. Supernatants were collected after 48 hours and analyzed by ELISA. (c) Cytotoxicity of CEA-specific CAR CD4⁺ Teffs, CD8⁺ Teffs, and Tregs towards CEA-expressing Capan-1 cells transfected with firefly luciferase, as measured by bioluminescence signal intensity. Control: Irrelevant CAR Teffs. *P < 0.05, **P < 0.01 compared to no T cells. CAR, chimeric antigen receptor; CEA, carcinoembryonic antigen; ELISA, enzyme-linked immunosorbent assay; Teffs, effector T cells; Tregs, regulatory T cells.

Figure 3

In vitro function of CAR Tregs. (a) In vitro cytokine secretion by irrelevant CAR Tregs, CEA-specific CAR Tregs, and CEA-specific CAR Teffs. CAR T cells were cocultured with CEA-expressing murine colorectal tumor cells, and supernatants were collected after 48 hours. (b) CEA-specific CAR Teffs prelabeled with CFSE were cocultured with CEA-expressing tumor cells and increasing numbers of CAR Tregs. CFSE signals were analyzed after 7 days as a measure of Teff proliferation. (c) The proliferation index, shown as percent of maximum proliferation (no Tregs), was extracted from CFSE histograms by calculating the sum of the cells in all generations divided by the calculated number of original parent cells. (d) Treg-mediated suppression of IL-2 secretion. CEA-specific CAR Teffs were stimulated to secrete IL-2 and increasing amounts of CAR Tregs were added. One experiment representative of several independent experiments is shown. Error bars represent the standard error of the mean of samples within a group (n = 3–4). *P < 0.05, **P < 0.01, ***P < 0.001. CAR, chimeric antigen receptor; CEA, carcinoembryonic antigen; Teffs, effector T cells; Tregs, regulatory T cells.

Figure 4

Induction of colitis in mice by adoptive transfer of CEA-specific CAR Teffs. (a,b) CEABAC-10 mice were preconditioned with 0, 2.5, or 5 Gy of TBI, followed by transfer of CD4⁺ or CD8⁺ CEA-specific CAR Teffs 24 hours later (day 0). Weight progression (a) and day 7 endoscopic colitis scores (b) are shown. (c) CEABAC-2 and CEABAC-10 mice and their wild-type littermates were administered with CEA-specific CD4⁺ CAR Teffs, and colitis was monitored endoscopically over time. Error bars represent the standard error of the mean of samples within a group. **P < 0.01, ***P < 0.001. (d) Histological sections from colons of CEABAC-10 mice at different time points following colitis induction (acute: day 10; chronic: day 45). Bar = 80 μm. (e) Bioluminescence monitoring of mice injected with CEA-specific CD4⁺ CAR Teffs. CAR, chimeric antigen receptor; CEA, carcinoembryonic antigen; TBI, total body irradiation; Teffs, effector T cells; Tregs, regulatory T cells.

Figure 5

CEA-specific CAR Tregs protect mice from transfer colitis-related death. Lethal colitis was induced in CEABAC-10 mice by 5 Gy of TBI followed by transfer of 1.5 × 10⁶ CEA-specific CAR CD4⁺ Teffs. Some mice also received high dose (1.5 × 10⁶ cells) or low dose (0.75 × 10⁶) infusion of CEA-specific or irrelevant CAR Tregs. CAR, chimeric antigen receptor; CEA, carcinoembryonic antigen; TBI, total body irradiation; Teffs, effector T cells; Tregs, regulatory T cells.

Figure 6

Suppression of transfer colitis by CEA-specific CAR Tregs. Colitis was induced in CEABAC-10 mice by 4 Gy of TBI followed by transfer of 1.5 × 10⁶ CEA-specific CAR CD4⁺ Teffs. Recipient mice received no Tregs, irrelevant CAR Tregs, or CEA-specific CAR Tregs. (a) Bioluminescence images taken on day 6 show a dramatic Treg-mediated suppression of CD4⁺ CAR Teff expansion. (b) Average total bioluminescence signal from the abdominal section in each of the three groups. (c) Weight progression. (d) Average colitis scores, and (e) representative colonoscopic images. (f) Bioluminescence monitoring of luciferase expressing Tregs in gut and spleen of mice sacrificed at different time points after infusion. One representative experiment of several independent experiments is shown. Error bars represent the standard error of the mean of samples within a group (n = 7). *P < 0.05, **P < 0.01. CAR, chimeric antigen receptor; CEA, carcinoembryonic antigen; TBI, total body irradiation; Teffs, effector T cells; Tregs, regulatory T cells.

Figure 7

Adoptive transfer of CEA-specific CAR Tregs ameliorates colitis and colitis-associated colorectal cancer in mice. Colitis and colorectal tumors were induced in mice with AOM–DSS. (a) Immunohistochemistry staining reveals elevated CEA levels in inflamed (right) compared to normal (center) CEABAC-2 colons. Bar = 100 μm. (b) Representative day 10 colonoscopic images from mice treated with irrelevant CAR Tregs or CEA-specific CAR Tregs. (c) Average colitis scores. (d) Day 56 colonoscopic images showing decreased tumor burden in mice treated with CEA-specific CAR Tregs compared to mice treated with irrelevant CAR Tregs. (e) Relative tumor scores. Data are cumulative from two independent experiments. Error bars represent the standard error of the mean of samples within a group (n = 13). *P < 0.05. AOM–DSS, azoxymethane–dextran sodium sulfate; CAR, chimeric antigen receptor; CEA, carcinoembryonic antigen; Teffs, effector T cells; Tregs, regulatory T cells.