Thymosin α1 protects from CTLA-4 intestinal immunopathology

Abstract

The advent of immune checkpoint inhibitors has represented a major boost in cancer therapy, but safety concerns are increasingly being recognized. Indeed, although beneficial at the tumor site, unlocking a safeguard mechanism of the immune response may trigger autoimmune-like effects at the periphery, thus making the safety of immune checkpoint inhibitors a research priority. Herein, we demonstrate that thymosin α1 (Tα1), an endogenous peptide with immunomodulatory activities, can protect mice from intestinal toxicity in a murine model of immune checkpoint inhibitor-induced colitis. Specifically, Tα1 efficiently prevented immune adverse pathology in the gut by promoting the indoleamine 2,3-dioxygenase (IDO) 1-dependent tolerogenic immune pathway. Notably, Tα1 did not induce IDO1 in the tumor microenvironment, but rather modulated the infiltration of T-cell subsets by inverting the ratio between CD8+ and Treg cells, an effect that may depend on Tα1 ability to regulate the differentiation and chemokine expression profile of DCs. Thus, through distinct mechanisms that are contingent upon the context, Tα1 represents a plausible candidate to improve the safety/efficacy profile of immune checkpoint inhibitors.

Conflict of interest statement

A patent application on “Thymosin α1 for use in treatment of checkpoint inhibitor immune adverse events” by G Renga, MM Bellet, M Pariano, C Costantini, E Garaci, and L Romani is pending (IT201900016310).

© 2020 Renga et al.

Figures

Figure 1.. Thymosin α1 (Tα1) protects mice from DSS-induced colitis.

C57BL/6 mice were subjected to DSS-induced colitis for 1 wk followed by a recovery period of another week. Fresh DSS solution was added at day +3. Tα1 was administered every other day at the dose of 200 μg/kg, intraperitoneally. (A, B, C, D, E, F, G, H) Mice were evaluated for (A) % weight change, (B) % survival, (C) gross pathology, (D) colon length (cm), (E) colon histology (hematoxylin and eosin staining), (F) histology score, (G) levels of inflammatory cytokines, and (H) IL-10 in colon homogenates and Ido1 expression. Cytokines were determined by ELISA and gene expression was performed by RT-PCR (data are presented as mean ± SD of three independent experiments). Images were taken with a high-resolution microscope (Olympus BX51), 20× magnification (scale bars, 200 μm). For histology, data are representative of three independent experiments. Each in vivo experiment includes four mice per group. Weight change and survival are calculated on a total of 12 mice per group. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, Tα1-treated versus untreated (DSS) mice. Two-way ANOVA, Bonferroni or Tukey’s post hoc test. None, mice with DSS colitis only. Naїve, untreated mice.

Figure 2.. Thymosin α1 (Tα1) protects mice from DSS plus anti–CTLA-4–induced colitis.

C57BL/6 mice were subjected to DSS-induced colitis for 1 wk followed by a recovery period of another week and administered 100 μg of anti–CTLA-4 mAb or isotype control twice (at days 0, 4, and 8 after DSS administration). Tα1 was administered every other day at a dose of 200 μg/kg, intraperitoneally. (A, B, C, D, E, F) Mice were evaluated for (A) % weight change, (B) % survival, (C) disease activity index, (D) rectal bleeding, (E) gross pathology, and (F) colon length (cm). Each in vivo experiment includes four mice per group. Weight change and survival are calculated on a total of 12 mice per group. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, Tα1-treated versus untreated (DSS plus anti–CTLA-4 only) mice. Two-way ANOVA, Bonferroni or Tukey’s post hoc test. None, mice with DSS colitis only. Naїve, untreated mice.

Figure 3.. Thymosin α1 (Tα1) prevents epithelial damage in DSS plus anti–CTLA-4–induced colitis.

C57BL/6 mice were subjected to DSS plus anti–CTLA-4–induced colitis and administered Tα1 as described in the legend of Fig 2. (A, B, C, D, E) Mice were evaluated for (A) colon histology (periodic acid-Schiff staining), (B) histology score, (C) Ki-67 expression and TUNEL staining, (D) dextran-FITC levels in the serum, and (E) fungal growth (log10 CFUs) in the mesenteric lymph nodes. For immunofluorescence, nuclei were counterstained with DAPI. Photographs were taken with a high-resolution microscope (Olympus BX51), 20× magnification (scale bars, 200 μm). For histology and immunofluorescence, data are representative of two independent experiments. In vivo experiment includes four mice per group. *P < 0.05, Tα1-treated versus untreated (DSS plus anti–CTLA-4 only) mice. One-way ANOVA, Bonferroni or Tukey’s post hoc test. None, mice with DSS colitis only. Naїve, untreated mice.

Figure 4.. Thymosin α1 (Tα1) prevents epithelial damage in DSS plus anti–CTLA-4–induced colitis via IDO1.

C57BL/6 or Indo−/− mice were subjected to DSS plus anti–CTLA-4–induced colitis and administered Tα1 as described in the legend of Fig 2. (A, B, C, D, E) C57BL/6 mice were evaluated for (A) Ido1 gene and (B) IDO1 protein expression, (C) kynurenine (Kyn), tryptophan (Trp) levels, and Kyn/Trp ratio, (D) IL-10, IL-1β and IL-17A levels in colon homogenates, and (E) Ikaros and Foxp3 expression in mesenteric lymph node. (F, G, H)Indo−/− mice were evaluated for (F) colon histology (periodic acid-Schiff staining) and expression of Foxp3- and Rorγt-positive cells, (G) histology score and (H) Il10 and Foxp3 expression in colon. Cytokines were determined by ELISA and gene expression was performed by RT-PCR (data are presented as mean ± SD or ± SEM of three independent experiments). For immunofluorescence, nuclei were counterstained with DAPI. Photographs were taken with a high-resolution microscope (Olympus BX51), 20× (scale bars, 200 μm) and 40× magnification (scale bars, 100 μm) for histology and immunofluorescence, respectively. For histology and immunofluorescence, data are representative of two independent experiments. In vivo experiment includes four mice per group. *P < 0.05, Tα1-treated versus untreated (DSS plus anti–CTLA-4 only) mice. t test or one-way ANOVA, Bonferroni or Tukey’s post hoc test. None, mice with DSS colitis only. Naїve, untreated mice.

Figure 5.. Thymosin α1 (Tα1) preserves anti–CTLA-4 antitumor activity in melanoma.

C57BL/6 or Indo−/− mice were subcutaneously injected with B16 tumor cells and administered 100 μg anti–CTLA-4 mAb or isotype control intraperitoneally four times at 3-d intervals up to 15 d. Tα1 was administered every other day at a dose of 200 μg/kg, intraperitoneally. (A, B, C, D, E, F, G, H, I) Mice were evaluated for (A) tumor growth, (B) histology (H&E staining), CD8+ cells infiltration and tumor cell death (TUNEL), (C) local expression of GzmB and Perforin, (D, E) frequency of CD8+CD4+ T cells and (F, G) CD25+Foxp3+ T cells in tumor homogenates, quantified by flow cytometry (H) Ido1 gene and (I) kynurenine (Kyn), tryptophan (Trp) levels and Kyn/Trp ratio. Gene expression was performed by RT-PCR (data are presented as mean ± SD of two independent experiments). Tumor growth data are presented as mean ± SEM of two independent experiments. Black arrows in histology sections indicate the presence of dark pigmented cells. For immunofluorescence, nuclei were counterstained with DAPI. Photographs were taken with a high-resolution microscope (Olympus BX51), 10× magnification (scale bars, 500 μm) for histology, and TUNEL assay, 40× magnification (scale bars, 100 μm) for immunofluorescence. For histology and immunofluorescence, data are representative of two independent experiments. In vivo experiment includes 3–6 mice per group. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, Tα1-treated versus untreated (anti–CTLA-4) mice. t test or one-way ANOVA, Bonferroni or Tukey’s post hoc test. None, untreated mice.

Figure S1.. C57BL/6 mice were intravenously injected with LLC1 tumor cells and administered 200 μg anti-PD1 mAb or isotype control intraperitoneally four times at 3-d intervals up to 15 d.

Thymosin α1 was administered every other day at the dose of 200 μg/kg, intraperitoneally. (A, B, C, D, E) Mice were evaluated for (A) lung weight, (B) gross pathology (scale bars, 1 mm), (C) local expression of GzmB and Perforin, and (D, E) frequency of CD25+Foxp3+ T cells in lung homogenates.

Figure 6.. Thymosin α1 (Tα1) induces the differentiation of DC from bone marrow precursors with a specific gene expression profile.

(A) Chemokine gene expression in C57BL/6 mice with B16 melanoma, treated with 100 μg anti–CTLA-4 mAb with and without Tα1 as in legend to Fig 5. Gene expression was performed by RT-PCR. Data are presented as mean ± SD of two independent experiments. (B, C) Frequency of MHCII+-CXCL9+ cells in tumor homogenates, quantified by flow cytometry. (D, E) Bone marrow precursor cells were treated with Tα1 alone (Tα1-DC), GM-CSF/IL-4 (GM-DC), or FLT3 ligand (FL-DC) and evaluated for (D) morphology by light and electron microscopy (representative images of two independent experiments) and (E) expression of CD11c, B220, and CD11b by flow cytometry. (F) Percentage of CD11b+/CD11c+ cells obtained from primary cultures of bone marrow cells prepared from mice deficient of selected TLRs as well as associated adapters and exposed to Tα1 evaluated by flow cytometry. (G) Gene expression in DC by microarray. The genes up-regulated, or down-regulated, in Tα1-DC compared to both GM-DC (black dots) and FL-DC (white dots) are indicated. Each dot represents a biological replicate.

Figure S2.

Thymosin α1 (Tα1) does not modify the frequency of CD11b+F4/80+ cells in melanoma. (A, B) Frequency of CD11b+F4/80+ cells in tumor homogenates, quantified by flow cytometry.

Figure S3.

Cytokine and receptor microarrays in bone marrow–derived DCs. (A, B) Cytokines (A) and receptors (B) microarrays in DCs from bone marrow precursors treated with thymosin α1 (Tα1) alone (Tα1-DC), GM-CSF/IL-4 (GM-DC), or FLT3 ligand (FL-DC), as detailed in the Materials and Methods section.

Figure S4.. Gating strategy for flow cytometry analysis.