The PI3K pathway as a therapeutic intervention point in inflammatory bowel disease

Abstract

With glucose being the preferred source of energy in activated T cells, targeting glycolysis has become an attractive therapeutic intervention point for chronic inflammatory bowel diseases (IBD). The switch to glycolysis is mediated by phosphoinositide-3-kinases (PI3K) which relay signals from surface receptors to the AKT pathway. We first confirmed by analysis of the oxygen consumption rate (OCR) and extracellular acidification rate (ECAR) that metabolism is shifted towards glycolysis in IBD patients as compared to non-IBD donors. In contrast to non-IBD donors, OCR correlated with ECAR (IBD: cor = 0.79, p = 2E-10; non-IBD: cor = 0.37, p = n.s.), in IBD patients. Second, we tested the PI3K inhibitor copanlisib as a potential therapeutic. Ex vivo, copanlisib suppressed the ECAR significantly in T cells activated by anti-CD3 antibodies and significantly decreased ECAR rates in the presence of copanlisib (anti-CD3: 58.24 ± 29.06; copanlisib: 43.16 ± 20.23, p < .000. In addition, copanlisib impaired the activation of CD4⁺ CD25⁺ T cells (anti-CD3: 42.15 ± 21.46; anti-CD3 + copanlisib: 26.06 ± 21.82 p = .013) and the secretion of cytokines (IFNγ: anti-CD3: 6332.0 ± 5707.61 pmol/ml; anti-CD3 + copanlisib: 6332.0 ± 5707.61, p = .018). In vivo, copanlisib significantly improved the histological scores (ethanol: 8.5 ± 3.81; copanlisib: 4.57 ± 2.82, p = .006) in the NSG-UC mouse model. Orthogonal partial least square analysis confirmed the efficacy of copanlisib. These data suggest that the PI3K pathway provides an attractive therapeutic intervention point in IBD for patients in relapse. Targeting metabolic pathways have the potential to develop phase dependent therapies.

Keywords: NSG-UC mouse model; PI3K; copanlisib; immune-metabolism; inflammatory bowel disease; ulcerative colitis.

Figure 1

Activation status of T cells and metabolic activity is affected in IBD patients. (A) Flow cytometric analysis of early‐activated (CD4⁺ CD69⁺) and late‐activated (CD4⁺ CD134⁺) in PBMCs from IBD (number of donors N = 77) and non‐IBD (number of donors N = 36). Frequencies are depicted as Cumming plots. The upper part of the plot presents each data point in a swarm plot. The mean and SD of each group is plotted as a gapped line, where the vertical lines correspond to the mean ± SD and the mean itself is depicted as a gap in the line. In the lower panel of the plots, the effect sizes are shown. The 0 point of the difference axis is based on the mean of the reference group (control). The dots show the difference between groups (effect size/mean difference). The shaded curve shows the entire distribution of excepted sampling error for the difference between the means (the higher the peak, the smaller the error). The error bar in the filled circles indicates the 95% confidence interval (bootstrapped) for the difference between means. (B) Representative energy maps of IBD patients and non‐IBD patients determined by Seahorse technology. (C) Pearson's product‐moment correlation analysis of ECAR and OCR rates (IBD: number of donors N = 10, experiments were performed in triplicates or quintuplicate, number of samples n = 42; non‐IBD: number of donors N = 6, number of samples n = 18). Cor, correlation coefficient; ECAR, extracellular acidification rate; OCR, oxygen consumption rate

Figure 2

Copanlisib affects glycolysis in activated CD4⁺ T cells. OCR and ECAR analysis of PBMCs performed with two different donors (N = 2) incubated for 2h in the presence or absence anti‐CD3 monoclonal antibodies (61 ng/ml) and copanlisib (100 nM). Measurements were taken over the course of 1 h at 10 consecutive time points. The copanlisib group was measured in duplicates. Number of measurements (control: n = 20; anti‐CD3: n = 20; copanlisib: n = 40) Groups compared: untreated control (control), activated CD4⁺ T cells (anti‐CD3) and activated CD4⁺ T cells + copanlisib (anti‐CD3 + copanlisib). (A) Representative energy map of control‐, anti‐CD3, and anti‐CD3 + copanlisib groups. (B) Measurements of consecutive time points depicted as Cumming plots. The upper part of the plot presents each data point in a swarm plot. The mean and SD of each group is plotted as a gapped line, where the vertical lines correspond to the mean ± SD and the mean itself is depicted as a gap in the line. The 0 point of the difference axis is based on the mean of the reference group (control). The dots show the difference between groups (effect size/mean difference). The shaded curve shows the entire distribution of excepted sampling error for the difference between the means (the higher the peak, the smaller the error). The error bar in the filled circles indicates the 95% confidence interval (bootstrapped) for the difference between means. ECAR, extracellular acidification rate; OCR, oxygen consumption rate; PBMC, peripheral blood mononuclear cell

Figure 3

Copanlisib impairs the activation of T cells and secretion of cytokines. PBMCs were incubated in the presence or absence of anti‐CD3 antibodies (62 ng/ml) and copanlisib (100 nM) for 72 h. The experiments were performed on different days. Groups compared: untreated control (control), activated CD4⁺ T cells (anti‐CD3) and activated CD4⁺ T cells + copanlisib (anti‐CD3 + copanlisib). (A) Flow cytometric analysis of frequencies of CD4⁺CD25⁺, CD4⁺CD69⁺, CD4⁺CD103⁺, and CD4⁺CD134⁺ T cells depicted as Cumming plots. Experiments were performed with seven donors (N = 7) in triplicates or quadruplicates, number of experiments n = 23). (B) Levels of cytokines depicted as Cumming plots. Supernatants of cell cultures were analyzed by Luminex assays (control: experiments were repeated with two donors (N = 2, and performed in triplicates, number of experiments n = 6; anti‐CD3 and copanlisib: number of donors N = 3, experiments were performed in triplicates or quadruplicate, number of experiments n = 9, n = 10 respectively). The upper part of the plot presents each data point in a swarm plot. The mean and SD of each group is plotted as a gapped line, the vertical lines correspond to the mean ± SD and the mean itself is depicted as a gap in the line. For the comparison of the groups. In the lower panel of the plots, the effect sizes are shown. The 0 point of the difference axis is based on the mean of the reference group (control). The dots show the difference between groups (effect size/mean difference). The shaded curve shows the entire distribution of excepted sampling. PBMC, peripheral blood mononuclear cell

Figure 4

Copanlisib ameliorates symptoms and pathological phenotype in NSG‐UC mice. NSG‐UC mice were engrafted with PBMCs derived from two UC patients, challenged with 10% ethanol at Day 7, and 50% ethanol on Day 14 and treated with 6 mg/kg copanlisib or carrier on Days 6, 7, and 14–16. Groups compared: unchallenged control (control, number of donors N = 1, number of mice n = 8), ethanol challenged control (EtOH, number of donors N = 2, number of mice n = 14), ethanol challenged and treated with copanlisib (copanlisib, numer of number of donors N = 2, number of mice, n = 14). (A) Clinical‐, colon‐ and histological scores depicted as Cumming plots. The upper part of the plot presents each data point in a swarm plot. The mean and SD of each group is plotted as a gapped line, where the vertical lines correspond to the mean ± SD and the mean itself is depicted as a gap in the line. In the lower panel of the plots, the effect sizes are shown. The 0 point of the difference axis is based on the mean of the reference group (control). The dots show the difference between groups (effect size/mean difference). The shaded curve shows the entire distribution of excepted sampling error for the difference between the means (the higher the peak, the smaller the error). The error bar in the filled circles indicates the 95% confidence interval (bootstrapped) for the difference between means. (B) Representative macrophotographs of colons. (C) Histological manifestations. (a) Hematoxilin (HE), (b) Masson Goldner Trichrome (MGT), (c) Periodic Acid Schiff (PAS). Solid arrows indicate edema and influx of inflammatory cells, dashed arrows epithelial erosions, bold arrows show fibrosis, and arrows with a circle arrowhead indicate goblet cell loss. PBMC, peripheral blood mononuclear cell; UC, Ulcerative colitis

Figure 5

Copanlisisib affects frequencies of early activated CD4⁺ T cells in vivo. Mice were treated as described in Figure 4. Flow cytometric analysis of frequencies of CD4⁺CD69⁺, CD4⁺CD25⁺, CD4⁺CD103⁺, and CD4⁺CD134⁺ T cells depicted as Cumming plots. The upper part of the plot presents each data point in a swarm plot. The mean and SD of each group is plotted as a gapped line, the vertical lines correspond to the mean ± SD and the mean itself is depicted as a gap in the line. For the comparison of the groups. In the lower panel of the plots, the effect sizes are shown. The 0 point of the difference axis is based on the mean of the reference group (control). The dots show the difference between groups (effect size/mean difference). The shaded curve shows the entire distribution of excepted sampling

Figure 6

oPLS‐DA analysis support in vivo efficacy of copanlisib. Mice were treated as described in Figure 4. Clinical, colonic, and histological scores and frequencies of CD4⁺ CD69⁺ and CD4⁺ CD134⁺ cells were selected as variables. Orthogonal partial least square discrimination analysis (o‐PLS‐DA). Top left: inertia barplot; Top right: significance diagnostic: the pR2Y and pQ2 of the model are compared with the corresponding values obtained after random permutation of the y response. Bottom left: Outlier diagnostics. Bottom right: X‐score plot: the number of components and the cumulative R2X, R2Y, and Q2Y are indicated below the plot. R2Y: fraction of the variation of the X variables explained by the model, R2X: fraction of the variation of the Y variables explained by the model. Q2Y: fraction of the variation of the Y variables predicted by the model, root mean square error of estimation (RMSEE) value