Background: We have recently shown that Tregs isolated from the peripheral blood of patients with Crohn's Disease (CD) play a critical role in controlling both phenotype and expansion of auto-reactive T cells (1). This is a critical step to developing cell based therapy for Crohn's disease, where where primary and secondary non response rates to biologics remain unacceptably high. The immune system used retinoic acid (ATRA) to induce a4b7 and thus prime Tregs to home to the gut. We sought use ATRA in-vitro in order to engineer gut specific Tregs for our Phase 1 trial in Crohn's Disease. We then validated our findings in-vitro and in-vivo.
Methods: Tregs were isolated from peripheral blood of CD patients. ATRA supplementation was tested in standard culture conditions. The expression of a4b7 was assessed by flow cytometry. Suppression assays were performed using autologous effector T cells (Teff). An ibidi flow chamber system coated with recombinant human MAdCAM-1 was used for in-vitro trafficking experiments. SCID mice xenografted with foetal intestinal small bowel were used for in-vivo experiments. Parametric and non-parametric data were calculated as the mean ± s.d. and median (IQR). For comparison of parametric and non-parametric data, t-test, or ANOVA were used.
Results:Ex-vivo expansion of Tregs in the presence of ATRA significantly induced the expression of a4b7 compared to Rapamycin alone (5.57%±3.12 vs 82.8%±9.5, p=0.0057) Cells treated with Rapa+ATRA maintained their superior suppressive ability compared to Rapamycin treated Tregs (95.8%±3.5% vs 91.15%±10.1% p=ns; at Treg:Teff 1:1 ratio). RAPA+ATRA Tregs did not produce IFNy or IL17 under pro-inflammatory cytokine challenge. When flowed through a MAdCAm-1 coated chamber, significantly higher numbers of Rapa+ATRA treated cells were observed to roll (Rapa 0.83±0.40 vs Rapa+ATRA 10.17±2.54 p=0.005), crawl (Rapa 0 vs Rapa+ATRA 4±0.89 p=0.001) and firmly adhere (Rapa 0.33±0.21 vs Rapa+ATRA 36.8±1.78, p<0.001) than those treated with Rapa alone. When Tregs were transferred into mice, a higher proportion of Tregs were found in xenografts of animals treated with Rapa+ATRA Tregs compared Rapa Tregs (12.10 (7.54–22.83) vs 4.97 (1.72–7.63), p=0.0056). Importantly there was a higher proportion of Tregs in inflamed xenografts of animals treated with Rapa+ATRA Tregs compared to those treated with Rapa Tregs (18.35 (12.95–28.63) vs 6.78 (2.65–9.61), p=0.0095).
Conclusions: The addition of ATRA to Treg culture in-vitro confers a gut homing phenotype. This is functionally relevant in-vitro and in-vivo. The treatment maintains the highly suppressive and phenotypically stable phenotype of these cells. These gut specific Tregs will be implemented in our first in man trial Treg therapy for Crohn's disease.