Adoptive T cell transfer-mediated colitis was first described by Morrissey et al. in 1993 [J. Exp. Med. 178, 237-244: 1993], who reported inflammatory changes in the large bowel following the transfer of CD4+CD45RBhi T cells.
CD45RBhi T cells promote pro-inflammatory, Th1 immunity that is dependent on production of IFN-γ, IL-12, IL-18 and TNF-α; in contrast, CD45RBlo T cells promote humoral, Th2 immunity that is mediated by immunoglobulin production by B cells.
The colitis observed involved epithelial hyperplasia and a substantial influx of inflammatory cells into the mucosa.
Subsequently, the model has been accepted as demonstrating many features that are characteristic of human Crohn's disease: it is dependent on the response of CD4+ T cells to antigens derived from the gut flora [Niess et al. J. Immunol. 180, 559-568: 2008]; Th17 cells are induced, giving rise to chronic disease [Leppkes et al. Gastroenterol. 136, 257-267: 2008]; and gene expression profiles are similar to human Crohn's disease [te Velde et al. Inflamm. Bowel Dis. 13, 325-330: 2007].
At Epistem, we have been running the model using CD4+CD62L+ T cells from BALC/c donor mice, and CB-17 Prkdcscid mice as recipients, as first described by Wirtz and colleagues [Wirtz et al. J. Immunol. 168, 411-420: 2002].
This model gives reproducible colitis, with a consistent onset time for disease symptoms and severity of microscopic histopathology.
Diarrhoea is evident from approximately 2 weeks post-T cell transfer, with the subsequent development of weight loss, relative to untreated mice.
Marked epithelial hyperproliferation and inflammatory cell infiltration are characteristic of the colitis that develops following adoptive transfer of CD4+CD62L+ T cells; other features of the colitis include increased frequency of crypt fission, erosions, ulceration and the presence of crypt abscesses.
Both CD62L and CD45RBhi are markers of naïve T cells and show overlapping expression [Wirtz et al. J. Immunol. 168, 411-420: 2002]. In a head-to-head study, using cells of the same donor origin and age- and sex-matched recipients, we have observed similar disease time-course and severity using either of these two T cell populations. Data for changes in body weight, diarrhoea score and histopathology in in the mid-colon are illustrated in Figure 1, Figure 2 and Figure 3, respectively.
T cell-recipient mice demonstrate a period of sub-clinical disease with no symptoms and are typically observed for a period of 2-3 weeks with clinical disease, which is mild/moderate in its severity.
This model offers the ability to study test items using a range of dosing regimens, from dosing in a prophylactic manner through to examining efficacy in ameliorating early stage or established disease. The adoptive T cell transfer model of colitis is usually run over 4-5 weeks, in order to achieve sufficient disease penetrance (>90%) and observe test item efficacy.
Colitis results from the activation of donor T cells by luminal antigens resulting in chronic inflammation and damage to the large bowel mucosa. T cell-recipient mice develop diarrhoea and demonstrate weight loss relative to untreated mice, from 2 weeks post-T cell transfer.
Histologically, the colitis is characterised by discontinuous lesions in the large bowel, with epithelial hyperplasia, erosion and occasional ulceration; affected areas demonstrate significant infiltration of lymphoid cells into the mucosa and sub-mucosa.
The model is responsive to a range of therapies including monoclonal antibodies targeting inflammatory cytokines (Figure 4); the reference item employed is a monoclonal antibody to the common p40 sub-unit of IL-12 and IL-23, the human equivalent, Ustekinumab, has recently been approved by the FDA for treatment of moderate/severe Crohn's disease. Other agents, such as sphingosine-1-phosphate (S1P) receptor modulator, FTY720 and anti-TNF therapies also demonstrate efficacy in the model; locally-acting steroids (budesonide) have some limited efficacy.