TL1A: A mediator of gut inflammation

In this issue of PNAS, a team of investigators led by Giorgos Bamias and Fabio Cominelli at the University of Virginia Health Science Center (Charlottesville, VA) in collaboration with the laboratory of Giorgos Kollias at the Alexander Fleming Biomedical Sciences Research Center (Athens, Greece) report on the role of TL1A (a TNF superfamily member) in two different models of chronic murine ileitis (1). These mouse models are considered to be good models for human Crohn’s disease, a severe inflammatory bowel disorder that strikes young adults (ages 20–30) and is thought to originate from predisposing genetic and environmental factors that cause an imbalance of effector (proinflammatory) and regulatory T cell responses, resulting in inflammation of the gastrointestinal mucosa and disease. Bamias et al. found that, in both models of murine ileitis described in ref. 1, TL1A and its receptor [death receptor 3 (DR3)] are up-regulated in the inflamed gut mucosa.

Death receptors and their ligands play a key role in the maintenance of tissue homeostasis and the physiological regulation of programmed cell death. Binding of a death ligand induces oligomerization of the receptor, recruitment of an adapter protein via a conserved cytoplasmic signaling element termed the death domain, activation of caspases, and induction of apoptosis. Although death receptors such as Fas/Apo-1/CD95, TNF-R1, TRAIL-R1, TRAIL-R2, or DR3 were initially characterized as inducers of apoptosis, there is growing evidence that these receptors also have nonapoptotic functions, including regulation of the adaptive immune response. Bamias et al. (1) report that TL1A is expressed by lamina propia dendritic cells and that it functions by increasing the proliferation of memory cells, but not naïve CD4⁺ T cells, and synergizes with IL-12 and/or low-dose stimulation of the T cell receptor to strongly enhance IFN-γ gene expression. IFN-γ expression in the gut has been considered a marker of inflammation, and many strategies for treating Crohn’s disease rely on broad attempts to suppress the immune-activated state. However, such approaches (steroid treatment and immunosuppressive drugs) do not focus on the gut specifically and thus have their own complications. Targeted therapies based on the use of antagonists of TNF-α were introduced with success in the 1990s, and the results reported in ref. 1 suggest that therapy directed specifically against TL1A or its receptor may provide an alternative targeted therapy for this debilitating disorder.

Although TL1A was first cloned in 2002 by a team at Human Genome Sciences (Rockville, MD) (2), the work on the involvement of TL1A in inflammatory gut disease has been has been carried out by the Cominelli laboratory and the laboratory of Stephan Targan (Cedars Sinai Medical Center, Los Angeles). Both laboratories have found an association of the expression of TL1A and inflammatory bowel disease (3, 4). TL1A has also been reported to be angiostatic and to induce metalloproteinase and IL-8 gene expression (5, 6). Indeed, TL1A and DR3 may be involved in the pathogenesis of atherosclerosis by increasing the production of proinflammatory cytokines and chemokines and decreasing plaque stability by inducing extracellular matrix-degrading enzymes (6). There is also evidence to suggest that TL1A/DR3 is involved in the etiology of rheumatoid arthritis (7).

However, as reported in ref. 1 and elsewhere by both the Cominelli and Targan laboratories, TL1A seems to have a most profound effect when expressed in the gut during inflammation. Both laboratories have shown that TL1A synergizes in the induction of IFN-γ expression in human T cells when combined with IL-12/18, although increased expression can also be observed in natural killer cells (2, 8, 9). The article by Bamias et al. (1) is the first report of a similar observation in mouse models of Crohn’s disease and extends earlier data by showing that the synergy occurs when the T cell receptor is weakly stimulated or T cells are treated with IL-12. Although in ref. 1 no synergy is observed when TL1A treatment is combined with IL-18, this result may not be surprising because both IL-18 and TL1A signal through NF-κB. Whereas the initial report (2) on TL1A demonstrated that it was a T cell costimulatory signal, Bamias et al. demonstrate that it is the memory T cell that most strongly responds, consistent with the increased capacity of this T cell population to express IFN-γ. Because this population does proliferate, it also expresses higher levels of the TL1A receptor, thus further enhancing the ability of the cells to proliferate and express IFN-γ. This finding might be considered somewhat surprising given that the only known receptor of TL1A is DR3, a death domain-containing receptor, and it might have been hypothesized that triggering this receptor would lead to cell death. However NF-κB-dependent antiapoptotic genes, such as inhibitor of apoptosis 2, have been shown to be induced by TL1A (10), and therefore triggering of apoptosis vs. proliferation may be cell-type dependent.

What is particularly interesting about TL1A is that it appears to represent one of the earliest members of the TNF superfamily. TNF itself has not been found in chickens, but recently the cloning of TL1A from chickens has been reported (11). The finding implies that TL1A is part of the early immune system and raises the question of what are the endogenous triggers of TL1A expression. Although it has been reported that IL-1 and TNF can induce TL1A expression (12), one might expect that activation of toll receptors, either directly or through the nucleotide-binding oligomerization domain (NOD) pathway, will trigger TL1A expression. TL1A expression is tightly regulated under physiological conditions, and it will be of importance to understand the ways by which TL1A expression is triggered and whether differences in TL1A expression between individuals can account for increased susceptibility to and development of Crohn’s disease. The authors here have shown that transmembrane DR3 is up-regulated during chronic ileitis, through alternative splicing of DR3 mRNA, in preference to the soluble form of the receptor. This observation is in contrast to the majority of the members of the TNF-R superfamily, where soluble receptors are generated by proteolytic cleavage. DR3 expression is also regulated by methylation of specific CpG islands in its promoter and by the silencer of death domains protein that binds to the cytoplasmic portion of DR3 to inhibit signaling in the absence of ligand. The DR3 decoy receptor has been reported to be up-regulated by LPS in human cell lines, and the inflamed gut also shows increased decoy receptor expression (13), indicating that there is a homeostatic regulation of the effects of TL1A. Why this decoy fails to control the excessive inflammation observed during disease progression remains a subject of exploration. It will also be of interest to define the downstream signaling pathway used in activated T cells after binding of TL1A to functional DR3 receptors and whether DR3 activates TRAF2 and NF-κB via the TAK1/ASK1-MKK4/MKK7-JNK2 kinase cascade as previously reported for DR3-mediated IL-8 gene expression (5).

Since its approval in 1998, infliximab (Remicade; Centocor, Inc., Horsham, PA), a chimeric monoclonal antibody against TNF-α, has revolutionized the treatment of Crohn’s disease. Although generally well tolerated, infliximab can cause recrudescence of tuberculosis infection and worsening of heart failure, demyelinating disease, and an increased incidence of lymphoma. Other targeted approaches, including inhibition of IL-12, are being considered and have shown clinical promise (14), and, based on ref. 1 and other previously published work, TL1A should now be added to the list of promising candidates for therapy (Table 1). Many approaches appear promising, including the use of a soluble DR3 receptor or blocking antibodies to DR3, soluble DcR3 receptor, or neutralizing monoclonal antibodies specific for TL1A. The outcome of trials using these reagents could provide new hope to patients stricken with this debilitating disease.

Table 1.

Treatments for Crohn’s disease

Current treatments	Specificity	Future treatments	Specificity
Antiinflammatories (e.g., sulfasalazine)	Low	Anti-TL1A	High
Cortisone or steroids (e.g., prednisone)	Low	Anti-IL-12 (14)	High
Immune system suppressors (e.g., 6-mercaptopurine)	Low	Anti-α4 integrin (15, 16) (e.g., natalizumab)	Potentially high (JC virus reactivation may occur)
Antibiotics	Very low	Histone deacetylase inhibitors (HDAC) (17)	Low
Surgery	Very low
Anti-TNF(e.g., infliximab)	High (not approved as a first-line therapy)	Lactococcus lactis expressing IL-10 (18)	High