Feeding intravenously saves the patient but starves the gut: consequences for the intestinal barrier

During the late 1960s, the medical breakthrough of providing a patient's complete daily nutritional requirement via intravenous catheter was successfully developed (Dudrick, 2009). When coupled with a growing understanding of the role of optimal nutrition in the recovery of patients unable to consume food, total parenteral nutrition (TPN) has saved numerous lives. However, as with any advanced medical therapy, TPN comes with a number of important complications, many of which are life-threatening. One specific consequence of TPN is gut mucosal atrophy, in which the intestinal mucosa is reduced in thickness because of the importance of luminal nutrients on mucosal growth. Furthermore, there is an associated breakdown in intestinal barrier function, which is critical because the mucosal barrier is the most important line of defence against microorganisms within the intestinal lumen. Bacterial translocation can in turn trigger sepsis, an important cause of mortality in gravely ill patients. However, the mechanisms of intestinal barrier breakdown in patients on TPN are poorly understood. The intestinal mucosal barrier is composed of a single layer of columnar epithelial cells attached side-by-side by a series of interepithelial junctions, including the apical-most tight junctions. Over the last 10–15 years, tight junctions have been the subject of intensive study because although they are ‘tight’ they are also the weakest link in the barrier. Tight junctions are composed of interdigitating extracytoplasmic loops of junctional proteins, including occludin, claudins and junction adhesion molecules. Loss of barrier function at the level of the tight junctions is thought to be a pivotal event in the pathophysiology of a number of important disease syndromes, including infectious diarrhoeal disease, ischaemia/reperfusion injury and inflammatory bowel disease (Hering et al. 2012). However, the mechanisms whereby intestinal barrier function is pathologically altered in intestinal disease are not well defined. In this issue of The Journal of Physiology, Feng & Teitelbaum (2013) have built on their previous findings of barrier disruption in experimental animals on TPN by showing that a complex interplay of tumour necrosis factor (TNFα) and its two receptors (TNFR1 and TNFR2) is responsible for the breakdown of intestinal epithelial tight junctions. Experiments using mutant mice in which TNFR1, TNFR2, or both receptors were knocked out revealed that the absence of either TNF receptor partially prevented the effects of TPN on tight junctions. Additionally, genetic deletion of both TNF receptors prevented any notable effect of TPN on intestinal barrier function. This is interesting not only to researchers studying the mechanisms of TPN-induced barrier disruption, but also to others who have focused on mechanisms whereby cytokines such as TNFα disrupt epithelial barrier function. The TPN model is particularly interesting because it involves selective loss of tight junction integrity independent of inflammatory cell infiltration. In this way, TPN models may allow more precise studies assessing the effect of select factors on tight junction integrity in vivo which are complicated in many models by inflammation. From cell studies, it is well known that cytokines, including TFNα, interferon-γ and interleukin-1-β, can disrupt epithelial barrier function. For example, initial studies assessing the role of cytokines in cell culture models revealed that TNFα dose-dependently reduced epithelial barrier function associated with morphological disruption of tight junctions using electron microscopy (Schmitz et al. 1999). With advancing technology, studies of specific tight junction proteins have shown that TNFα induces endocytosis of occludin as a consequence of activation of the cytoskeletal-associated enzyme myosin light chain kinase (MLCK). Interestingly, Feng et al. noted increased phosphorylation of myosin light chain by MLCK in their TPN mouse model. MLCK has a physiological role in ‘opening’ tight junctions in response to sodium-coupled glucose intestinal epithelial transport, but has more recently been closely linked to pathological states involving disruption of tight junctions. TNFα has also been shown to have alternate mechanisms of barrier disruption. For example, recent studies have shown that TNFα disrupts the highly controlled process of cell shedding of the most superficial surface epithelium resulting in excessive cell loss that can be seen in patients with inflammatory bowel disease (Watson et al. 2012). Such cell loss is also associated with perturbation of tight junctions, allowing cell detachment. Which of the actions of TNFα is critical to loss of barrier function in TPN-treated mice is not entirely clear, although the tight junctions and MLCK are certainly involved. As far as the translational relevance of TNFα in disruption of the gut barrier is concerned, Feng & Teitelbaum (2013) have additionally shown that the TNFα inhibitor etanercept prevents loss of barrier function in TPN-infused mice. Etanercept is a TNFR fusion protein that was originally studied for the treatment of shock (Fisher et al. 1996) and has been marketed for the treatment of select autoimmune diseases. Considering the fact that placement of a patient on TPN is elective, there may be the opportunity to intervene prior to onset of intestinal barrier disruption. None.