点击显示 收起
1Department of Pathology and 2Department of Medicine, The University of Chicago, Chicago, Illinois, USA. 3Department of Nephrology, Hannover Medical School, Hannover, Germany.
Address correspondence to: Jerrold Turner, The University of Chicago, 5841 South Maryland, MC 1089, Chicago, Illinois 60637, USA. Phone: (773) 702-2433; Fax: (773) 834-5251; E-mail: jturner@bsd.uchicago.edu.
We read with interest the commentary (1) accompanying our article (2) in the October issue of the JCI. We appreciate Michael Field’s opinion that based on our data, the concept of secretory diarrhea must be revised to include the idea that net water secretion, or diarrhea, can occur in the absence of active ion secretion. However, we disagree with his conclusion that TNF must affect capillary permeability or smooth muscle contractility to generate increased interstitial pressure and a hydraulic driving force for water secretion.
In contrast to Field’s suggestion, our study of 2 TNF superfamily members, TNF and LIGHT (lymphotoxin-like inducible protein that competes with glycoprotein D for herpesvirus entry mediator on T cells), demonstrates that water secretion occurs as a consequence of epithelial Na+ malabsorption and increased epithelial paracellular permeability. This is supported by the observation that although TNF and LIGHT cause quantitatively similar paracellular permeability increases, only TNF causes diarrhea. The key difference that accounts for this is that only TNF activates PKC, thereby inhibiting Na+/H+ exchanger isoform 3 (NHE3); TNF did not induce diarrhea when PKC-mediated NHE3 inhibition was prevented by either pharmacological inhibitors or PKC knockout. Could these approaches to PKC inhibition have also prevented PKC-dependent changes in capillary permeability or smooth muscle contractility? To test this we took advantage of the observation that LIGHT does not activate PKC. LIGHT did cause diarrhea when PKC was activated pharmacologically, confirming that LIGHT was fully able to recapitulate TNF-like diarrhea when coupled with PKC activation. More importantly, direct NHE3 inhibition, either pharmacological or genetic, showed that this was the critical second component necessary for LIGHT to cause diarrhea. This was not due to inhibition of endothelial or smooth muscle NHE3, since neither tissue expresses NHE3. The unmistakable conclusion is that the necessary role of PKC in TNF-induced diarrhea is to inhibit epithelial NHE3 and not to effect changes in capillary permeability or smooth muscle contractility.
Why, then, is NHE3 so critical? Is it simply that Na+ absorption provides an osmotic driving force for water absorption? This alone is an inadequate explanation, as inhibition of absorption is not necessarily equivalent to induction of secretion. We propose that loss of NHE3-mediated Na+ absorption results in dissipation of the normal hyperosmolarity of the upper villus that drives water absorption (3). It then follows that increased paracellular permeability allows Na+ to leak from tissue to lumen; without NHE3, Na+ cannot be reabsorbed. Increased epithelial paracellular permeability may also allow water and other solutes to flow into the lumen, thereby compounding loss of the villous osmotic gradient that drives water absorption. In terms of absorption, the model gains further support from the observation that LIGHT-induced increases in permeability enhance water absorption (when NHE3 is active). Thus, this model fully explains the data without the need to invoke increased interstitial pressure to force water through paracellular or transcellular water channels.
Is it therefore impossible for elevated interstitial pressure, secondary to increased capillary permeability or smooth muscle contraction, to contribute to TNF-induced diarrhea? It is not. However, if elevated venous or interstitial pressure were able to cause diarrhea by the mechanism proposed by Yablonski and Lifson (4) that is cited by Field (1), it is surprising that diarrhea is not a common clinical characteristic of patients with portal hypertension and that jejunal water and electrolyte transport are normal in these patients (5).
We therefore conclude that TNF causes acute diarrhea via myosin light chain kinase–dependent increases in epithelial paracellular permeability coupled with PKC-dependent NHE3 inhibition. Although intriguing as a hypothesis, no data support the role of increased interstitial pressure physically squeezing water into the lumen as proposed by Field. However, existing models of epithelial transport do provide an ample explanation for the observed synergy between increased epithelial paracellular permeability and Na+ malabsorption.
References
Field M. 2006. T cell activation alters intestinal structure and function. J. Clin. Invest. 116:2580–2582 doi: 10.1172/JCI29985.
Clayburgh D.R., Musch M.W., Leitges M., Fu Y.-X., Turner J.R. 2006. Coordinated epithelial NHE3 inhibition and barrier dysfunction are required for TNF-mediated diarrhea in vivo. J. Clin. Invest. 116:2682–2694 doi: 10.1172/JCI29218.
Hallback D.A., Jodal M., Mannischeff M., Lundgren O. 1991. Tissue osmolality in intestinal villi of four mammals in vivo and in vitro. Acta Physiol. Scand. 143:271–277.
Yablonski M.E. and Lifson N. 1976. Mechanism of production of intestinal secretion by elevated venous pressure. J. Clin. Invest. 57:904–915.
Norman D.A., Atkins J.M., Seelig L.L., Gomez-Sanchez C., Krejs G.J. 1980. Water and electrolyte movement and mucosal morphology in the jejunum of patients with portal hypertension. Gastroenterology. 79:707–715.