Diarrhea associated with inflammatory bowel diseases has traditionally been attributed to stimulated secretion. The purpose of this study was to determine whether chronic stimulation of intestinal mucosa by interferon-γ (IFN-γ) affects expression and function of the apical membrane Na+/H+ exchangers NHE2 and NHE3 in rat intestine and Caco-2/bbe (C2) cells. Confluent C2 cells expressing NHE2 and NHE3 were treated with IFN-γ for 2, 24, and 48 h. Adult rats were injected with IFN-γ intraperitoneally for 12 and 48 h. NHE2 and NHE3 activities were measured by unidirectional 22Na influx across C2 cells and in rat brush-border membrane vesicles. NHE protein and mRNA were assessed by Western and Northern blotting. IFN-γ treatment of C2 monolayers caused a >50% reduction in NHE2 and NHE3 activities and protein expression. In rats, region-specific, time- and dose-dependent reductions of NHE2 and NHE3 activities, protein expression, and mRNA were observed after exposure to IFN-γ. Chronic exposure of intestinal epithelial cells to IFN-γ results in selective downregulation of NHE2 and NHE3 expression and activity, a potential cause of inflammation-associated diarrhea.
- inflammatory bowel disease
- sodium transport
- sodium absorption
- water and electrolyte transport
- intestinal adaptation
two major functions of the intestinal epithelium are to provide a selective barrier to luminal contents and to transport water, nutrients, and electrolytes in a vectorial manner (6). In chronic inflammatory bowel diseases (IBDs), aberrations in barrier and transport functions result in malabsorption and diarrhea, with the latter hypothesized to arise from active anion secretion stimulated by the actions of numerous immune and inflammatory mediators (20). This hypothesis is based on numerous experimental observations that these agents can stimulate active anion secretion and assumes that the intestinal epithelium of chronically inflamed mucosa is operationally intact. However, recent clinical observations and in vivo studies do not support this notion and, in fact, suggest that the chronically inflamed intestinal epithelium may have not only defective absorption but also impaired secretion and diminished barrier function (3, 11-13, 23).
Apical membrane Na+/H+ exchange of intestinal epithelial cells is the major route for electroneutral, non-nutrient-dependent Na+ absorption (17). In contrast to the ubiquitously expressed basolateral membrane Na+/H+ exchanger (NHE1) (5), the brush-border membrane Na+/H+ exchangers NHE2 and NHE3 (5, 14, 29) appear to be active even under basal conditions in the intestine, where they serve an important purpose acting as transport pathways whenever luminal Na+ is present (17, 18).
In this study, we examined the effect of chronic inflammation as mediated by interferon-γ (IFN-γ), a well-characterized cytokine produced by Th1 lymphocytes and present in high quantities in the gut of patients with IBD that has many effects on intestinal epithelial cells (1, 4, 9, 16, 31) and on Na+ absorption by NHE2 and NHE3 in both an in vitro cell culture system and physiologically relevant brush-border membranes of rat intestine. In both cases, there was a time- and dose-dependent downregulation of NHE2 and NHE3 activity, protein expression, and mRNA expression. These findings support the hypothesis of impaired or downregulated intestinal epithelial function associated with chronic inflammatory states. They further implicate IFN-γ as an important mediator of these effects and in the pathogenesis of inflammation-associated diarrhea.
MATERIALS AND METHODS
Caco-2/bbe (C2) cells generously provided by Dr. Mark Mooseker (Yale University, New Haven, CT) were grown as confluent monolayers on Transwells in DMEM supplemented with 10% fetal bovine serum (FBS), 2 mM l-glutamine, 10 μg/ml transferrin (GIBCO, Grand Island, NY), 50 U/ml penicillin, and 50 μg/ml streptomycin in a humidified atmosphere of air with 5% CO2.
Apical membrane unidirectional 22Na influx as a measure of NHE activity.
C2 cells were grown on Transwells for 14 days before experiments began and were treated with DMEM supplemented with glutamine, penicillin, streptomycin, and transferrin as described in Cell culture; however, the FBS was increased to 30% (vol/vol) for 4 days to induce apical membrane NHE2 and NHE3 activity. Two days before flux measurements were taken, cells were treated with 3 or 30 ng/ml of IFN-γ for 2, 24, and 48 h. The lots of IFN-γ used in the present studies were all 107 U/mg and were obtained from Endogen (Woburn, MA). Unidirectional apical membrane Na+ uptake was determined in flux buffer (130 mM choline chloride, 5 mM KCl, 1 mM MgCl2, 2 mM CaCl2, 15 mM HEPES, pH 7.4, and 20 mM NaCl, with 1 μCi/ml 22NaCl) for 10 min. Na+ influx was stopped by four washes in cold buffer (140 mM NaCl, 5 mM KCl, 15 mM HEPES, pH 7.4, and 1 mM Na3PO4) and was calculated by dividing the accumulated disintegrations per minute by the specific Na+activity in the medium. Dimethylamiloride (DMA; 500 μM) and HOE-642 (30 μM) were used to distinguish NHE2 and NHE3 activities; the former was defined as the HOE-642-sensitive, and the latter as the HOE-642-insensitive, components of the DMA-inhibitable unidirectional22Na influx. All fluxes were measured under acid-loaded conditions as previously described (18). Briefly, cells were incubated at 37°C for 60 min in acidifying saline (in mM: 50 NH4Cl, 70 choline chloride, 5 KCl, 2 CaCl2, 1 MgCl2, 5 glucose, and 15 HEPES, pH 7.0). Fluxes in experimental groups were expressed as a percentage of fluxes in control wells with each dose and time point having its own control value.
For the animal experiments, adult Sprague-Dawley rats (200–255 g) were injected with 10,000 or 25,000 units (U) of IFN-γ IP for 12 and 48 h before death, while controls received saline injections of equal volume for the same time points. The activity of NHE2 and NHE3 in the brush-border membranes was measured as previously described (7, 8). Briefly, ileal and colonic mucosal scrapings were weighed and added to 30 ml of hypotonic lysis buffer (10 mM Tris, pH 7.4, and 3 mM EDTA, with protease inhibitors as described previously) and homogenized for 30 s at a speed of 15,000 rpm in an Ultra-Turrax homogenizer. Samples were taken for enzyme enrichment studies, and the samples were spun at 2,000 g for 5 min at 4°C to remove nuclei and unbroken cells. The supernatants were removed and spun at 10,000 g for 10 min at 4°C to remove mitochondria. The supernatants were removed, and 15 mM CaCl2 was added. Samples were gently stirred in the cold for 15 min and then spun at 8,000 g for 8 min to remove the endoplasmic reticulum, Golgi, and basolateral membranes. The supernatants were spun at 45,000 g for 45 min at 4°C to obtain brush-border membranes. The membranes were resuspended in a small volume of intravesicular transport buffer (10 mM MES, pH 6.1, 3 mM EDTA, and 80 mM mannitol) and resuspended using a Teflon pestle homogenizer. A sample was removed for protein determination and enzyme enrichment studies. Five microliters of the vesicles were added to forty-five microliters of extravesicular transport buffer (10 mM HEPES, pH 7.4, 1 mM Na, with 1 μCi/ml 22Na giving a specific activity of 2,200 dpm/nmol, and 80 mM mannitol). All brush-border uptakes were of 10-s duration. Uptake of Na+ into the vesicles was stopped by the addition of 2 ml of ice-cold 90 mM mannitol and immediate placement onto a 0.45-μm cellulose filter (HAWP; Millipore, Milford, MA). The filter was washed once with 4 ml of ice-cold 90 mM mannitol, and the filter was removed and solubilized in liquid scintillation fluid. 22Na was determined by liquid scintillation spectroscopy. For the present experiments, the22Na uptakes were always performed with both HOE-642 (30 μM) and DMA (500 μM) so that NHE2 and NHE3 activities could be distinguished. With 1 mM Na, NHE2 is completely inhibited by the amiloride analog HOE-642 at 30 μM, whereas NHE3 is inhibited <5% (18). Both exchangers are sensitive to DMA. Fluxes in experimental rats were expressed as a percentage of the flux in untreated control rats.
C2 cells grown on Transwell membranes were harvested by scraping in PBS, washed with PBS, lysed with hypotonic buffer (10 mM Tris, 5 mM MgSO4, 5 U/ml RNase, 50 U/ml DNase, 10 μg/ml leupeptin, 10 μg/ml aprotinin, and 1 mM phenylmethylsulfonyl fluoride), and allowed to sit on ice for 15 min. An aliquot was removed for protein determination, and the remaining cell protein was then solubilized in Laemmli stop buffer. Proteins were separated by 7.5% SDS-PAGE and transferred to the polyvinylidene difluoride membranes. After blocking with Tris-buffered saline (TBS)-0.1% Nonidet P-40 (NP-40)-5% milk, blots were incubated with primary antibody diluted in TBS-1% BSA overnight at 4°C. Blots were incubated with specific polyclonal antisera developed and characterized in our laboratory to NHE2 and NHE3 (5, 18). After three washes with TBS-0.1% NP-40-5% milk and one wash with TBS-0.1% NP-40-1% milk, blots were incubated with horseradish peroxidase-conjugated secondary antibody diluted in TBS-0.1% NP-40-1% milk for 1 h at 25°C. After three washes with TBS-0.1% NP-40-1% milk and one wash with TBS-0.1% NP-40, the membrane was developed by using an enhanced chemiluminescence system. For rat intestinal scrapings, NHE2 and NHE3 protein levels were analyzed by taking an aliquot of the brush-border membranes used for flux studies. The brush-border proteins were solubilized in Laemmli stop solution and resolved on a 7.5% SDS-PAGE, and Western blots were performed as described above.
For RNA isolation, intestinal scrapings were homogenized in Trizol. RNA was then extracted once with acid phenol-chloroform, reprecipitated, and quantified by absorbance at 260 nm. Twenty micrograms were size-separated on a denaturing formaldehyde agarose gel, transferred to a positively charged nylon membrane by capillary action, and RNA-linked to the membrane by ultraviolet light. Blots were prehybridized and hybridized in XOTCH solution as previously described (5) with the use of the cDNA probes for rat NHE2 and NHE3. Glyceraldehyde phosphate dehydrogenase was used as a constitutive control. Probes were labeled with [32P]dCTP by random prime labeling. Blots were hybridized at 55°C overnight and then washed up to a stringency of 0.5× saline sodium citrate-0.5% SDS at 55°C.
Effect of IFN-γ on differentiation.
To determine whether IFN-γ had effects on the differentiation of the C2 cells or the rat intestine, we measured activities of the brush-border enzymes sucrase and alkaline phosphatase. Additionally, levels of the microvillus structural protein villin were determined by Western blotting.
For the C2 cells, cells were harvested from Transwells 14 days postplating and, when appropriate, treated with IFN-γ. Cells were scraped from the filter in PBS and lysed in 10 volumes of hypotonic buffer (10 mM Tris and 3 mM EDTA with protease inhibitors as described previously). Nuclei, unbroken cells, and mitochondria were removed by centrifugation (10,000 g for 5 min), and the microsomal membrane fraction was obtained by centrifugation (100,000 gfor 20 min). The amount of protein in this membrane fraction was measured with the bicinchoninic acid procedure, and alkaline phosphatase and sucrase activities were immediately measured. Alkaline phosphatase was measured colorimetrically by usingpara-nitrophenol phosphate and measuring absorbance at 490 nm as described by Cox and Griffin (10). Sucrase activities were measured as described in the microassay procedure of Messer and Dahlquist (19) by measuring the glucose generated by sucrase activity. Glucose oxidase andortho-dianisidine were used to generate a colored end product, and the absorbance was measured at 450 nm. To determine the effect of IFN-γ on differentiation in the rat intestine, we measured alkaline phosphatase activities in brush borders of both the ileum and colon that had been isolated for use in Na+ flux studies.
As another marker of differentiation, the C2 cells as well as the brush-border membranes from rat ileum and colon were analyzed for villin by Western blotting. The protocol used was essentially the same as that used for NHE Western blots, except that a monoclonal anti-villin antibody (Transduction Labs, Lexington, KY) was used.
Effect of IFN-γ on C2 cells.
The time- and dose-dependent effects of IFN-γ on apical membrane NHE2 and NHE3 expression and function were studied in human colonic C2 cell monolayers. NHE2 and NHE3 are the two major NHE isoforms responsible for non-nutrient-dependent Na+ absorption across the brush-border membrane of enterocytes. C2 cell monolayers were selected for these studies because they form tight, polarized monolayers, display phenotypic properties characteristic of mature absorptive epithelium, and correctly sort NHE2 and NHE3 to the apical membrane, where they are found in intestinal enterocytes (5, 14,29).
As shown on Fig. 1, cell monolayers treated with 30 ng/ml IFN-γ for 2 h exhibited little change in NHE2 activity (88.6 ± 3.8% compared with controls). However, as time progressed, IFN-γ treatment resulted in a progressive decrease in NHE2 activity as shown by a reduction to 53.1 ± 6.0% of control values observed after 24 h. A further decrease of NHE2 function to 44.6 ± 6.4% of control values was observed by 48 h of treatment. At a lower dose of IFN-γ (3 ng/ml), similar, but more slowly developing, effects on NHE2 activity were observed (Fig. 2). NHE2 activity at 24 h was 112 ± 11.3% compared with controls; however, by 48 h, NHE2 activity decreased to 50.4 ± 8.1% of control values (Fig. 2).
Time- and dose-dependent effects with IFN-γ treatment on apical membrane NHE3 activity also were observed. As shown in Fig. 1, after 2 h of exposure of monolayers to 30 ng/ml IFN-γ, little change in apical membrane NHE3 activity was observed (97.4 ± 2.6% of control values). However, by 24 h, NHE3 activity was significantly reduced to 34.1 ± 4.9% of control values with a further decrease to 18.0 ± 2.0% of control values at 48 h. As shown in Fig.2, lower doses of IFN-γ (3 ng/ml) had no effect on NHE3 activity at 24 h (96.6 ± 1.8% compared with control), but, by 48 h, a decrease in NHE3 activity to 39.7 ± 2.4% of control values was observed.
To determine whether the IFN-γ effects on apical membrane NHE activity were secondary to altered protein expression, we performed Western blot analysis of NHE2 and NHE3 with isoform-specific polyclonal antibody. As shown by the representative immunoblots in Figs.3 and 4, the effects of both low (3 ng/ml)- and high (30 ng/ml)-dose IFN-γ treatment of C2 cells for periods of 24, 48, and 72 h can be observed. At 24 h, there is no change in the expression of NHE2 or NHE3 at either dose of IFN-γ. By 48 h, treatment of C2 monolayers with 30 ng/ml IFN-γ caused a significant decrease in both NHE2 and NHE3 protein expression, whereas 3 ng/ml IFN-γ had little effect. At 72 h, both doses of IFN-γ showed dramatic and nearly equivalent decreases in NHE2 and NHE3 protein expression.
In vivo effects of IFN-γ on rat intestinal apical membrane NHE activity and expression.
To determine whether the in vivo effects of IFN-γ reflected changes observed in vitro with C2 cells, we measured the regional activity and expression of rat intestinal NHE2 and NHE3 in control and IFN-γ-treated rats. As shown in Fig.5, NHE2 activity of brush-border membrane vesicle decreased to 83.5 ± 24.7% of control values in the jejunum in response to treatment with 25,000 U of IFN-γ IP for 12 h, although this change was not statistically significant. However, a significant inhibition of NHE2 activity to 34.2 ± 6.6% of control values was observed in the jejunum of IFN-γ-treated rats by 48 h. IFN-γ had a similar effect on ileal and colonic NHE2 activity: there was little effect at 12 h, but at 48 h there was a dramatic decrease in NHE2 activity. This effect was dose dependent as well, since treatment with 10,000 U of IFN-γ at 48 h produced a smaller effect compared with 25,000 U of IFN-γ at the same time point. After 48 h of treatment, for instance, this dose of IFN-γ reduced jejunal, ileal, and colonic NHE2 activities to 56.5 ± 11.5%, 62.0 ± 11.5%, and 53.2 ± 10.2% of control values, respectively (Fig. 6).
As shown in Figs. 5 and 6, IFN-γ treatment of rats also caused a significant decrease in intestinal brush-border membrane NHE3 activity. Jejunal and ileal NHE3 activity were reduced to a statistically significant level of 74.9 ± 4.9% and 74.1 ± 10.8%, respectively, after 12 h of IFN-γ (25,000 U) treatment. However, by 48 h, NHE3 activity of jejunal and ileal brush-border membranes was further reduced to 34.1 ± 6.5% and 35.3 + 5.5%, respectively. On the other hand, colonic NHE3 activity was profoundly decreased by IFN-γ treatment to 48.1 ± 11.3% at 12 h with no further decrease after 48 h (51.8 ± 5.8%) (Fig. 5) Similar to NHE2, a dose-dependent response to IFN-γ treatment was observed (Fig. 6). With 10,000 U of IFN-γ, NHE3 activity was significantly reduced to 49.1 ± 8.6% of control values in the jejunum and 51.5% ± 7.8% of control values in the ileum, a lesser inhibition than was observed with 25,000 units of IFN-γ. However, colonic mucosal NHE3 activity was reduced to 53.2% ± 11.4% of control levels, a response that approximated the changes observed induced by the higher dose of 25,000 U of IFN-γ.
Western blot analyses were also performed to assess changes in ileal and colonic NHE2 and NHE3 protein expression. As shown in Fig.7, 25,000 U of IFN-γ significantly decreased NHE2 protein expression in both ileal and colonic mucosa after 48 h of treatment (32.6 ± 7.7% and 35.6 ± 9.6% of control densitometric units, respectively, corresponding to changes observed in the functional activity measurements). NHE3 protein expression in the rat intestine was affected in a similar fashion by exposure to 25,000 U of IFN-γ for 48 h, decreasing to 36.3 ± 9.5% and 27.1 ± 10.6% of control densitometric units in the ileum and colon, respectively (Fig.8).
Northern blot analyses were also performed to assess whether there were changes in mucosal NHE2 and NHE3 mRNA abundance, as shown in Figs.9 and 10. NHE3 mRNA was significantly reduced after 48-h treatment with 25,000 U of IFN-γ in both the ileum and the colon (28.4 ± 2.7% and 44.1 ± 14.9% of control densitometric units, respectively), consistent with changes observed in protein expression and functional activity. In contrast, changes in NHE2 mRNA abundance in ileum and colon induced after treatment for 48 h with 25,000 U of IFN-γ (19.1 ± 4.6% and 65.9 ± 7.8%, respectively) did not correlate in magnitude to the changes observed in protein expression and functional activity. These data may indicate differences in IFN-γ regulation of NHE2 expression in different segments of bowel.
Effect of IFN-γ on differentiation.
To determine whether the effect of IFN-γ was specific, we measured activities of the brush-border enzymes sucrase and alkaline phosphatase. Additionally, levels of the microvillus protein villin were analyzed by Western blots. IFN-γ did not affect the activities of sucrase or alkaline phosphatase, or the expression of villin, in either the C2 cells or the rat ileum or colon (Fig.11). It should be noted that the C2 cells used for these studies were plated onto collagen-coated Transwells at confluence. This promotes a rapid differentiation, and thus, over the 4-day course of treatment with media with 30% serum, no changes in sucrase, alkaline phosphatase, or villin were generally observed.
IBDs are characterized by intestinal mucosal destruction and functional impairment caused by the chronic effects of immune and inflammatory mediators, with the clinical consequences often being the development of severe diarrhea and malabsorption. For many years, the etiology of inflammation-associated chronic diarrhea was attributed to stimulated anion secretion, a notion based on experimental observations demonstrating the acute prosecretory effects of various immune and inflammatory mediators on normal intestinal tissues (2, 6,20). However, these studies failed to take into account the fact that IBDs are chronic diseases and that the effects of these agents might have different consequences over time. Sandle et al. (23), for instance, demonstrated that in colonic mucosa from IBD patients, Na+ absorptive capacity was diminished. The Cl− exchanger protein DRA is also decreased in areas of intestinal inflammation (30). In animal models of chronic enteritis, Na+-dependent glucose absorption (26), anion exchange, and Na+/Cl−transport were impaired (27), albeit potential causative agents for these effects were not identified. Furthermore, Sundaram and West (27) reported decreased intestinal mucosal Cl−/HCO exchange, but no change in NHE activity, in a model of coccidiococcus-induced enteritis. Finally, studies of trinitrobenzene sulfonic acid-induced chronic colitis showed that while basal electrolyte transport was normal, the mucosa was less responsive to secretagogues (3).
In this study, we have shown that IFN-γ that is present at high levels in IBD tissues can significantly downregulate intestinal epithelial NHE2 and NHE3 protein and mRNA expression in C2 cells and in rat intestinal brush-border membranes and that the effects are both time and dose dependent. Although we believe that these changes may be due to IFN-γ-induced decreases in NHE gene transcription, the possibility that posttranscriptional mechanisms have a role in these effects cannot be ruled out. This may be particularly true for NHE2 where disproportionate decreases in protein and mRNA expression in certain regions of the bowel were observed. This IFN-γ effect appears to be relatively specific, since concomitant changes in the expression of the epithelium-specific protein, villin, and the specific activities of brush-border hydrolases were not observed. Additionally, IFN-γ induces the expression of class II myosin heavy chain, suggesting a phenotypic switch rather than a downregulation of all functions (1, 9, 16, 22, 31). Furthermore, IFN-γ at the doses used in this study has no effects on mucosal histology in vivo (data not shown) and has no effect on sucrase, alkaline phosphatase, or villin in either the C2 cells or rat intestine.
Considerable differences in NHE2 and NHE3 functional regulation have been noted, both acutely and chronically. With respect to the latter, NHE3 appears to be highly sensitive to a variety of systemic and luminal stimuli (7, 8, 21, 28), whereas NHE2 expression appears to be relatively stable under physiological situations. The finding that IFN-γ affects the expression of both NHE2 and NHE3 was therefore surprising but could play a major role in causing many of the mucosal transport alterations associated with chronic inflammation. IFN-γ is known to chronically downregulate active anion secretion and barrier function in intestinal epithelial and T84 cells (1, 4, 9,16, 25, 31). Although one group has reported a decrease in cystic fibrosis transmembrane conductance regulator (CFTR) expression and no change in Na+-K+-ATPase activity (4), others have observed the opposite (9,16), i.e., downregulated Na+-K+-ATPase activity is decreased and CFTR expression is unchanged. IFN-γ-stimulated decreases in active anion secretion were thus attributed to a diminished Na+ gradient and internalization of apical membrane CFTR. Diminished anion secretion, however, would not explain the development of inflammation-associated diarrhea. The importance of this study therefore lies in its examination of the chronic effects of IFN-γ on apical membrane NHEs, the major routes for non-nutrient-dependent, electroneutral Na+ absorption. The contributions of NHE2 and NHE3 to intestinal Na+absorption have been determined in only a limited number of studies. Recent studies in our laboratory (21) using in vitro measurements of Na+ absorption in Ussing chambers have demonstrated that 25% of the Na+ absorption is mediated by NHE2 and 45% by NHE3. In in vivo studies of intestinal perfusions in canine small intestine, Maher et al. (17) determined that NHE3 mediates basal Na+ absorption, while NHE2 plays little role. The differences may relate to differing species, techniques, or segments of the intestine. It should be emphasized that apical Na+/H+ exchange is an important route for Na+ absorption, and the downregulation by IFN-γ would be anticipated to diminish the Na+ absorbed.
One mechanism of action for IFN-γ may be downregulation of transcription of NHE2 and NHE3 genes. The mRNA decreased for both the exchangers, which would occur if transcription had decreased or if the mRNA was being degraded at a quicker rate for the same rate of transcription. IFN-γ may be a potent regulator of the activity of a variety of transcription factors. IFN-γ may directly act through second messenger pathways to regulate activity of certain transcription factors. However, other factors may play a role in the transcriptional effects of IFN-γ. Our laboratory has recently shown that chronic IFN-γ stimulation of human intestinal C2 cells causes a dose- and time-dependent downregulation of activity and expression of several transport- and barrier-related proteins including the Na+-K+-ATPase (25). Further examination of this phenomenon suggests that the concomitant increase in intracellular sodium was responsible for the downregulatory effects, since treatment with ouabain, a specific inhibitor of the Na+-K+-ATPase, and Na+ ionophores reproduced the IFN-γ effect, while measures that lowered Na+ blunted it. Increases in Na+ have been implicated as an activating factor for several transcription factors such as p38 and c-Jun NH2-terminal kinase (15). Inhibition of Na+/H+exchange leading to increased Na+ also has been shown to activate stress protein kinases in U-937 cells (24). Therefore, it is possible that the mechanism of IFN-γ downregulation of NHE expression and function is mediated by this cell signaling pathway.
What purpose would diminished intestinal epithelial transport and barrier serve in a state of chronic inflammation? We speculate three potentially mutually inclusive possibilities. The first possibility involves the initiation of watery diarrhea expulsing pathogens and noxious agents from the intestinal lumen and delivering them to the lumen antimicrobial agents and antibodies. The second possibility is that these changes are part of a phenotypic shift by enterocytes preparing them for a critical role in wound healing and host defense. Finally, we speculate that enterocytes specifically downregulate non-life essential functions to reduce metabolic requirements or to shift them to processes necessary for phenotypic shift. The maintenance of active electrolyte and nutrient transport and a normal Na+ gradient represents a major metabolic demand on most cells, particularly under conditions of stress. Therefore, their downregulation during conditions of sustained inflammation can conserve sufficient energy for survival.
In summary, this study implicates IFN-γ as an important cytokine that causes selective downregulation of intestinal Na+absorption, in part by decreasing expression and activity of the apical membrane Na+ transporters NHE2 and NHE3. This effect may underlie the development of mucosal dysfunction and diarrhea often found to be associated with chronic inflammatory diseases of the bowel.
This work was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grants DK-47722 (E. B. Chang) and DK-38510 (E. B. Chang), National Institutes of Health Digestive Disease Research Core Center Grant DK-42086, National Cancer Institute Grant CA-14599 (Cancer Research Center, University of Chicago), and the Gastrointestinal Research Foundation of Chicago.
F. Rocha and M. W. Musch contributed equally to this work.
Address for reprint requests and other correspondence: E. B. Chang, The Martin Boyer Laboratories, Dept. of Medicine, MC 6084, The Univ. of Chicago, 5841 South Maryland Ave., Chicago, IL 60637 (E-mail:).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2001 the American Physiological Society