HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 50, 49036-49046, December 13, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/50/49036    most recent M205288200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Woo, A. L. Articles by Shull, G. E. Search for Related Content PubMed PubMed Citation Articles by Woo, A. L. Articles by Shull, G. E. Social Bookmarking                      What's this?

In Vivo Evidence for Interferon--mediated Homeostatic Mechanisms in Small Intestine of the NHE3 Na⁺/H⁺ Exchanger Knockout Model of Congenital Diarrhea^*

Alison L. Woo, Lucy A. Gildea^§, Leslie M. Tack^§, Marian L. Miller^¶, Zachary Spicer, David E. Millhorn, Fred D. Finkelman^§, Daniel J. Hassett, and Gary E. Shull^**

From the Departments of  Molecular Genetics, Biochemistry, and Microbiology, ^§ Medicine, and ^¶ Environmental Health and the  Genome Research Institute, University of Cincinnati College of Medicine, Cincinnati, Ohio 45267-0524

Received for publication, May 29, 2002, and in revised form, October 4, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Mice lacking NHE3, the major absorptive Na⁺/H⁺ exchanger in the intestine, are the only animal model of congenital diarrhea. To identify molecular changes underlying compensatory mechanisms activated in chronic diarrheas, cDNA microarrays and Northern blot analyses were used to compare global mRNA expression patterns in small intestine of NHE3-deficient and wild-type mice. Among the genes identified were members of the RegIII family of growth factors, which may contribute to the increased absorptive area, and a large number of interferon--responsive genes. The latter finding is of particular interest, since interferon- has been shown to regulate ion transporter activities in intestinal epithelial cells. Serum interferon- was elevated 5-fold in NHE3-deficient mice; however, there was no evidence of inflammation, and unlike conditions such as inflammatory bowel disease, levels of other cytokines were unchanged. In addition, quantitative PCR analysis showed that up-regulation of interferon- mRNA was localized to the small intestine and did not occur in the colon, spleen, or kidney. These in vivo data suggest that elevated interferon-, produced by gut-associated lymphoid tissue in the small intestine, is part of a homeostatic mechanism that is activated in response to the intestinal absorptive defect in order to regulate the fluidity of the intestinal tract.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

A balance between absorption and secretion of ions in the intestinal tract is necessary to maintain proper ionic and fluid composition of the luminal contents and to recover electrolytes, acid-base equivalents, and water. The absorption of NaCl, HCO, and water is due, in part, to apical Na⁺/H⁺ exchange. There are five known plasma membrane Na⁺/H⁺ exchangers (NHEs),¹ three of which are expressed in the intestine (1-3). NHE2 and NHE3 are localized to the apical membranes of the intestinal epithelia (4), whereas NHE1 is localized to basolateral membranes (5). Previous studies of NHE3-deficient (Nhe3^/) mice have shown that NHE3 is the major absorptive Na⁺/H⁺ exchanger in the intestine (6, 7) and that NHE2 provides little, if any, compensation for the loss of NHE3 (8).

The NHE3 knockout has a severe absorptive defect in the intestine (7) and is currently the only animal model of congenital diarrhea. Loss of NHE3 results in sharp up-regulation of the epithelial Na⁺ channel and colonic H⁺,K⁺-ATPase in colon (7), the induction of an amiloride-sensitive Na⁺ absorptive mechanism in the small intestine (6), and enlargement of all segments of the intestinal tract. These observations suggest that compensation occurs by alterations in ion transporter activities and a modest increase in the absorptive area of the intestine (7). In the colon, increased aldosterone is involved in the induction of the apical Na⁺ channel, K⁺ channel, and colonic H⁺,K⁺-ATPase (7, 9-11), which operate in concert to provide an alternative Na⁺-absorptive mechanism (12). Little is known, however, about the underlying genetic regulatory mechanisms that are activated in the small intestine during chronic diarrheas. To address this issue, global mRNA expression patterns were examined in small intestines of Nhe3^/ and Nhe3^+/+ mice using cDNA microarrays. We were particularly interested in genes that have the potential to regulate either ion transport processes or growth and development, such as signal transduction molecules, transport proteins, and growth or transcription factors. Some of the most interesting genes identified included putative growth factors of the regenerating gene subtype III family and, surprisingly, a large number of IFN--responsive genes.

A number of investigators have shown previously that IFN- regulates the expression and activity of ion transporters involved in intestinal absorption and secretion (13-17). The identification of so many IFN--responsive genes altered in the Nhe3^/ small intestine suggested that elevated IFN- might serve as a mechanism for regulating the balance between absorption and secretion, thereby adjusting the fluidity of the intestinal contents to a more appropriate level for digestive functions and reducing the loss of electrolytes and acid-base equivalents. The apparent IFN- response was puzzling, however, because IFN- has been associated almost invariably with inflammatory responses, yet previous studies provided no evidence for inflammation in Nhe3^/ mice (7). Analysis of serum cytokine levels in Nhe3^/ mice revealed that IFN- levels were increased, but levels of other cytokines characteristic of known models of gut inflammation were unchanged. Moreover, there was no histological or molecular evidence for inflammation in small intestine and no signs of pathogenic bacteria detected by our methods. These data, showing that IFN- levels were elevated and that the expression of numerous IFN--responsive genes was altered in the Nhe3^/ small intestine, in the absence of any detectable inflammation, suggest the possibility of a homeostatic function for IFN- in chronic diarrhea.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Generation and Care of Experimental Mice-- Colonies of NHE3-deficient mice were generated previously (7) and maintained on a mixed 129SvJ and Black Swiss background because viability was severely impaired on an inbred background (8). All experimental pairs of Nhe3^/ and Nhe3^+/+ mice were 8-12 weeks old and were littermates matched by both age and sex. Colonies of ROMK-deficient mice were generated and maintained as previously described (18). Mice were housed in a barrier facility, in individually ventilated microisolation cages. To ensure a pathogen-free environment, sentinel mice housed on each rack were tested every three months for viral infections, including, but not limited to, mouse hepatitis virus, rotavirus, hantavirus, adenovirus, cytomegalovirus, and parvovirus. A viral infection was never detected in this room.

Microarray Analysis-- Total RNA was extracted from pooled small intestines of three Nhe3^+/+ and three Nhe3^/ mice using Tri-Reagent (Molecular Research Center, Inc., Cincinnati, OH), and poly(A) RNA was purified using the MicroPoly(A) pure small scale mRNA purification kit (Ambion, Inc., Austin, TX). Two separate sets of poly(A) RNAs from Nhe3^/ and Nhe3^+/+ small intestines were sent to IncyteGenomics (St. Louis, MO), where three experiments were performed using the mouse GEM2 gene expression array. In experiment 1A, the first set of poly(A) RNAs from Nhe3^/ and Nhe3^+/+ small intestine pooled samples were labeled with Cy3 and Cy5, respectively. In experiment 1B, the same set of poly(A) RNAs were used, but the Cy3 and Cy5 labels were reversed. In the third experiment, experiment 2, the second set of pooled poly(A) RNAs from Nhe3^/ and Nhe3^+/+ small intestines were labeled with Cy3 and Cy5, respectively. The first two hybridizations used GEM2.08, and the third hybridization used GEM2.25.

Microarray data were analyzed using Incyte Gemtools software and GeneSpring software (Silicon Genetics, Redwood City, CA). Only those samples that passed the quality control specifications outlined by IncyteGenomics, determined using Gemtools, were incorporated in the study. The Cy3 and Cy5 signals for each data set were normalized using a balancing coefficient determined from the total signal intensity of each dye on the microarray. In addition, an intensity-dependent normalization procedure, provided by the GeneSpring software, was used to normalize data at lower fluorescence levels, where nonlinearity in the ratios of the two dyes is often observed (19). Differential mRNA expression levels were calculated from the ratio of Nhe3^/ to Nhe3^+/+ fluorescence measurements and are referred to as -fold changes. -Fold changes greater than +1 represent up-regulation in the knockout compared with wild type, whereas -fold changes less than 1 represent down-regulation in the knockout compared with wild type. Lists of up-regulated and down-regulated genes were compiled by determining the average -fold changes in gene expression for experiments 1A and 1B (collectively referred to as experiment 1) and then calculating the average -fold changes in gene expression for experiments 1 and 2. Values that exceeded an average of 1.4-fold differential expression, the value recommended by IncyteGenomics, are reported in Tables I and II. The recommended differential expression level of 1.4 was based on an IncyteGenomics GEM Microarray Validation study showing that the percentage coefficient of variation (the relative S.D. value) between two experiments was 12%; thus, fewer than 1% of genes that are not differentially expressed will vary by as much as 1.4 in a single experiment. The reproducibility between experiments 1A and 1B and between experiments 1 and 2 was excellent, with coefficients of variation of 10.1 and 9.1%, respectively. Northern blot analyses of a third distinct set of RNA samples was performed to confirm the reliability of microarray data obtained from the first two sets of samples.

Northern Blot Analysis-- Total RNA was extracted from the pooled small intestines of three Nhe3^+/+ and Nhe3^/ mice using Tri-Reagent (Molecular Research Center). These samples were from a different set of animals than those used for the microarrays. Total RNA (10 µg/sample) was mixed with Glyoxal sample buffer (BioWhittaker Molecular Applications, Rockland, ME), separated by electrophoresis in 1% agarose, and transferred to Hybond-N+ nylon membrane (Amersham Biosciences). Northern blots were screened using ³²P-labeled cDNA probes, all generated from I.M.A.G.E. consortium clones except for Na,K-ATPase ₁ (20), Na,K-ATPase ₁ (20), and NKCC1 (mouse codons 427-695) probes. I.M.A.G.E. clones were purchased from IncyteGenomics or the University of Cincinnati Genomics and Microarray Laboratory and were sequence-verified. Quantitation of mRNA levels was determined by PhosphorImager analysis (Amersham Biosciences) using ImageQuant software (Amersham Biosciences). Gene expression levels were normalized using the L32 ribosomal protein mRNA as a loading control and reported as mean volume-integrated values for replicate samples from four lanes.

In Vivo Cytokine Measurements and Lipopolysaccharide (LPS) Stimulation-- The Cincinnati Cytokine Capture Assay (an enzyme-linked immunosorbent assay) was used for in vivo measurement of serum IFN-, TNF-, IL-2, IL-4, and IL-6 in five pairs of Nhe3^/ and Nhe3^+/+ mice and of IFN- in four pairs of Romk^/ and Romk^+/+ mice (21). All mice were housed in a barrier facility until the day of the experiment. Biotin-labeled anti-IFN- mAb (R4-6A2), anti-TNF- mAb (TN3) (Pharmingen), anti-IL-4 mAb (BVC4-1D11), anti-IL-2 mAb (JES6-5H5) (Pharmingen), and anti-IL-6 mAb (Pharmingen) were administered via the lateral tail vein, and mice were housed overnight in a clean facility. Blood was collected 24 h later, and serum cytokine levels were analyzed by enzyme-linked immunosorbent assay as described previously (21). LPS stimulation of mice that had been injected with biotinylated antibody was achieved by intraperitoneal injection of 50 µg of Salmonella enteritidis LPS (Sigma) (22, 23). Blood was collected 6 h later, and serum cytokine levels were determined. Statistics were performed using Student's t test, and changes were considered significant if p was <0.05.

Morphometry and Histology-- Whole small intestines and colons were dissected from Nhe3^/ and Nhe3^+/+ mice, fixed in 10% buffered formalin phosphate (Fisher), dehydrated, and embedded in paraffin for hematoxylin and eosin staining. Paraffin sections were cut at 5 µm. Using a camera lucida, Summagraphics digitizing tablet, and SigmaScan Pro software (Jandel Scientific), the following measurements were recorded. The area of Peyer's patches (µm²) and linear distance of submucosa (µm) was digitized, and plasma cells and inflammation (i.e. neutrophils, eosinophils, and small mononuclear cells) were assessed semiquantitatively on a 0-5+ scale with 5+ being most severe. In these analyses, n refers to the number of mice examined, and all segments of the small and large intestine were analyzed. Data were analyzed using SAS 8e for the PC, using the general linear model. Changes were considered significant if p was <0.05.

Semiquantitative RT-PCR-- Total cellular RNA was extracted from small intestine, colon, spleen, and kidneys of three individual mice of each genotype (these were different mice than those used for the microarray and Northern blot analysis) using Tri-Reagent (Molecular Research Center). 5 µg of each RNA sample was converted to cDNA in a total volume of 21 µl, using the SuperScript first-strand synthesis system for RT-PCR (Invitrogen). cDNA products were amplified by PCR using the following gene-specific primers: IFN- (5'-TGGCTGTTTCTGGCTGTTACTG-3' and 5'-AATCAGCAGCGACTCCTTTTCC-3') (24), IL-1 (5'-TGACGGACCCCAAAAGATGAAG-3' and 5'-CTGCTTGTGAGGTGCTGATGTA-3') (25), IL-18 (5'-GGCCCAGGAACAATGGCTGCC-3' and 5'-GGGTCACAGCCAGTCCTCTTAC-3') (26), IL-12p40 (5'-AGATGACATCACCTGGACCT-3' and 5'-GCCATGAGCACGTGAACCGT-3') (26), transforming growth factor- (5'-GCGGACTACTATGCTAAAGAGG-3' and 5'-GTTGTGTTGGTTGTAGAGGGCA-3') (24), and -actin (5'-ATGTACGTAGCCATCCAGGC-3' and 5'-AAGGAAGGCTGGAAAAGAGC-3') (27). PCR conditions were as follows: 94 °C for 2 min and then 30 cycles at 94 °C for 30 s, 58 °C (IL-1 and -actin) or 60 °C (IFN-, IL-18, IL-12p40, transforming growth factor-) for 1 min, and 72 °C for 1 min, followed by 72 °C for 10 min. PCR products were separated by agarose gel electrophoresis and visualized by ethidium bromide staining.

Real Time PCR-- The same cDNA samples used for semiquantitative PCR were also used for real time PCR. Reverse-transcribed cDNA was amplified using gene-specific primers for IFN- (5'-TCAAGTGGCATAGATGTGGAAGAA-3' and 5'-TGGCTCTGCAGGATTTTCATG-3', 92-base pair product) (28) and IL-1 (5'-ACACTCCTTAGTCCTCGGCCA-3' and 5'-CCATCAGAGGCAAGGAGGAA-3', 82-base pair product) (29). PCRs were carried out in a Cepheid Smart Cycler (Cepheid, Sunnyvale, CA) as described previously (30) and contained the following components in a total volume of 25 µl: 1× LightCycler-DNA Master SYBR Green I (Roche Diagnostics), 10 pmol of each primer, and 0.05-1 µg of RNA-equivalent cDNA. The annealing temperature was 60 °C for both primer sets, fluorescence was recorded during the 72 °C extension step of each cycle, and threshold cycle values were calculated from the second derivative of the growth curve as described previously (30). The line equations of the linear regions for each curve were used to calculate the relative -fold change between Nhe3^/ and Nhe3^+/+ tissues.

Analysis of Bacterial Flora in Small Intestine-- Mice were euthanized by CO₂ inhalation and placed in an anaerobic chamber. The contents of the small intestines were gently squeezed out and weighed in preweighed 1.5-ml microcentrifuge tubes. Intestinal contents were diluted in PRAS anaerobic medium (Randolf Biomedical, West Warwick, RI) to a final concentration of 0.4 mg/ml. The intestinal suspension was vortexed for several minutes and used for 10-fold serial dilutions in PRAS medium. 50 µl of each dilution was spread onto a plate and grown anaerobically. Bacteria were cultured on Brucella blood agar (Anaerobe Systems, Morgan Hill, CA) and egg yolk agar (Anaerobe Systems), both enriched nonselective media, to obtain the total number of bacteria/g of intestinal contents. Bacteria were also cultured on Bacteroides bile esculin agar and cycloserine-cefoxitin fructose agar (Anaerobe Systems) for the isolation and identification of Bacteroides fragilis and Clostridium difficile, respectively. All plates were cultured at 37 °C for 24-48 h under anaerobic conditions, until distinct colonies were formed. The average number of colonies/g of intestinal contents ± S.E. for three pairs of mice was calculated.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

mRNA Expression Profile of Nhe3^/ Small Intestine-- cDNA microarray analyses of mRNAs from Nhe3^/ and Nhe3^+/+ small intestines were performed to identify differentially regulated genes. The GEM2 microarrays contained 10,200 elements (IncyteGenomics). In each of the three competitive hybridization experiments, a number of elements did not meet the appropriate hybridization criteria. These elements were excluded, leaving 8234, 8044, and 9257 elements that passed all criteria in experiments 1A, 1B, and 2, respectively. All elements that changed an average of 1.4-fold or greater in experiments 1 and 2 were identified, resulting in 53 known genes that were up-regulated (Table I) and 28 known genes that were down-regulated (Table II).

View this table: [in this window] [in a new window]   Table I Up-regulated genes in the Nhe3/ small intestine Microarrays were used to identify up-regulated genes in the Nhe3/ compared with the Nhe3+/+ small intestine. -Fold changes are reported for experiments 1 and 2, and the average -fold change in both experiments is also listed. ND, -fold change not determined because the gene was not represented on that version of the microarray. Expt., experiment.

View this table: [in this window] [in a new window]   Table II Down-regulated genes in the Nhe3/ small intestine Microarrays were used to identify down-regulated genes in the Nhe3/ compared with the Nhe3+/+ small intestine. -Fold changes are reported for experiments 1 and 2, and the average -fold change in both experiments is also listed. ND, -fold change not determined because the gene was not represented on that version of the microarray. Expt., experiment.

As discussed in more detail below, expression levels of 16 differentially expressed genes were examined by Northern blot analyses of a third set of pooled RNA samples. We included two genes that were just below our cut-off of 1.4 (CFTR, 1.30; trefoil factor 1, 1.32) and one gene, small proline-rich protein 2a, which was identified as up-regulated in experiments 1A and 1B but down-regulated in experiment 2. The latter gene has no known function in the intestine but has been implicated in barrier function in squamous epithelium (31) and was identified as up-regulated in an Nhe3^/ small intestine subtraction/suppression library.² The Northern blot data (Figs. 1-4) confirmed the microarray data for all but one of the genes. For metallothionein I, the microarrays detected a 1.90-fold up-regulation in Nhe3^/ small intestines, whereas Northern blots revealed only a 1.1-fold increase (Fig. 1). These experiments show that there is a good correlation between the microarray and Northern blot data, although the Northern blots tended to yield higher levels of induction or repression.

View larger version (55K): [in this window] [in a new window]   Fig. 1.   Northern blot analysis of gene expression levels in Nhe3/ and Nhe3+/+ small intestine. Each lane contains 10 µg of RNA from the pooled tissues of three mice. Replicate blots were used to quantitate expression of different genes, and samples were run in quadruplicate on each blot. Hybridization signals were quantitated by PhosphorImager analysis and normalized to the signal for L32 ribosomal subunit mRNA. The -fold change in expression levels in Nhe3/ versus Nhe3+/+ samples, determined by Northern blot (N) or microarray (M) analysis, are shown beside the corresponding blot. Gran. B, granzyme B; crp duc, crp-ductin; MTI, metallothionein 1; galectin, lectin- and galactose-binding, soluble 9; alc.deh., alcohol dehydrogenase 1 complex; cyt. P450, cytochrome P450, steroid-inducible 3a11; TFF1, trefoil factor 1; TFF2, trefoil factor 2 (spasmolytic protein 1); TFF3, trefoil factor 3; L32, L32 ribosomal subunit (loading control).

Up-regulation of the Regenerating Gene Subtype III Family-- Two of the largest changes in gene expression identified by the microarrays were members of the regenerating (Reg) gene subtype III family, RegIII (also called pancreatitis-associated protein, or PAP) and RegIII (also called islet neogenesis-associated protein-related protein). The physiological functions of these proteins are unknown; however, on the basis of their homology to the well characterized growth factor Reg (32, 33), they are hypothesized to be involved in tissue regeneration and growth in the small intestine. Northern blot and microarray expression data were identical for RegIII and RegIII, ~8- and ~5.4-fold up-regulated, respectively, in Nhe3^/ small intestine (Fig. 2). The RegIII gene family also includes two other genes, RegIII and RegIII, neither of which were represented on the microarrays. RegIII is up-regulated in another example of intestinal growth, the adapting remnant of the small intestine in a resection model of intestinal adaptation (34). To gain insight into how this gene family is regulated in the Nhe3^/ small intestine, expression of RegIII was examined and found to be up-regulated 12-fold (Fig. 2). The up-regulation of the RegIII genes in both the resection model and in the enlarged Nhe3^/ small intestine suggests that they have roles as intestinal growth factors, similar to the role of Reg in pancreas (32, 33).

View larger version (72K): [in this window] [in a new window]   Fig. 2.   Northern blot analysis of members of the RegIII family in Nhe3/ small intestine. Blots were prepared and analyzed as described in the legend to Fig. 1. The -fold change in expression levels for RegIII, -, and - in Nhe3/ versus Nhe3+/+ small intestine, determined by Northern blot (N) or microarray (M) analysis, is shown beside the corresponding blot. RegIII was only represented on GEM2.08 and not on GEM2.25; therefore, the average -fold change in the microarray column represents data from experiments 1A and 1B. L32, L32 ribosomal subunit (loading control).

Differential Expression of IFN--inducible GTPases and Other IFN--regulated Genes-- mRNAs encoding IFN--inducible GTPases were also up-regulated in the Nhe3^/ small intestine. Northern analyses showed that IFN--induced GTPase, T cell-specific GTPase, and IFN--inducible 47-kDa protein were up-regulated 7.6-, 1.65-, and 5.47-fold, respectively (Fig. 3). IFN--induced GTPase and IFN--inducible 47-kDa protein were up-regulated to a much greater extent than detected by the microarrays.

View larger version (76K): [in this window] [in a new window]   Fig. 3.   Northern blot analysis of IFN--inducible GTPases in the Nhe3/ and Nhe3+/+ small intestine. Blots were prepared and analyzed as described in the legend to Fig. 1. The -fold change in expression levels in Nhe3/ versus Nhe3+/+ samples, determined by Northern blot (N) or microarray (M) analysis, is shown beside the corresponding blot. IGTP, IFN--induced GTPase; TGTP, T cell-specific GTPase; IRG-47, IFN--inducible protein, 47 kDa; L32, L32 ribosomal subunit (loading control).

In addition to these three IFN--inducible GTPases, 24 other genes from Table I and four genes from Table II have been reported to be responsive to IFN- (Table III). The identification of such a large number of IFN--responsive genes with altered expression levels in the Nhe3^/ small intestine was surprising, because the original analysis of these mice (7) did not reveal evidence of inflammation, one of the major causes of elevated IFN- (35). It seemed possible that alterations in expression of these IFN--responsive genes might be part of an adaptive response to the chronic diarrhea. Consistent with this hypothesis, IFN- has been implicated as an important molecule in regulating a number of ion transporters, including CFTR and the basolateral Na⁺-K⁺-2Cl cotransporter (NKCC1), in cultured intestinal epithelial cells (13-16).

View this table: [in this window] [in a new window]   Table III IFN--responsive genes identified by microarrays All genes that were either up-regulated or down-regulated in Nhe3/ small intestine and are responsive to IFN- are listed in alphabetical order.

mRNA Expression Levels of Ion Transporters Involved in Absorption and Secretion-- The absorption of NaCl, HCO, and water in the small intestine is regulated in part by NHE3. To test the possibility that the loss of NHE3 might alter the mRNA expression of intestinal ion transporters involved in absorption or secretion, Northern analysis was performed (Fig. 4). Expression of the basolateral Na⁺,K⁺-ATPase ₁ and ₁ subunits, which contribute to both absorption and secretion, was unchanged. Similarly, the expression of NKCC1, which functions in both secretory and nonsecretory cells, was unchanged. mRNA encoding CFTR, the apical Cl channel required for secretion in the intestine, was down-regulated ~30-40% in the Nhe3^/ small intestine.

View larger version (77K): [in this window] [in a new window]   Fig. 4.   Northern blot analysis of ion transporters in Nhe3/ small intestine. Blots were prepared and analyzed as described in the legend to Fig. 1. The -fold change in expression levels for NKA1, NKA1, NKCC1, and CFTR in Nhe3/ versus Nhe3+/+ small intestine, determined by Northern blot (N) or microarray (M) analysis, is shown beside the corresponding blot. NKA1, Na+-K+-ATPase 1 isoform; NKA1, Na+-K+-ATPase 1 isoform; NKCC1, Na+-K+-2Cl cotransporter isoform 1; L32, L32 ribosomal protein (loading control).

Serum Levels of IFN- Are Elevated, but Levels of Other Cytokines Are Unaltered in Nhe3^/ Mice-- To determine whether the apparent IFN- response was, in fact, due to elevated IFN-, we analyzed serum levels of this cytokine in NHE3-deficient and wild-type mice. Serum IFN- levels were increased 5-fold in Nhe3^/ mice (Fig. 5), suggesting that elevated IFN- was most likely responsible for the altered expression of IFN--responsive genes. Increased IFN- is characteristic of inflammation, such as that seen in inflammatory bowel disease, in which it is invariably accompanied by changes in other cytokine levels (36). To better understand the cytokine status of Nhe3^/ mice, serum levels of TNF-, IL-4, IL-2, and IL-6 were measured. IL-2 and IL-6 levels are elevated in inflammatory bowel disease, whereas IL-4 levels are decreased (36). The levels of TNF-, IL-4, IL-2, and IL-6 were not significantly different in Nhe3^/ mice (Fig. 5), suggesting that the elevated IFN- levels were due not to inflammation but to some other disorder in NHE3-deficient mice.

View larger version (17K): [in this window] [in a new window]   Fig. 5.   Serum levels of INF-, TNF-, IL-4, IL-2, and IL-6 in Nhe3/ and Nhe3+/+ mice. The serum levels of cytokines in five pairs of Nhe3/ and Nhe3+/+ mice were tested using the Cincinnati cytokine capture assay (see "Experimental Procedures"). Data represent means ± S.E. *, p < 0.00005 compared with the corresponding Nhe3+/+ mice, using Student's t test.

Elevated Serum IFN- Does Not Occur in Response to a Deficit in Extracellular Fluid Volume-- In addition to the diarrheal state, another major systemic disorder in NHE3-deficient mice is extracellular fluid volume depletion (7). To determine whether the elevated IFN- levels might be a general response to volume depletion rather than being due to the loss of NHE3 and the resulting diarrhea, we analyzed serum IFN- levels in ROMK-deficient mice, another model of severe extracellular fluid volume depletion (18). Serum IFN- levels were not significantly different in Romk^/ and wild-type mice (Fig. 6), suggesting that a reduction in extracellular fluid volume has no affect on IFN- levels.

View larger version (9K): [in this window] [in a new window]   Fig. 6.   Serum levels of INF- in Romk/ and Romk+/+ mice. The serum levels of INF- in four pairs of Romk/ and Romk+/+ mice were measured using the Cincinnati Cytokine Capture Assay (see "Experimental Procedures"). Data are means ± S.E. A Student's t test revealed no statistically significant differences.

Histological Analyses Provide No Evidence for Inflammation in Nhe3^/ Small Intestines-- To further assess the possibility that inflammation was involved in the IFN- response, histological analysis of the duodenum, ileum, jejunum, and colon of both genotypes was performed. As shown for jejunum and ileum in Fig. 7, Nhe3^/ and Nhe3^+/+ small intestines did not contain an inflammatory infiltrate in any of the sections examined. The area of Peyer's patches, which comprise the lymphoid tissue in the intestinal mucosa, was quantified, and a semiquantitative value for inflammatory cells, such as eosinophils, neutrophils, and plasma cells, was assigned. Although the difference was not statistically significant, there were more Peyer's patches in Nhe3^/ small intestines (Nhe3^+/+, 3.45 ± 2.00 µm², n = 7; Nhe3^/, 11.37 ± 5.5 µm², n = 7). The numbers of infiltrating inflammatory cells (Nhe3^+/+, 0.05 ± 0.028%, n = 7; Nhe3^/, 0.07 ± 0.04%, n = 7) and plasma cells (Nhe3^+/+, 0.69 ± 0.17%, n = 7; Nhe3^/, 0.81 ± 0.27%, n = 7) were similar in both genotypes. In addition, there were no significant differences between the Nhe3^+/+ and Nhe3^/ colons in the relative number of Peyer's patches (Nhe3^+/+, 3.10 ± 1.58 µm², n = 6; Nhe3^/, 5.67 ± 5.4 µm², n = 5), the number of infiltrating cells (Nhe3^+/+, 0.06 ± 0.04%, n = 6; Nhe3^/, 0.05 ± 0.04%, n = 5), or the number of plasma cells (Nhe3^+/+, 0.12 ± 0.05%, n = 6; Nhe3^/, 0.21 ± 0.07%, n = 5). In summary, these histological analyses revealed no signs of inflammation in the Nhe3^/ small intestine or colon, and, as noted under "Discussion," the possible increase in Peyer's patches is unlikely to contribute to IFN- production.

View larger version (139K): [in this window] [in a new window]   Fig. 7.   Histology of Nhe3/ and Nhe3+/+ small intestine. Representative sections of jejunum and ileum stained with hematoxylin and eosin. A and B, cross-sections of Nhe3/ jejunum and ileum, respectively. C and D, longitudinal sections of Nhe3/ jejunum and ileum, respectively. E and F, cross sections of Nhe3+/+ jejunum and ileum, respectively. G and H, longitudinal sections of Nhe3+/+ jejunum and ileum, respectively. Bar, 100 µm.

Quantitation of Cytokine Expression in Nhe3^/ Small Intestines-- To generate molecular evidence for the presence or absence of inflammation and to determine whether the elevated serum IFN- levels resulted from a localized increase in expression in the small intestine, expression levels of cytokines characteristic of inflammation were analyzed by RT-PCR. Inflammatory bowel disease is characterized by increases in IFN- and IL-1 throughout the gut (36), whereas the loss of transforming growth factor- has been shown to contribute to the development of inflammatory bowel disease (37). In addition, IL-12 and IL-18, both potent inducers of IFN- production by T cells (38), have been shown to synergistically induce intestinal inflammation by IFN--dependent mechanisms (39). These cytokines were therefore examined in the small intestine, colon, spleen, and kidney of both Nhe3^/ and Nhe3^+/+ mice. IFN- mRNA was up-regulated in Nhe3^/ small intestine, but an increase was not evident in colon, spleen, or kidney (Fig. 8A). IL-1 expression appeared slightly up-regulated in Nhe3^/ small intestine and colon, whereas the expression of IL-12 and IL-18 did not appear to be up-regulated in any tissue (Fig. 8A).

View larger version (31K): [in this window] [in a new window]   Fig. 8.   Semiquantitative and quantitative RT-PCR analysis of cytokine expression in small intestine (Sm. Int.), colon, spleen, and kidney from Nhe3/ and Nhe3+/+ mice. RNA samples were isolated from the small intestine, colon, spleen, and kidneys from three pairs of Nhe3/ and Nhe3+/+ mice. A, semiquantitative RT-PCR using gene-specific primers. PCR products were separated by electrophoresis on a 2% agarose gel and visualized by ethidium bromide staining. -Actin was used as a loading control. B, quantitation of IFN- and IL-1 mRNA levels determined by real time PCR. **, p < 0.005; *, p < 0.02 compared with the corresponding Nhe3+/+ controls, using Student's t test.

In order to quantitate this apparent up-regulation of IFN- in small intestine and IL-1 in colon, real time PCR was employed. IFN- mRNA was up-regulated 3.14 ± 0.62-fold in Nhe3^/ small intestine, but an increase was not evident in colon, spleen, or kidney (Fig. 8B). IL-1 mRNA was up-regulated 2.31 ± 0.42-fold in colon but was not induced in small intestine, suggesting that an IFN--specific response and not a general inflammatory response was occurring in the small intestine (Fig. 8B). Conversely, there appeared to be a trend toward down-regulation of IFN- and IL-1 expression, 1.45-fold and 1.42-fold, respectively, in kidney (Fig. 8B), suggesting that the IFN- response in NHE3-deficient small intestine has no counterpart in the kidney, which also expresses NHE3. Together, these data provide no molecular evidence for inflammation in the small intestine, despite the localized up-regulation of IFN- expression.

Analysis of the Bacterial Flora in Nhe3^/ and Nhe3^+/+ Small Intestine-- Immune responses to normal bacterial flora and infections from pathogenic bacteria can induce IFN- and other inflammatory cytokine levels (40). Viral infections can also induce IFN-, although it seems unlikely that this occurred, because the Nhe3^/ mice were housed in a barrier facility that was screened routinely for viral pathogens (as described under "Experimental Procedures"). Small intestinal contents of Nhe3^/ and Nhe3^+/+ mice were cultured under anaerobic conditions to examine their bacterial flora. When analyzed on two different enriched, nonselective mediums, the total number of bacteria/g of small intestinal contents was not significantly different in Nhe3^/ mice (Fig. 9). To look more closely at whether different bacterial populations were present in Nhe3^/ and Nhe3^+/+ small intestines, media specific for growth of C. difficile and B. fragilis were also employed. Cultures for C. difficile, a Gram-positive, anaerobic pathogen that causes diarrhea and colitis in patients (41), were negative for all of the Nhe3^/ and Nhe3^+/+ small intestines examined. On the other hand, cultures for B. fragilis, a Gram-negative obligate anaerobe, revealed higher numbers in the Nhe3^/ small intestine. Three of four Nhe3^/ small intestines contained B. fragilis (1.5 × 10⁶ ± 1.4 × 10⁶ colonies/g of contents, n = 3), whereas only two of four Nhe3^+/+ small intestines had any (1.90 × 10⁴ ± 1.60 × 10⁴ colonies/g of contents, n = 2).

View larger version (12K): [in this window] [in a new window]   Fig. 9.   Analysis of bacterial flora in Nhe3/ and Nhe3+/+ small intestines. Small intestines were individually removed in a sterile, anaerobic environment from three pairs of Nhe3/ and Nhe3+/+ mice. Intestinal contents were extruded, and bacteria were cultured under anaerobic conditions at 37 °C on both Brucella blood agar and egg yolk agar. The number of colony-forming units (CFU) denotes the total number of colonies/g of intestinal contents. Data represent means ± S.E. Student's t test revealed no statistically significant differences.

Unlike Gram-positive bacteria, Gram-negative strains produce LPS that contains lipid A or endotoxin, which can elicit a dramatic immune response in animal cells (42). It seems unlikely that B. fragilis LPS is involved in the IFN- response in Nhe3^/ mice, because it induces only a weak cytokine response in mice, and this response is not limited to IFN- but also includes increases in TNF- and IL-6 (43). Although the total numbers of bacteria in Nhe3^/ or Nhe3^+/+ small intestines are not significantly different, it is possible that there are more Gram-negative bacteria in the Nhe3^/ small intestine. If this were the case, then the LPS produced by these bacteria could lead to elevated IFN- levels. Primary exposure to LPS induces gene expression of multiple inflammatory and immunological genes, including cytokines (44). Repeated exposure to LPS results in LPS tolerance, a phenomenon whose characteristics include a lower induction of INF-, TNF-, and IL-6 levels (45).

To investigate whether Nhe3^/ mice exhibit signs of LPS tolerance as a result of more LPS-producing bacteria in their intestine, the cytokine response to LPS stimulation was determined in both genotypes. The cytokine response of Nhe3^/ mice to LPS stimulation was the same as that of Nhe3^+/+ mice, and both produced equally elevated levels of IFN-, TNF-, and IL-6, whereas IL-2 and IL-4 were unchanged (Fig. 10). These data showed that the LPS-induced response of Nhe3^/ mice did not differ from that of Nhe3^+/+ mice, suggesting that the Nhe3^/ mice did not have higher levels of LPS exposure. A preliminary study using another model of antigen stimulation, injection of affinity-purified goat antibody to mouse IgD (46), also revealed no difference in either the level of induction of IL-4 or IgE in the two genotypes (data not shown). Together, these data provide evidence that the elevated IFN- levels in Nhe3^/ mice were not part of an inflammatory response, suggesting a possible homeostatic role for this cytokine in a state of chronic diarrhea.

View larger version (12K): [in this window] [in a new window]   Fig. 10.   Cytokine profile of Nhe3/ and Nhe3+/+ mice after LPS stimulation. 50 µg of S. enteritidis LPS was injected intraperitoneally into Nhe3/ and Nhe3+/+ mice (n = 5 for each genotype). Serum was collected 6 h later, and IFN-, TNF-, IL-4, IL-2, and IL-6 levels were analyzed using the Cincinnati cytokine capture assay (see "Experimental Procedures"). Data represent means ± S.E. Student's t test revealed no statistically significant differences.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Gene-targeted mice lacking NHE3 are currently the only experimental model of congenital diarrhea. Given the severity of the intestinal absorptive defect in these animals (7), we anticipated that there might be alterations in gene expression in the small intestine that would provide information about either homeostatic mechanisms involved in regulating the normal balance between absorption and secretion or compensatory mechanisms that directly blunt the severity of the diarrhea. An understanding of such mechanisms is important, because diarrheal diseases kill 5-8 million people, mostly children, each year in underdeveloped countries and sicken hundreds of millions more throughout the world (47).

The identification of many IFN--responsive genes in the small intestine was at first surprising, because earlier studies of these mice had revealed no signs of inflammation (7), one of the major stimuli for elevated IFN- levels. Initially, we considered the possibility that some of the same systems that can be regulated by IFN- might have been activated by the intestinal absorptive defect and the resulting diarrhea rather than by IFN-; however, cytokine analysis demonstrated that serum IFN- was, in fact, elevated. In addition to the diarrheal state, NHE3-deficient mice have a metabolic acidosis, which is very mild in the awake animal (8), and severe extracellular fluid volume depletion that results from the loss of NHE3 in both intestine and kidney (7). However, we observed no change in IFN- levels in ROMK-deficient mice, which exhibit both metabolic acidosis and severe fluid volume depletion as a result of renal salt and water wasting (18). Thus, our data suggest that it is the diarrheal state resulting from the absorptive defect in the intestine, and not the extracellular fluid volume depletion or mild metabolic acidosis, that is responsible for the elevated IFN- levels.

Although the increased serum IFN- levels suggested the possibility of inflammation or infection, serum levels of proinflammatory cytokines such as TNF- and IL-6, which are invariably elevated during inflammation, were unchanged. Microarray analyses showed that expression of RANTES (48) and serum amyloid A3 (49), which are induced by IFN- and have roles in inflammation, were increased in the small intestine, but the levels of induction were less than 2-fold. Also, there was no up-regulation of mRNAs encoding trefoil factors (Fig. 1), which are frequently induced during inflammation (50); in fact, TFF1 and TFF2 were down-regulated. Although the differences were not significant, morphometric analyses of Nhe3^/ small intestine and colon indicated that the numbers of Peyer's patches might be increased in the small intestine. It is unlikely that increased numbers of Peyer's patches could be responsible for the increase in IFN- levels, because dendritic cells from Peyer's patches produce IL-10 and stimulate naive T cells to produce IL-4, IL-6, and IL-10, ultimately inhibiting production of IFN- (51, 52). Moreover, serum IL-4 and IL-6 levels were not elevated in Nhe3^/ mice (Fig 5), suggesting that activation of naive T cells by dendritic cells from Peyer's patches was not occurring. There were no significant changes in the numbers of infiltrating inflammatory cells or plasma cells in small intestine or colon, and finally, there was no evidence of pathogenic bacteria in the NHE3-deficient intestine. Thus, the molecular, histological, and microbiological data do not support the hypothesis that the IFN- response is the result of inflammation.

The effects of IFN- on Nhe3^/ small intestine appear to be very broad, as indicated by the number of IFN--responsive genes listed in Table III. One of the most interesting genes in this list is RegIII, the most highly up-regulated gene in the Nhe3^/ small intestine identified by the microarrays. Two other genes of this family, RegIII and RegIII, were also highly up-regulated. The biological functions of the RegIII gene products are poorly understood, yet based on homology to Reg, which has been shown to be involved in pancreatic islet cell proliferation (33) and the healing of gastrointestinal mucosal lesions (32), they are thought to function in growth and proliferation. The up-regulation of these putative growth factors could therefore be involved in the enlargement of the intestinal tract observed in both Nhe3^/ mice (7) and in the adapting remnant in the small bowel resection model (34). Up-regulation of members of the REG family, including the human homologue of mouse RegIII has also been observed during microarray analysis of intestinal RNA from inflammatory bowel disease patients (53, 54), consistent with functions in growth and repair.

PCR analysis indicated that IFN- mRNA was up-regulated in the small intestine but not in colon or kidney, tissues in which the loss of NHE3 also causes severe absorptive defects. In a similar microarray study of mRNA from NHE3-deficient kidneys,³ we observed no IFN- response, indicating that elevated serum IFN- alone does not cause significant changes in the expression of IFN--responsive genes in the kidney and that the response is specific for the small intestine. Other cytokines, including IL-12 and IL-18 that are known to synergistically induce IFN- production in T cells (38), were unchanged in small intestine. Although the possibility of a generalized inflammatory reaction or infection that has been overlooked and that also does not induce TNF-, IL-6, or other major cytokines cannot be ruled out entirely, our data suggest that IFN- is specifically up-regulated in the small intestine in response to the intestinal absorptive defect.

Given the 5-fold elevation of serum IFN-, it was surprising that IFN- mRNA was induced only 3-fold in small intestine. This local increase in IFN- mRNA in the intestine does not seem sufficient to explain the observed systemic levels of IFN- protein. However, there is recent evidence that the 5'-untranslated region of IFN- mRNA contains a pseudoknot structure that can cause a 30-fold change in translational efficiency via a mechanism involving phosphorylation of translation initiation factor 2 (55). Thus, it is possible that the levels of IFN- produced in the small intestine are regulated by multiple mechanisms and that the induction of IFN- protein is much higher than indicated by the level of mRNA induction.

A function for IFN- that does not involve immunological or inflammatory processes is unusual but not unprecedented. For example, some of the enzymes and transcription factors involved in fat metabolism are decreased by IFN- (56, 57), and IFN- also inhibits adipocyte differentiation (58). Interestingly, Nhe3^/ mice have very little fat compared with their wild-type littermates,⁴ and proteins involved in lipid metabolism, such as fatty acid transporter 2 and liver fatty acid-binding protein 1, are down-regulated in the knockout. We are currently examining the possibility that this could be the result of chronically elevated IFN- levels.

The physiological defect in the NHE3-deficient intestine is a reduction in the absorption of NaCl, with accompanying water. Given the elevated IFN- levels in Nhe3^/ mice and the large number of IFN--responsive genes identified in the intestine, it is conceivable that IFN- functions in the homeostatic mechanisms that maintain the appropriate fluidity of the intestinal contents. Little is known about how fluidity is regulated under normal physiological conditions; however, the transport proteins that mediate absorption and secretion are the effector molecules that directly control fluidity of the intestinal contents, and some of the cytokines produced by gut-associated lymphoid tissue are known to regulate the activity of these transporters. Although their effects are generally studied under pathophysiological conditions such as secretory diarrheas and inflammatory bowel disease, there is no evidence that regulation of ion transporters by cytokines is necessarily limited to disease states.

In fact, a number of investigators have demonstrated that IFN- regulates ion transport in cultured intestinal epithelial cells or in isolated sheets of small intestine in vitro. Most of these studies suggest that exposure to IFN- causes a decrease in secretion, which would directly counter a decrease in absorption. The mechanism of the IFN--mediated decrease in secretion includes down-regulation of apical Cl channel activity (15, 59), which directly reduces secretion, and a reduction in activities of the basolateral K⁺ channel (15), Na⁺-K⁺-2Cl cotransporter (15), and Na⁺,K⁺-ATPase (14, 15, 59), which are involved in maintaining the intracellular Cl concentrations and membrane potential required for anion secretion. The effects on secretion in these cultured cell systems are mediated at least in part by alterations in mRNA or protein levels, since IFN- down-regulates expression of the CFTR Cl channel mRNA (16) and both the Na⁺-K⁺-2Cl cotransporter (14) and Na⁺,K⁺-ATPase (14) protein. The antisecretory effects of IFN- in cell culture are the opposite of those for the proinflammatory cytokines TNF- (60) and IL-1 (61), which increase secretion.

Northern blot and microarray analyses suggested that CFTR mRNA is slightly down-regulated in the NHE3-deficient small intestine, but mRNAs for NKCC1 and the Na,K-ATPase  and  subunits were not changed. This suggests either that NKCC1 and Na,K-ATPase mRNA levels are not regulated by IFN- or that any changes occurring in secretory cells are masked by expression of these mRNAs in absorptive or other cell types that are not involved in secretion. Despite these inconclusive results, a recent study showed that the jejunum of NHE3-deficient mice exhibits a 35-40% reduction in cAMP-stimulated CFTR activity, consistent with the observed reduction in CFTR mRNA, and a 55-60% reduction in NKCC1-dependent anion secretion (62). These observations provide a strong in vivo correlate of the IFN- mediated down-regulation of secretion in cultured intestinal cells and support the possibility that the elevated IFN- levels in Nhe3^/ mice might play a role in compensatory down-regulation of secretion.

There is at least one study indicating that IFN- may down-regulate absorptive processes in the intestine, although this clearly does not occur in the Nhe3^/ mouse. Treatment of Caco-2 cells with IFN- led to a reduction in Na⁺/H⁺ exchange activity and protein levels of both NHE2 and NHE3, and intraperitoneal injection of IFN- into rats resulted in down-regulation of NHE2 and NHE3 mRNA, activity, and protein in the intestine (17). The authors suggested that chronic exposure to IFN- reduces apical Na⁺/H⁺ exchange in the intestine, resulting in inflammation-associated diarrhea (17). Those results conflict with the notion that IFN- might play an important homeostatic function in maintaining an appropriate balance between absorption and secretion in vivo; however, it should be noted that expression of the Nhe3 gene is up-regulated 3-fold in the small intestine of NHE3-deficient mice (although the mutant mRNA does not encode a functional protein) and that Nhe2 gene expression is unchanged (8). It is possible that the elevated serum aldosterone levels (7, 8), which are known to induce NHE3 mRNA, protein, and activity in the proximal colon (63), override any negative regulatory effects of IFN- on NaCl absorption.

A compensatory Na⁺-absorptive mechanism is activated in the small intestine of Nhe3^/ mice (6). The identity of the transporter responsible for this activity and whether IFN- has any role in its regulation are not yet known. Nevertheless, it is clear that elevated IFN- levels in Nhe3^/ mice do not prevent the activation of this alternative Na⁺-absorptive mechanism; nor, as discussed above, do they lead to down-regulation of apical Na⁺/H⁺ exchanger genes in the small intestine. On the contrary, the elevated IFN- levels in Nhe3^/ mice correlate with up-regulation of the Nhe3 gene (8) and the alternative absorptive pathway (6). They also correlate with a modest reduction in CFTR mRNA (Fig. 4), CFTR-mediated Cl secretion in response to cAMP, and NKCC1-dependent anion secretion (62), as also seen in cultured epithelial cells (15, 16, 59). Thus, the alterations in both absorptive and secretory mechanisms in the small intestine of Nhe3^/ mice would tend to blunt the severity of the absorptive defect, thereby contributing to maintenance of the appropriate fluidity of the luminal contents.

In addition to ion transporters and putative growth factors, IFN- regulates immune accessory molecules, including MHC class I and II molecules and the IgA receptor, in cultured epithelial cells (15). Our data showing that IFN- levels are elevated in NHE3-deficient mice and that there are changes in the expression of IFN--responsive and other genes involved in immunological functions (Table I-III) provide an in vivo correlate of the results of these in vitro studies. It is unusual, but not unprecedented, to see activation of the mucosal immune system in the absence of any apparent inflammation. For example, the T cell receptor-positive subpopulation of lymphocytes is dramatically increased in the sheep uterus during normal pregnancy (64); the authors suggested that these cells might serve a physiological rather than pathophysiological function during pregnancy. Similar suggestions have been made for intraepithelial T cells in the intestine (65, 66).

Mouse small intestinal intraepithelial lymphocytes express predominately T cell receptors (67). T cells are CD8⁺ and express either CD8 heterodimers or CD8 homodimers (68) and produce IFN- preferentially (69). It should be noted that T cells, with or without CD8 homodimers, and T cells with CD8 homodimers are expressed in small intestine, but not in colon (70), which did not express elevated levels of IFN- mRNA. Recently, it was shown that the CD8 complex of intestinal intraepithelial lymphocytes interacts directly with a tetramer consisting of ₂-microglobulin and thymus leukemia antigen, a nonclassical major histocompatibility complex class I molecule encoded by the T18^d gene and expressed in small intestinal epithelial cells (71). This interaction inhibits proliferation and cytotoxicity of intraepithelial lymphocytes, which would prevent disruption of the epithelium by dividing lymphocytes and destruction of healthy epithelial cells (71), and also stimulates release of IFN- (71). Interestingly, CD8, ₂-microglobulin, and the  chain of the T cell receptor are mildly up-regulated (1.39-, 1.53-, and 1.44-fold, respectively) in the Nhe3^/ small intestine. Furthermore, when we placed the NHE3-null mutation on a C57BL/6 background, which lacks the T18^d gene encoding the thymus leukemia antigen (72), null mutants exhibited abdominal bloating and died before weaning.⁵ In future studies, it will be important to test the hypothesis that the interaction between CD8, most likely expressed on T cells, and the ₂-microglobulin/thymus leukemia antigen complex is part of an early signaling event linking the absorptive defect of the Nhe3^/ small intestine and the increase in IFN- production.

In conclusion, our data demonstrate that many of the changes in gene expression in the small intestine of the Nhe3^/ congenital diarrhea model, in which inflammation was not observed, are in response to an increase in the production of IFN-. Although the in vivo data are only correlative at the present time, when viewed in the light of in vitro studies showing that IFN- reduces secretion (59) and the down-regulation of secretion in the Nhe3^/ small intestine (62), this observation suggests that IFN-, produced by gut-associated lymphoid tissue, plays a homeostatic role in controlling the fluidity of the intestinal contents. This is consistent with the hypothesis that the mucosal immune system, rather than acting solely in response to foreign antigens or pathogens, is directly involved in the normal physiological regulation of intestinal function.

	ACKNOWLEDGEMENTS

We thank Dr. Bruce Aronow for assistance in analyzing the microarray data and Dr. Karen Seta for developing the real time PCR methodology.

	FOOTNOTES

^* This work was supported by National Institutes of Health Grants RO1-DK50594, T32-DK07727, R124-DK58811, RO1-AI40541, RO1-AI44971, and R21-AI46972.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.

^** To whom correspondence should be addressed: Dept. of Molecular Genetics, Biochemistry, and Microbiology, University of Cincinnati College of Medicine, 231 Albert Sabin Way, Cincinnati, OH 45267-0524. Tel.: 513-558-0056; Fax: 513-558-1885; E-mail: shullge@ucmail.uc.edu.

Published, JBC Papers in Press, October 4, 2002, DOI 10.1074/jbc.M205288200

² A. L. Woo and G. E. Shull, unpublished observation.

³ M. Flagella, A. L. Woo, and G. E. Shull, unpublished observation.

⁴ A. L. Woo, J. N. Lorenz, and G. E. Shull, unpublished observations.

⁵ P. J. Schultheis and G. E. Shull, unpublished observation.

	ABBREVIATIONS

The abbreviations used are: NHE, Na⁺/H⁺ exchanger (the number following NHE refers to the specific isoform); Nhe3^+/+ and Nhe3^/, wild-type and homozygous mutant mice, respectively; IFN-, interferon-gamma; TNF-, tumor necrosis factor-; IL, interleukin (the number following IL refers to the specific interleukin); mAb, monoclonal antibody; LPS, lipopolysaccharide; NKCC1, isoform 1 of the Na⁺-K⁺-2Cl cotransporter; CFTR, cystic fibrosis transmembrane conductance regulator; RT, reverse transcription; RANTES, regulated on activation normal T cell expressed and secreted.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				E. M. Bradford, M. A. Sartor, L. R. Gawenis, L. L. Clarke, and G. E. Shull Reduced NHE3-mediated Na+ absorption increases survival and decreases the incidence of intestinal obstructions in cystic fibrosis mice Am J Physiol Gastrointest Liver Physiol, April 1, 2009; 296(4): G886 - G898. [Abstract] [Full Text] [PDF]

				H. Xu, H. Chen, J. Dong, J. Li, R. Chen, J. K. Uno, and F. K. Ghishan Tumor necrosis factor-{alpha} downregulates intestinal NHE8 expression by reducing basal promoter activity Am J Physiol Cell Physiol, March 1, 2009; 296(3): C489 - C497. [Abstract] [Full Text] [PDF]

				O. Gross, C. Grupp, C. Steinberg, S. Zimmermann, D. Strasser, N. Hannesschlager, W. Reindl, H. Jonsson, H. Huo, D. R. Littman, et al. Multiple ITAM-coupled NK-cell receptors engage the Bcl10/Malt1 complex via Carma1 for NF-{kappa}B and MAPK activation to selectively control cytokine production Blood, September 15, 2008; 112(6): 2421 - 2428. [Abstract] [Full Text] [PDF]

				D. Laubitz, C. B. Larmonier, A. Bai, M. T. Midura-Kiela, M. A. Lipko, R. D. Thurston, P. R. Kiela, and F. K. Ghishan Colonic gene expression profile in NHE3-deficient mice: evidence for spontaneous distal colitis Am J Physiol Gastrointest Liver Physiol, July 1, 2008; 295(1): G63 - G77. [Abstract] [Full Text] [PDF]

				M. Donowitz and X. Li Regulatory Binding Partners and Complexes of NHE3 Physiol Rev, July 1, 2007; 87(3): 825 - 872. [Abstract] [Full Text] [PDF]

				Md. R. Amin, P. K. Dudeja, K. Ramaswamy, and J. Malakooti Involvement of Sp1 and Sp3 in differential regulation of human NHE3 promoter activity by sodium butyrate and IFN-{gamma}/TNF-{alpha} Am J Physiol Gastrointest Liver Physiol, July 1, 2007; 293(1): G374 - G382. [Abstract] [Full Text] [PDF]

				S A Keilbaugh, M E Shin, R F Banchereau, L D McVay, N Boyko, D Artis, J J Cebra, and G D Wu Activation of RegIII{beta}/{gamma} and interferon {gamma} expression in the intestinal tract of SCID mice: an innate response to bacterial colonisation of the gut Gut, May 1, 2005; 54(5): 623 - 629. [Abstract] [Full Text] [PDF]

				M. Ohsugi, C. Cras-Meneur, Y. Zhou, E. Bernal-Mizrachi, J. D. Johnson, D. S. Luciani, K. S. Polonsky, and M. A. Permutt Reduced Expression of the Insulin Receptor in Mouse Insulinoma (MIN6) Cells Reveals Multiple Roles of Insulin Signaling in Gene Expression, Proliferation, Insulin Content, and Secretion J. Biol. Chem., February 11, 2005; 280(6): 4992 - 5003. [Abstract] [Full Text] [PDF]

				C. L. Brett, M. Donowitz, and R. Rao Evolutionary origins of eukaryotic sodium/proton exchangers Am J Physiol Cell Physiol, February 1, 2005; 288(2): C223 - C239. [Abstract] [Full Text] [PDF]

				L. R. Gawenis, H. Hut, A. G. M. Bot, G. E. Shull, H. R. de Jonge, X. Stien, M. L. Miller, and L. L. Clarke Electroneutral sodium absorption and electrogenic anion secretion across murine small intestine are regulated in parallel Am J Physiol Gastrointest Liver Physiol, December 1, 2004; 287(6): G1140 - G1149. [Abstract] [Full Text] [PDF]

				M. Ohsugi, C. Cras-Meneur, Y. Zhou, W. Warren, E. Bernal-Mizrachi, and M. A. Permutt Glucose and Insulin Treatment of Insulinoma Cells Results in Transcriptional Regulation of a Common Set of Genes Diabetes, June 1, 2004; 53(6): 1496 - 1508. [Abstract] [Full Text] [PDF]

				L. S. Bertelsen, L. Eckmann, and K. E. Barrett Prolonged interferon-{gamma} exposure decreases ion transport, NKCC1, and Na+-K+-ATPase expression in human intestinal xenografts in vivo Am J Physiol Gastrointest Liver Physiol, January 1, 2004; 286(1): G157 - G165. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 277/50/49036    most recent M205288200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Woo, A. L. Articles by Shull, G. E. Search for Related Content PubMed PubMed Citation Articles by Woo, A. L. Articles by Shull, G. E. Social Bookmarking                      What's this?