as cell bodies in the myentenc plexus showed faint staining. This was consistent with the staining pattern observed in the distal colon in which staining was also detected in submucosal plexus cell bodies. In fundus circular muscle, rat NmU-23 (0.1 nM- 0.1 g,M) had no effect on elecmcafiy-evoked, nerve-mediated contractile responses but concentration-dependently increased nerve-mediated contractile responses in distal colon preparations, ECso ~ 15 riM; n=6-8 (stomach) and n=3-4 (colon), each concentration. Rat NmU-23 (0.1 nM- 0.1 g-M) concentration-dependently contracted both mouse gastric fundns (ECs0 ~0.08 g.M) and distal colon smooth muscle, EC~0 ~ 1 p,M; n = 12 (stomach) and n = 4 (colon), each concentration. CONCLUSIONS: NmU potently contracts mouse gastric fundns but not colon smooth muscle and has an ability to exert prokinetic-like activity in colon but not gastric fundus preparations. Based on localisation studies, these effects may be mediated via NmU-R2, although NmU- Rl-mediated effects cannot be excluded.

$1029

Endothelin Causes Contraction of Human Esophageal Muscularis Mucosae through Interaction with Both ETA and ETB Receptors Shih-Che Huang, Bee-Song Chang

Endothelin (ET) causes contraction of the museularis mucosae of the guinea pig esophagus, but its role in human esophagus remains unknown. To investigate effects of ET in human esophagus, we measured contraction of human esophageal muscularis mucosae strips caused by ET related peptides and binding of ~251-ET-I to cell membranes prepared from the human esophageal muscularis mucosae. Isolated muscularis mucosae strips from the distal third of human esophagus were obtained from patients undergoing esophagectomy for cancer. Autoradiography demonstrated ~:SI-ET-1 binding to the museularis mucosae. ET-1 caused tetrodotoxin and atropine-insensitive contraction of muscularis mucusae strips. In terms of the mammal tension of contraction, ET-1 and ET-2 were equal in efficacy. The efficacies for ET-1 and ET-2 were 52 "4" 8% and 46 +- 9% of the tension caused by i g-M carbacbol, respectively. The relative potencies for ET isopeptides to cause contraction were ET-1 (EC50 = 11 +_ 2 n M ) = ET-2 (ECSO = 23 -+ 9 nM) > ET-3 (EC50 = 43 +_ 27 riM) > sarafotoxin $6c (SX6c) (EC50 = 153 _+ 73 nM), an ETB receptor agonist. ET-1 caused contraction was only slightly inhibited by BQ-123 (potent ETA receptor antagonist) and not by BQ-788 (potent ETB receptor antagonist). It was moderately inhibited by the combina- tion of both (EC50 of ET-1 plus BQ-123 and BQ-788 = 67 -+ 22 nM, p < 0.05, compared with ET-1 alone), indicating synergistic inhibition. Furthermore, desensitization to SX6c with SX6c pre-treatment failed to abolish the response to ET- 1. These indicate the involvement of both ETA and ETB receptors in the contraction. Binding of 12~I-ET-1 to cell membranes of the mnscularis mucosae was saturable and specific. Analysis of dose-inhibition curves demonstrated the presence of ETA and ETB receptors. This study demonstrates that, the mnscularis mucosae of the human esophagus, similar to that of the guinea pig esophagus, possesses both ETA and ETB receptors mediating muscle contraction with cooperation, g,

$1030

The nuclear Estrogen-Related Receptora (ERRoO modulates intestinal fat metabolism in mice Julie Carrier, Vincent Glguere

ERRa orphan receptor is a member of the nuclear receptor superfamily initially identified in a search for genes related to the estrogen receptor. In mice, ERRc~ is expressed at high levels during mid- to late embryonic intestinal development and at moderate levels in adult intestinal epithelial cells. ERRa is also expressed in tissues which preferentially metabolize fatty acids and has been shown to regulate PGC-lu (PPAR Gamma Coactivator-1) target genes involved in mitochondrial metabolism. Preliminary phenotypic analysis of ERRa null mice has revealed a growth retardation as well as a reduced body fat content. The aim of this study is to identify possible physiologic functions for ERRs in the intestine. METHODS: Microarray analysis was performed using paired RNA isolated from pooled ERRs -/- and wild-type mice ileum(chips prepared using the 7K mouse Array-Ready Ofigo set from Operon). Three microarray profiles were performed using 129/Sv mice, female C57B1/6 mice and male C57B1/6 mice. For each experiment, we selected genes with a relative fold change outside one standard deviation from the mean ratio. Real-time RT-PCR was performed to verify the differential expression of a subset of genes. In addition, fecal fat content was measured in 10 day old pups. RESULTS: ERRs -/- mice do not have an overt intestinal phenotype and histological analysis revealed normal mucosal structure. Microarray expression profiling in the 3 experiments consistently showed 2 genes upregulated and 15 genes downregnlated in the intestine of ERRs -/- mice in comparison to wild-type mice. A subset of downregulated genes are involved in oxidative processes: ATP synthase, cytochrome c oxidase, NADH dehydrogenase. A number of other downregulated genes are associated with fat digestion and absorption (ratio from Q-RT-PCR in the 3 sets of samples used for microarray: apoa4: -3, -25.9, - 10; fabpl : - 10, -38, -565; fabp2: -1.2, -1.4, -1.7 and pnliprp2: - 1.3, -7.1, -23). Although adult ERRs -/- mice do not have steatorrhea, 10 day old pups have 27% higher fat in colonic content compared to control pups (22.5% versus 17.7%, p<0.05). In addition, knockout pups weigh 11% less than their control fitermates (-/-: 5.51g, n=24; control: 6.16g, n = 33. p<0.05). We conclude that ERRs plays a role in intestinal lipid metabolism. Further studies are needed to evaluate whether differentially expressed genes are direct targets of ERRs or are modulated by metabolic events occurring secondary to ERRc~ deficiency.

$1031

The Spontaneous Dwarf Rat (SDR) has Selectively Reduced Apoptosis and Proliferation in Response to AOM Jason Collins, Young Choi, Jai Marchandani, Guy Caiafa, Stephen Swanson, Robert Carroll

Background: The SDR rat possesses a point mutation m the GH gene and produces negligible amounts of GH and reduced levels of IGF-1. We have previously shown a reduction in both aberrant crypt loci (ACF) and tumor number in the distal colon of SDR rats compared to Sprague-Dawley (WT) l i t termates exposed to the carcinogen AOM (Gas- tro2002:122:A460). To further evaluate the role of the GHAGF axis in early carcinogenesis we compared apoptosis, proliferation, and B-catenin expression in response to AOM in both animal genotypes. Methods: 6 rats of each genotype received 30 mg/kg of AOM and were sacrificed along with control animals at 6 and 72 hrs. Apoptosis and proliferation was quantified in the proximal and distal colon by TUNEL staining, BRDU labeling a nd flow cytometry. 1GF-1 was measured by RIA. IGF1-R, EGF-R, p27, p21 were assayed by Western blotting. Beta-catenin expression was qualitatively evaluated in rumor and colon specimens from AOM treated and control animals. Results: Apoptosis rates of AOM treated SDR animals were similar to WT in the proximal (6%) colon but reduced in the distal colon (3% vs 7%) Apoptosis was undetectable in control animals. Proliferation was significantly reduced (-60%) at 6hr from control animals in both groups, however the subsequent proliferative response in the distal colon at 72 hrs was 50% lower in the SDR animals. IGF-1 expression was 50% less in SDR animals but did not change in response to AOM. EGF-R expression declined from controls at 6 hrs in both gen otypes but increased 4 fold in the distal colon of WT animals. This was associated with a selective increase in 1GF-1R expression in distal colon greater in WT vs SDR animals, p21 expression increased at 6 hrs in SDR animals but became almost undectable in the WT rats. p27 expression was 2X higher in the proximal colon of both genotypes. Diffuse B-cateinn staining was seen in both WT and SDR tumors. Foci of diffuse staining crypts were idemihed in the proximal and distal colon of AOM treated animals of both groups but not in their respective controls. Conclusions: The distal colon of SDR animals have reduced levels of apoptosis and a blunted proliferative response to carcinogen. 1GF- 1 receptor induction may be an important proliferative response in malignant transformation of the colon. Reduced IGF levels may alter apoptotic injury through p21. p27 expression may dictate that only beta-catenin mutations in the distal colon subsequently progress to colon tumors.

$1032

Uroguanylin but Not Guanylin Knockout Mice Have Diminished Sodium Excretion in Response to an Enteral Salt Load Noeet Elitsur, John Lorenz, Phil Sanford, Jennifer Hawkins, Micheile Nieman, Alison Woo, Gary Shun, Mitchell B. Cohen

Background: E. cofi heat stable toxin (ST) is a worldwide cause of secretory diarrhea. Guanylin (GN) and uroguanylin (UGN) are homologous peptides produced in the mammalian intestine, which are thought to mimic the action of ST and modulate intestinal secretion without causing diarrhea. To explore the in vivo role of GN and UGN we created gene-targeted mice. In initial experiments we did not identify intestinal secretory defects in these knockout (KO) mice. However, these peptides also circulate; therefore, based on data suggesting that GN and UGN could he natriuretic, we hypothesized that urinary Na + excretion would be diminished when KO mice were placed on a high salt diet. Methods: To test this hypothesis we separately evaluated GN and UGN wildtype (WT), heterozygote (HET) and KO mice (n = 6-9 per genotype). Mice were housed in metabolic cages where urine and fecal matter were separated and then collected every 24 hrs. Mice were fed a 1% NaC1 (control) diet for 3 days (days 1-3) followed by a 5% NaCl diet for 3 days (days 4-6). Food and water intake as well as urine and stool excretion were measured daily. Urine and fecal Na + content were measured by flame photometry. Results: All groups of mice ate and drank similar amounts of food and water. Water consumption and urine output increased on the high salt diet in all groups. However, urine and total (urine + fecal) Na + excretion in the UGN KO mice was 33% less on the high Na + diet (days 4-5) compared to WT (p<O.03 on day 5) and did not approach control levels until the 6th study day (p = NS) (see Table). In contrast, Na + excretion in GN KO mice was similar to WT controls. Parallel experiments in anesthetized UGN KO mice also demonstrated diminished urinary Na + excretion of a rapid enteral but not an identical intravenous salt load. Conclusions: GN and UGN KO mice are healthy and do not manifest uncompensated intestinal secretory defects. However, UGN KO mice cannot adequately excrete an acute enteral Na + load. UGN but not GN appears to play a significant role in an enteric-renal communication axis for Na + excretion. Our data are also consistent with the recent identification of a renal receptor for UGN distinct from the receptor for ST and GN.

Total Na* Excretion (pmolu Na~

1% NaCI diet 5% NaCI diet GenM'ype Day1 Day2 Day3 Day4 Day5 Day6

UGN-WT 353 370 491 1763 1951 1749 UGN-HET 374 347 402 1231 1315 1854 UGN-KO 278 336 271 1189 1288 1409

A G A A b s t r a c t s A - 1 4 0