November 27, 2004

Inhibition of Hedgehog Signaling Protects Adult Mice From Diet- Induced Weight Gain1


Hedgehog (Hh) signaling plays an important role in embryonic development of many tissues, including the gastrointestinal tract. Sonic Hh-and Indian Hh-deficient mice die before or soon after birth, precluding further study of this signaling pathway in the mature intestine. Maternal transfer of inactivating monoclonal antibodies to Hh proteins (anti-Hh moAb) during late stages of embryogenesis or to early postnatal mice produced intestinal villous abnormalities, progressive runting, and severe malabsorption of dietary fat. In the present study, we sought to determine the effect of inhibiting Hh signaling on weight gain and lipid absorption in adult mice. Anti-Hh moAb was administered to adult Balb/c mice fed either a low-fat, nonpurified diet or a high-fat, semipurified diet, and to adult ob/ob mice fed the low-fat, nonpurified diet. Weight gain was significantly inhibited by anti-Hh moAb treatment in Balb/ C mice fed the high-fat, but not the low-fat diet and in ob/ob mice. Further analysis of adult Balb/c mice fed the high-fat diet demonstrated that although total lipid absorption was normal, the rate of triglyceride absorption was significantly delayed in mice treated with anti-Hh moAb and they had significantly increased fecal FFA excretion. Hepatic steatosis, found in high-fat fed Balb/c mice treated with the control moAb, was abrogated by anti-Hh moAb administration. These findings point to a potential role for Hh signaling pathways in diet-induced abnormalities of lipid metabolism. J. Nutr. 134: 2979-2984, 2004.

KEY WORDS: * hedgehog signaling * intestinal triglyceride absorption * hepatic steatosis

The mammalian Hedgehog (Hh)3 family of proteins include Sonic Hh, Indian Hh, and Desert Hh. These proteins are produced, modified, and secreted from epithelial cells. They bind to receptors such as Patched on stromal cells, and activate a signaling cascade by inactivating the tonic inhibition of Patched on a second membrane receptor, Smoothened. Freeing Smoothened activates a complex signaling pathway, which results in the translocation of the Gli family of transcription factors into the nucleus and regulation of the expression of many downstream genes [reviewed in (1-3)].

Hh signaling plays an important role in the embryonic development of many tissues, including the gastrointestinal tract (4,5). Because Sonic Hh- and Indian Hh-deficient mice die at or shortly after birth (6-8), we previously used maternal transfer of inactivating monoclonal antibodies for Hh proteins (anti-Hh moAb) to study the role of Hh signaling in the gastrointestinal tract during late stages of embryogenesis and in early postnatal mice (9). We found that disruption of Hh signaling produced widespread alterations in villous morphology and resulted in alterations in lipid metabolism throughout the small intestine. These mice exhibited progressive runting and died before weaning due to severe malabsorption of dietary fat. In addition, the mice abnormally accumulated neutral lipid in enterocytes, suggesting a block in enterocytic lipid trafficking.

Although Sonic Hh and Indian Hh signaling genes are expressed in adult mice in many tissues including the intestine and liver (10), the role of Hh signaling in adult mice is unclear. In this study, we administered an inactivating anti-Hh moAb to adult mice to further define the role of Hh signaling in intestinal triglyceride absorption. Because dietary triglycerides are a major source of energy, we sought to determine whether inactivating Hh signaling protects adult mice from diet-induced weight gain. We also investigated whether inactivating Hh signaling protects adult mice from diet-induced hepatic steatosis.


Mice, diets and antibodies. Balb/cJ and ob/ob (C57BL/6 background) mice were from the Jackson Laboratory. All mice were housed in a patbogen-free barrier facility (12-h light:dark cycle) and fed either a low-fat, nonpurifted diet containing 4.5, 20.0, and 36.8 g/100 g from fat, protein, and carbohydrate, respectively (Picolab 20, Ralston-Purina), or a semipurified, high-fat diet containing 21, 19.5, and 49.1 g/100 g from fat, protein, and carbohydrate, respectively (Adjusted Calories Diet TD 88137, Harlan Teklad). The source of fat in the high-fat diet was anhydrous milk fat, the source of protein was casein, and the sources of carbohydrates were sucrose and cornstarch (34-1 and 15.0 g/100 g, respectively). In addition, the high-fat diet contained 0.15% cholesterol. Food intake and fecal collections were determined by housing mice individually in metabolic cages. All animal protocols were approved by the Washington University Animal Studies Committee and Biogen's Committee. The generation of anti-Hh moAb 5E1 and isotype-matched control moAb 1E6 was described previously (11). The anti-Hh moAb binds to all 3 Hh family members, including Sonic Hh, Indian Hh, and Desert Hh, and blocks their function as previously validated (11).

Experimental design. Three separate mouse studies were performed. In study 1, 3-wk-old Balb/cJ mice were fed a low-fat, nonpurified diet and injected i.p. 3 times/wk with either the control or an anti- Hh moAb (8 mg/kg body weight) for 23 wk (n = 5 mice). In study 2, 7- wk-old Balb/cJ mice were fed a high-fat diet and injected as above for 18 wk (n = 10 mice). In study 3, 9-wk-old ob/ob mice were fed a low-fat, nonpurified diet and injected as above for 6 wk (n = 5 mice). Body weights were recorded weekly. At the end of the high- fat feeding study with ad libitum consumption, mice were exsanguinated and tissues removed and stored at - 80C until analysis. Intestinal epithelial and stromal cells were isolated as previously described (12).

Microarray analysis and mRNA quantitation by real time Q-PCR. Total RNA was extracted from tissues or cells using Trizol reagent (Invitrogen). For the microarray analysis, equal amounts of total RNA from isolated enterocytes of 6 mice treated with either control or anti-Hh moAb after 18 wk of high-fat feeding were pooled. RNAs were hybridized using a Mouse NIA20K cDNA microarray (Vanderbilt Microarray Shared Resource). Target generation and microarray hybridization were conducted using Genisphere 3DNA labeling method (Genisphere). Array images were scanned on an Axon scanner. Array data were extracted and analyzed using GenePix Pro 4.1 software (Axon Instruments). mRNA transcripts demonstrating changes (+ or -2- fold differences) were confirmed by real-time PCR. cDNA was generated using Superscript II Reverse Transcriptase (Invitrogen). Real-time PCR reactions were performed on SDS 7000 (Applied Biosystems) using 2X Sybr Green Master Mix (Applied Biosystems) as directed by the manufacturer. Relative gene expression was determined using the comparative Ct method (User Bulletin #2, Applied Biosystems). PCR primers used were as follows (5'->3'): apolipoprotein (apo) AIV, CAA TGC CAA GGA GGC TGT AGA and AGT TTG TCC TTG AAG AGG GTA CTG A apo B, TGA ATG CAC GGG CAA TGA and GGC ATT ACT TGT TCC ATG GTT CT; diacylglycerol acyltransferase 1, TCC GCC TCT GGG CAT TC and GAA TCG GCC CAC AAT CCA; diacylglycerol acyltransferase 2, AGA ACC GCA AAG GCT TTG TG and AGG AAT AAG TGG GAA CCA GAT CAG; fatty acid synthetase, GGC ATC ATT GGG CAC TCC TT and GCT GCA AGC ACA GCC TCT CT; 3-hydroxy-3-methyl-glutaryl (HMG) CoA synthetase HMGS, TGG TGG ATG GGA AGC TGT CTA and TTC TTG CGG TAG GCT GCA TAG; microsomal triglyceride transfer protein, AAG ACA GCG TGG GCT ACA AAA and TCA TCA TCA CCA TCA GGA TTC C; uncoupling protein 2, TCA CTG TGC CCT TAC CAT GCT and AGG CAT GAA CCC CTT GTA GAA G; bone morphogenctic protein 4, TGG GCT GGA ATG ATT GGA TT and CAG TCC CCA TGG CAG TAG AAG; Gli 1 GCT TGG ATG AAG GAC CTT GTG and GCT GAT CCA GCC TAA GGT TCT C; Gli 2, GGC AGC TTG CAT CTT GAA G and AAA AAG CTC TGA AAA CTC GTC CAT; Gli 3, CCA GCC GAA AAC GTA CAC TGT and GGG ATG TTC TTA TCA TGG TCT GAA; Patched, CCT GCA AAC CAT GTT CCA GTT and TCG TAG CCC CTG AAG TGT TCA; Patched 2, CCA GGC TGC ATT TGA CCA A and TGG GCA TTC CCG GTT TG; Smoothened, GAG GGT GGC CTG ACT TTC TG and GCT GTT GAA CTT GAT GTT TTG TAC CT; 18S, CGG CTA CCA CAT CCA AGG AA and GCT GGA ATT ACC GCG GCT.

Serum, fecal, and tissue analyses. For study 2, serum FFA, total cholesterol, and triglyceride concentrations were determined using NEFA C, Cholesterol E, and L-type TG-H kits, respectively (Wako Chemicals). Feces were collected daily from individual mice housed in metabolic cages for 5 d. Fecal lipid analysis was performed as previously described (13). In addition, an aliquot of the lipid extract was dried under nitrogen and resuspended in 2% Triton X- 100. Fecal FFA, total cholesterol, and triglyceride concentrations were determined enzymatically as described above. Tissue lipid analysis of liver and enterocytes was performed using enzymatic methods as previously described (14). Hematoxylin and eosin staining was performed on formalin-fixed liver, intestine, and stomach sections. Oil Red O staining was performed on frozen sections of liver and intestine (9). Protein homogenates were prepared from liver and intestine in the presence of protease inhibitors (Complete- Mini tablet (Roche Diagnostics). Proteins were separated by SDS- PAGEand immunoblotted using the indicated antibody as previously described (15) Anti-Hsp40 IgG was purchased from StressGen Biotechnologies and anti-albumin antiserum from ICN.

Intestinal triglyceride absorption in vivo. On wk 16 of study 2, control and anti-Hh moAb treated mice were food-deprived for 16 h, anesthetized, weighed, and injected i.v. with 20 mg (500 mg/kg body weight) Triton WR1399 (Tyloxapol, Sigma) in 100 L saline. Mice received an i.g. bolus of 200 L corn oil. Blood was collected at 0, 1, 2, 3, and 4 h, and triglyceride concentration determined enzymatically as described above.

Statistical analyses. Values in the text are means SD or SEM, as indicated. Means were compared with a Student's t test and differences were considered significant at P


Inhibition of Hh signaling protects Balb/c mice fed a high-fat diet and ob/ob mice fed a low-fat diet from body weight gain. Inactivating Hh signaling in adult mice fed the low-fat, nonpurified diet did not affect body weight gain (Fig. 1A). However, inactivating Hh signaling protected adult mice from diet-induced weight gain (Fig. 1B). The percentage body weight changes did not differ between control and anti-Hh moAb-treated mice fed the high- fat diet for the first 12 wk of the study. After 12 wk, the body weights of anti-Hh moAb-treated mice reached a plateau, whereas the body weights of the control moAb-treated mice continued to increase (Fig. 1B). Based on the results showing an abrogation of the weight gain induced by the high-fat diet, we examined whether inactivating Hh signaling protected adult mice from a genetic form of obesity, using leptin-deficient (ob/ob) mice. Anti-Hh moAb treatment of ob/ ob mice fed the low-fat diet decreased weight gain compared with controls. This decreased weight gain persisted throughout the 6 wk of treatment (Fig. 2).

To understand the mechanisms underlying the protection from the weight gain induced by the high-fat diet, we undertook a more detailed examination of these mice. Inactivating Hh signaling in adult mice fed the high-fat diet did not affect serum lipids. Serum triglyceride, cholesterol, and FFA concentrations were similar in control (n = 8) and anti-Hh moAb-treated (n = 9) mice fed the high- fat diet after 18 wk of treatment (4.2 2.3 vs. 3.6 2.3 mmol/L, P = 0.61; 2.0 0.34 vs. 2.0 0.42 mmol/L, P = 0.92; 0.44 0.14 vs. 0.42 0.13 mmol/L, P = 0.75, respectively).

The rate of intestinal triglyceride absorption is delayed in anti- Hh moAb- treated mice fed the high-fat diet. Differences in growth rates between control and anti-Hh moAb-treated mice could not be accounted for by differences in food intake or fecal output (Fig. 3A and B). Surprisingly, the protection from diet-induced weight gain was not due to dramatic decreases in mass of lipid absorbed, such as occurs in neonatal and early postnatal mice treated with anti-Hh moAb (9). Nevertheless, there was a significantly greater recovery of total fecal lipid in anti-Hh moAh-treated mice compared with control moAb-treated mice (Fig. 3B). To further characterize this result, we measured fecal triglyceride, cholesterol, and FFA excretions. Fecal cholesterol or triglyceride excretion did not increase (Fig. 3C); however, FFA excretion increased significantly (~7-fold; Fig. 3C). Despite these differences in fecal FFA and total lipid excretions, the triglyceride concentration of isolated enterocytes did not differ between control and anti-Hh moAb-treated mice (242.3 216.2 vs. 250.8 180.8 g/mE protein, P = 0.94.

FIGURE 1 Growth response of mice fed the low-fat, nonpurified diet (A) or the high-fat, semipurified diet (B), treated with either control or anti-Hh moAb (A). Values are means SEM, n = 5 (low-fat diet) or 10 (high-fat diet) mice. *Different from control, P

FIGURE 2 Growth response of ob/ob mice fed the low-fat, non- purified diet, treated with either control or anti-Hh moAb. Values are means SEM, n = 5 mice. *Different from control, P

FIGURE 3 Food, triglyceride, and cholesterol intake (A), total fecal and lipid excretion (B), and fecal triglyceride, cholesterol, and FFA excretion (C) in mice fed the high-fat, semipurified diet and treated with either control or anti-Hh moAb. Values are means SD, n = 5 mice. *Different from control, P

To extend these findings, we examined intestinal triglyceride absorption after bolus administration. The rate of triglyceride absorption was delayed in adult mice by inactivation of Hh signaling (Fig. 4A). Specifically, the slope of the line for the rate in anti- Hh moAb-treated mice indicated a significant reduction in the rate of appearance of triglyceride compared with control moAb-treated mice (Fig. 4A). In addition, serum apo AIV, a marker of chylomicron secretion, was 30% lower in anti-Hh than control moAb-treated mice, whereas serum apo AI levels were similar in the 2 groups (Fig. 4B). In addition, consistent with a reduced rate of intestinal triglyceride absorption and secretion, enterocyte apo AIV abundance was increased 70% in anti-Hh compared with control moAb-treated mice (Fig. 4C).

Intestinal gene expression. To gain a better understanding of the effect of inactivating Hh signaling at the cellular level, we undertook a microarray survey to examine mRNA expression of several candidate lipid metabolism genes in isolated enterocytes and also of Hh signaling genes in isolated intestinal stromal cells (data not shown). To confirm these results we performed real-time PCR (Table 1). There were alterations in lipid metabolism pathways, i.e., a significant, 3-fold increase in the expression level of HMG CoA synthetase, a gene involved in cholesterol biosynthesis, in epithelial cells from anti-Hh compared with control moAb-treated mice. However, there were no other significant changes in mRNA expression of lipid metabolism genes in isolated intestinal epithelial cells. These included apo B and microsomal triglyceride transfer protein, genes whose regulation and expression is tightly linked to intestinal lipoprotein assembly and secretion. To verify the anticipated disruption of Hh signaling in mice treated with anti- Hh moAb, we established that the expression of Gli1 and patched 2 mRNAs were only 10 and 20% of control, respectively, in isolated intestinal stromal cells.

FIGURE 4 Rate of triglyceride absorption (A), protein expression in liver (B) and intestine (C) for apo AIV, apo Al, albumin, and Hsp40 in mice fed the high-fat, nonpurified diet and treated with either control or anti-Hh moAb. Values are means SEM, n = 5 (triglyceride absorption) or 6 (protein expression) mice. *Different from control, P


Gene expression in enterocytes and intestinal stromal cells of control and anti-Hh moAb-treated mice fed a high-fat diet1

Anti-Hh moAb-treated mice are protected from diet-induced hepatic steatosis. Feeding a high-fat diet resulted in the abnormal accumulation of hepatic lipid, as evidenced by abundant neutral lipid staining in liver sections from moAb-treated control mice fed the high-fat diet (Fig. 5A). In contrast, less oil Red O staining was evident in liver sections from anti-Hh moAb-treated mice (Fig. 5B). These morphological changes were demonstrated biochemically by a decrease in hepatic triglyceride and cholesterol concentrations in anti-Hh moAb-treated mice (Fig. 5C).

To gain a better understanding of the effect of inactivating Hh signaling in the liver, we examined the mRNA abundance of lipid metabolism and Hh signaling genes in the liver by real-time PCR (Table 2). Although the expression of several of the lipid metabolism genes tended to be reduced by Hh signaling inactivation, there were no significant changes in any of the candidates examined. Verification of the interruption of Hh signaling was again demonstrated by the marked inhibition in the expression of 2 downstream genes; Gli 1 and bone morphogenetic protein 4 mRNAs were only 20 and 50% of control, respectively. Of note, Gli3 expression was markedly increased.


The Hh signaling pathway plays a critical role in gut morphogenesis and has well-established functions in embryogenesis (3). However, the importance and function of Hh signaling in the mature intestinal tract are unclear. Our previous studies (9) showed that administration of an anti-Hh moAb in utero or during early postnatal life inhibited intestinal fat absorption as manifested by steatorrhea, increased lipid accumulation in enterocytes, distortion of the villi, and decreased serum apo AIV levels, ultimately leading to severe runting and early postnatal death. In addition to the morphologic changes noted in the villous enterocytes, crypt cell proliferation was increased. An effect on gut proliferation was noted in other studies, e.g., of the stomach epithelium, and colon (16,17). For example, administration of cyclopamine, an inhibitor of Hh signaling, increased gastric epithelial cell proliferation (17). On the basis of those studies, we hypothesized that Hh signaling might play an important role in regulating fat absorption, perhaps associated with alterations in epithelial cell proliferation in the mature intestinal epithelium. This is a particularly relevant target because the small intestinal epithelium is a continually renewing structure with a defined program of crypt to villus maturation and turnover.

FIGURE 5 Representative sections from oil Red O staining (400) of neutral lipid in liver sections (A) and biochemical analysis of triglyceride and cholesterol concentrations in livers (B) of mice fed the high-fat, semipurified diet and treated with either control or anti-Hh moab. Values are means SD, n = 6 mice. *Different from control, P


Gene expression in liver of control and anti-Hh moAb-treated mice fed a high-fat diet1

We showed that administration of an anti-Hh moAb was effective in blocking Hh signaling in mature mouse intestine and liv\er, as determined by examining downstream gene expression (Tables 1 and 2), and by the presence of intestinal metaplasia in. the stomachs of anti-Hh moAb-treated mice (data not shown). Expression of Gli1, a transcriptional activator of Hh signaling was decreased, whereas Gli3, which can serve an activating or repressor function, was notably increased in the liver. This increase was likely a compensatory response to the inhibition of Hh and Gli 1 signaling. The functional counterpart accompanying interruption of Hh signaling included inhibition of weight gain in 2 mouse models. Mice treated with the anti-Hh moAb gained weight similarly to control moAb- treated mice until they were fully mature (i.e., at ~12 wk of age, when skeletal growth ceases), but then were protected from further increases in weight when fed a high-fat diet. Also, 9-wk-old leptin- deficient mice treated with the anti-Hh moAb gained weight in parallel with controls until ~12 wk of age, but further (abnormal) weight gain was inhibited in these mice.

Several observations suggest that blocking Hh signaling results in protection from weight gain in part by affecting lipid absorption. Although there was no change in total fat absorption in adult mice [unlike in prenatal or early postnatal mice (9)], there was a significant delay in the rate of triglyceride absorption in mice treated with anti-Hh moAb (Fig. 4A), accompanied by decreased serum and increased enterocytic apo AIV levels. The increase in fecal FFA in anti-Hh moAb-treated mice, although insufficient to fully explain the protection from weight gain, also suggests that an alteration in triglyceride absorption occurred. We speculate that the increased fecal excretion of FFA is derived from the action of bacterial lipases on residual luminal triglyceride. However, we were unable to provide a unifying mechanism for the apparent delay in intestinal triglyceride absorption because the candidate genes we examined revealed comparable mRNA abundance. It is possible that alterations in protein expression or compartmentalization may account for the subtle defect observed, but such speculation will require formal examination.

Blocking Hh signaling may affect weight gain by other mechanisms, such as decreasing absorption of nutrients other than fat, increasing total energy expenditure, or decreasing lean body mass. However, mice treated with anti-Hh moAb gave no evidence of generalized malabsorption. Fecal weights and food intake were unchanged, and the mice appeared otherwise normal. Although it is possible that blocking Hh signaling increases total energy expenditure, thus leading to protection from weight gain, normal weight gain was not diminished in anti-Hh moAb-treated mice consuming a low-fat, nonpurified diet. However, further experiments are required to rule out the possibility of Hh-specific alterations in energy expenditure that may be relevant to models of obesity.

A role for Hh signaling in weight regulation was shown by Makino et al. (18), who described a spontaneous mouse mutant mes, in which there is a deletion of the most C-terminal cytoplasmic domain of Patched. This deletion results in constitutively activated Hh signaling because the mutated Patched can no longer inhibit Smoothened. Mes mice demonstrate increased body weight due to enhanced proliferation of mesenchymal cells of the trunk, in the neural tube, esophagus, and aorta. Thus, these data suggest that Patched is a negative regulator of body weight. Others showed that intestinal epithelial cancers and epithelial cancer cell lines demonstrate activated Hh signaling and increased proliferation (19,20). In our experiments, crypt cell proliferation rates in the intestines of mice treated with anti-Hh moAb were lower than those in the control moAb-treated mice (data not shown). Determining whether there is a more general decrease in cellular proliferation affecting multiple tissues will require further analysis.

An effect on lipid trafficking induced by anti-Hh moAb treatment was also supported by our finding that blocking Hh signaling protected mice fed a high-fat diet from hepatic steatosis. We recognize that the reduced steatosis may have resulted from direct effects of blocking Hh signaling on the liver, or may be related to the lag in triglyceride absorption; it was not possible to resolve these possibilities with the current design. In addition, triglyceride uptake and metabolism in other tissues may have also been affected by blocking Hh signaling, thus contributing to the protection from weight gain. Nevertheless, hepatic steatosis is a serious complication of obesity and other metabolic disorders (21,22); our data suggest that inhibition of Hh signaling may provide novel targets for therapies designed to prevent hepatic steatosis.


We thank R. Ncwberry and K. McDonald for assistance with intestinal epithelial and stromal cell isolation; F. Nassir and E. Newberry for helpful discussions; and R. Shapiro and F. R. Taylor for making the 5E1 moAb.

1 Supported by National Institutes of Health grants HL-38180, DK- 56260 (N.O.D.), DK61216, DK46122 (D.C.R.), DK50466 (M.S.L.), and an NIH postdoctoral fellowship DK 07130 (K.K.B.). We acknowledge the DDRCC (DK52574) Morphology and Genomics Cores and the CNRU (DK56341) Analytical Core.

3 Abbreviations used: anti-Hh moAb, anti-Hedgehog monoclonal antibody; apo, apolipoprotein; Hh, Hedgehog; HMG, 3-hydroxy-3- methyl-glutaryl; moAb, monoclonal antibody.


1. Nybakken, K. & Perrimon, N. (2002) Hedgehog signal transduction: recent findings. Curr. Opin. Genet. Dev. 12: 503-511.

2. Ingham, P. W. & McMahon, A. P. (2001) Hedgehog signaling in animal development: paradigms and principles. Genes Dev. 15: 3059- 3087.

3. Cohen, M. M., Jr. (2003) The hedgehog signaling network. Am. J. Med. Genet. 123A: 5-28.

4. Ramalho-Santos, M., Melton, D. A. & McMahon, A. P. (2000) Hedge-hog signals regulate multiple aspects of gastrointestinal development. Development 127: 2763-2772.

5. Harmon, E. B., Ko, A. H. & Kim, S. K. (2002) Hedgehog signaling in gastrointestinal development and disease. Curr. Mol. Med. 2: 67-82.

6. St-Jacques, B., Dassule, H. R., Karavanova, L, Botchkarev, V. A., Li, J., Danielian, P. S., McMahon, J. A., Lewis, P. M., Paus, R. & McMahon, A. P. (1998) Sonic hedgehog signaling is essential for hair development. Curr. Biol. 8: 1058-1068.

7. St-Jacques, B., Hammerschmidt, M. & McMahon, A. P. (1999) Indian hedgehog signaling regulates proliferation and differentiation of chondrocytes and is essential for bone formation. Genes Dev. 13: 2072-2086.

8. Chiang, C., Litingtung, Y., Lee, E., Young, K. E., Corden, J. L., Westphal, H. & Beachy, P. A. (1996) Cyclopia and defective axial patterning in mice lacking Sonic hedgehog gene function. Nature (Lond.) 383: 407-413.

9. Wang, L. C., Nassir, F., Liu, Z. Y., Ling, L., Kuo, F., Crowell, T., Olson, D., Davidson, N. O. & Burkly, L. C. (2002) Disruption of hedgehog signaling reveals a novel role in intestinal morphogenesis and intestinal-specific lipid metabolism in mice. Gastroenterology 122: 469-482.

10. Motoyama, J., Heng, H., Crackower, M. A., Takabatake, T., Takeshima, K., Tsui, L C. & Hui, C. (1998) Overlapping and non- overlapping Ptch2 expression with Shh during mouse embryogenesis. Mech. Dev. 78: 81-84.

11. Wang, L. C., Liu, 2. Y., Gambardella, L., Delacour, A., Shapiro, R., Yang, J., Sizing, I., Rayhorn, P., Garber, E. A., Benjamin, C. D., Williams, K. P., Taylor, F. R., Barrandon, Y., Ling, L. & Burkly, L. C. (2000) Regular articles: conditional disruption of hedgehog signaling pathway defines its critical role in hair development and regeneration. J. Investig. Dermatol. 114: 901-908.

12. Newberry, R. D., Stenson, W. F. & Lorenz, R. G. (1999) Cyclooxygenase-2-dependent arachidonic acid metabolites are essential modulators of the intestinal immune response to dietary antigen. Nat. Med. 5: 900-906.

13. Schwarz, M., Lund, E. G., Setchell, K. D., Kayden, H. J., Zerwekh, J. E., Bjorkhem, I., Herz, J. & Russell, D. W. (1996) Disruption of cholesterol 7alpha-hydroxylase gene in mice. II. Bile acid deficiency is overcome by induction of oxysterol 7alpha- hydroxylase. J. Biol. Chem. 271: 18024-18031.

14. Carr, T. P., Andresen, C. J. & Rudel, L. L. (1993) Enzymatic determination of triglyceride, free cholesterol, and total cholesterol in tissue lipid extracts. Clin. Biochem. 26: 39-42.

15. Xie, Y., Nassir, F., Luo, J., Buhman, K. & Davidson, N. O. (2003) Intestinal lipoprotein assembly in apobec-1 -/- mice reveals subtle alterations in triglyceride secretion coupled with a shift to larger lipoproteins. Am. J. Physiol. 285: G735-G746.

16. Van Den Brink, G. R., Bleuming, S. A., Hardwick, J. C., Schepman, B. L., Offerhaus, G. J., Keller, J. J., Nielsen, C., Gaffield, W., Van Deventer, S. J., Roberts, D. J. & Peppelenbosch, M. P. (2004) Indian Hedgehog is an antagonist of Wnt signaling in colonie epithelial cell differentiation. Nat. Genet. 36: 277-282.

17. Van Den Brink, G. R., Hardwick, J. C., Nielsen, C., Xu, C., Ten Kate, F. J., Glickman, J., Van Deventer, S. J., Roberts, D. J. & Peppelenbosch, M. P. (2003) Sonic hedgehog expression correlates with fundic gland differentiation in the adult gastrointestinal tract. Mol. Pathol. 56: 150-155.

18. Makino, S., Masuya, H., Ishijima, J., Yada, Y. & Shiroishi, T. (2001) A spontaneous mouse mutation, mesenchymal dysplasia (mes), is caused by a deletion of the most C-terminal cytoplasmic domain of patched (ptc). Dev. Biol. 239: 95-106.

19. Thayer, S. P., di Magliano, M. P., Heiser, P. W., Nielsen, C. M., Roberts, D. J., Lauwers, G. Y., Qi, Y. P., Gysin, S., Fernandez- del Castillo, C., Yajnik, V., Antoniu, B., McMahon, M., Warshaw, A. L. & Hebrok, M. (2003) Hedgehog is an early and late mediator of pancreatic cancer tumorigenesis. Nature (Lond.) 425: 851-856.

20. Berman, D. M., Karhadkar, S. S., Maitra, A., Montes De Oca, R., Gerstenblith, M. R., Br\iggs, K., Parker, A. R., Shimada, Y., Eshleman, J. R., Watkins, D. N. & Beachy, P. A. (2003) Widespread requirement for Hedgehog ligand stimulation in growth of digestive tract tumours. Nature (Lond.) 425: 846-851.

21. Diehl, A. M. (2002) Nonalcoholic steatosis and steatohepatitis IV. Nonalcoholic fatty liver disease abnormalities in macrophage function and cytokines. Am. J. Physiol. 282: G1-G5.

22. Marchesini, G., Bugianesi, E., Forlani, G., Cerrelli, F., Lenzi, M., Manini, R., Natale, S., Vanni, E., Villanova, N., Melchionda, N. & Rizzetto, M. (2003) Non-alcoholic fatty liver, steatohepatitis, and the metabolic syndrome. Hepatology 37: 917- 923.

Kimberly K. Buhman,* Li Chun Wang,[double dagger] Yuzhu Tang,* Elzbieta A. Swietlicki,* Susan Kennedy,* Yan Xie,* Zhong-Yi Liu,* Linda C. Burkly,[double dagger] Marc S. Levin,* ** Deborah C. Rubin,* and Nicholas O. Davidson*[dagger]2

Departments of * Internal Medicine, and [dagger] Pharmacology and Molecular Biology, Washington University School of Medicine, St. Louis, MO 63110; ** Specialty Care Service Line, St. Louis VA Medical Center, St. Louis, MO; and [double dagger] Department of Exploratory Science, Biogen, Cambridge, MA

2 To whom correspondence should be addressed. E-mail: [email protected]

Copyright American Institute of Nutrition Nov 2004