The dynamic process of intestinal cell turnover following massive bowel resection is a function of crypt cell proliferation, migration along the crypt–villus axis, enterocyte differentiation, and cell death via apoptosis. Growing evidence suggests that cell proliferation after bowel resection exceeds cell apoptosis leading to villus hypertrophy (1–5); however, the mechanisms of this effect are poorly understood. Various signaling cascades have been implicated in the control of stem cell activity, proliferation, lineage commitment, terminal differentiation, and cell survival during normal development and tissue regeneration of the gastrointestinal epithelium. These include Wnt/β-catenin, Hedgehog, bone morphogenetic protein and Notch signaling pathways (6). The roles of these signaling cascades in stimulation of cell proliferation after massive small bowel resection are unknown.
There is a growing body of evidence suggesting that various hormones and growth factors may stimulate post-resection adaptive hyperplasia and modify the intestinal absorption of nutrients in patients with short bowel syndrome (SBS) (7). The transforming growth factor-beta (TGF-β) superfamily consists of approximately 30 structurally related proteins. These include 3 isoforms of TGF-β itself, 3 forms of activin, and >20 bone morphogenetic proteins. These growth factors play an important role in embryonic development and control a broad range of cellular functions including cell proliferation, growth, differentiation, and apoptosis. They are also responsible for the production of extracellular matrix proteins (8). Members of the TGF-β family exert their effect by binding to heteromeric complexes of 2 different kinds of serine/threonine kinase receptors denoted type I and type II (9). Seven different type I receptors and 4 different type II receptors have been identified to date (10).
Dietary formulas containing TGF-β2 has been proven to have significant clinical utility in patients with Crohn disease by minimizing intestinal damage and facilitating regeneration after mucosal injury (11,12). We have demonstrated beneficial effects of TGF-β2-enriched diet on intestinal recovery following chemotherapy-induced mucositis in a rat (unpublished data). In another experiment, we have shown that dietary supplementation with TGF-β2 inhibits intestinal adaptation after massive small bowel resection in a rat (13). The mechanism of this negative effect is poorly understood. The present study is an extension of the previous experiment. The purpose of the present study was to evaluate whether the effect of TGF-β2 on enterocyte turnover (proliferation and apoptosis) is correlated with TGF-β2 receptor expression along the villus–crypt axis after bowel resection in a rat.
Cell Culture and TGF-β2 Treatment
The human colon adenocarcinoma cell line (CaCo-2) was cultured in Dulbecco modified Eagle's medium supplemented with 10% fetal calf serum, 1% glutamine, 25 mmol/L HEPES buffer, 1% penicillin, and streptomycin and was incubated in a humidified incubator at 37°C in 5% CO2. CaCo-2 cells were seeded onto a 96-well plate with a density of 40 × 103 cells per well and further incubated under standard cultivation conditions (37°C, 95% air, 5% CO2). When the cells reached 80% confluence and after an initial 24-hour incubation to allow cellular attachment, cells were cultured in RPMI 1640 medium with 0.5% fetal calf serum and then treated with 0.1 and 0.5 ng/mL TGF-β2 for 48 hours or with cell culture medium for 48 hours (control). Alamar Blue reduction test was used for investigation of cell viability as described by Nakayama et al (14). After the treatment, Alamar Blue solution was added directly in a final concentration of 10% and the plate was further incubated at 37°C for 3 hours. Optical density of the plate was measured spectrophotometrically at a wavelength of 570 and 630 nm with a fluorescence reader (ELISA module, Anthos microplate spectrophotometer Zenyth 200, Anthos Labtec Instruments GmbH, Salzburg, Austria). Cell viability was calculated as percentage of the difference between the reductions of Alamar Blue in treated vs controls. As a negative control, Alamar Blue was added to the medium without cells.
Evaluation of cell apoptosis following TGF-β2 administration was assessed by flow cytometry of propidium iodide–stained nuclei. CaCo-2 cells were seeded onto a 12-well plate with a density of 350 × 103 cells per well and further incubated under standard cultivation conditions (37°C, 95% air, 5% CO2). After an initial 24-hour incubation to allow cellular attachment, cells were cultured in the medium with 0.5% fetal calf serum and then were treated with 0.1 and 0.5 ng/mL TGF-β2 for 72 hours or with cell culture medium (control). Following drug treatment, cells were harvested by trypsin, combined with medium containing floating cells, washed with phosphate-buffered saline, and stained with hypotonic propidium iodide solution (50 μg/mL propidium iodide in 0.1% sodium citrate plus 0.1% Triton X-100). The propidium iodide fluorescence of individual nuclei was recorded by using FL2-H histogram plot (cell Apoptosis analysis) for assessment of nuclear hypodiploidity (FACSCalibur; BD Biosciences, Franklin Lakes, NJ).
Animals and Diet
The present experiment has been reviewed and approved by the institutional animal care and use committee at the Techion-Israel Institute of Technology, the Ruth & Bruce Rappaport Faculty of Medicine (certificate no. IL-090-09-2008). Briefly, male rats weighing 240–260 g were housed in individual stainless steel cages with unlimited access to water in a room maintained at 22°C on a 12:12-hour light-dark cycle.
TGF-β2-enriched diet contains energy 420 kJ/100 mL, protein 3.6 g/100 mL (protein sources: caseinates), fat 4.7 g/100 mL (fat source: milk fat, soy, corn, MCT 25%), carbohydrates 11 g/100 mL (carbohydrate source: corn syrup, sucrose, trace lactose), Na 35 mg/100 mL, K 120 mg/100 mL, Zn 1 mg/100 mL, Fe 1.1 mg/100 mL, Phos 61 mg/100 mL, Ca 91 mg/100 mL. The amount of TGF-β2 in the TGF-β2-enriched diet was approximately 2 μg/g of protein and negligible in the normal chow.
Forty male Sprague-Dawley rats were randomly assigned to 1 of 4 groups: group A sham rats underwent laparotomy, bowel transection, and reanastomosis (n = 10); group B sham-TGF-β rats underwent bowel transection and were treated with TGF-β2 supplemented chow from day 3 through 14 (n = 10); group C SBS-animals underwent 75% bowel resection (n = 10); and group D SBS-TGF-β animals underwent bowel resection and were treated with TGF-β2 supplemented chow similar to group B (n = 10). Diet consumption increased gradually in sham rats from 13 g per rat per day (day 3) to 25 g per rat per day (day 14) and in SBS animals from 12 to 22 g per rat per day. The amount of TGF-β2 in the TGF-β2-enriched diet was about 2 μg/g of protein and negligible in the normal chow. Animals were checked daily for physical appearance, food consumption, consistency of feces, and diarrhea.
After overnight fasting, the animals were anesthetized with an intraperitoneal injection of ketamine (90 mg/kg) and xylasine (10 mg/kg). Using sterile techniques, the abdomen was opened using a midline incision. Resected rats had bowel 5 cm distal to the ligament of Treitz to 10 cm proximal to the cecum removed (75% mid-small bowel resection). Transected rats underwent bowel transection 10 cm proximal to the cecum followed by restoration of bowel continuity by end-to-end anastomosis using 6-0 absorbable suture (Vicryl; Ethicon Corporation, San Angelo, TX). For all of the operations, the abdominal cavity was closed in 2 layers with a running suture of 3-0 Vicryl (Ethicon Corporation). Postoperative rats were allowed ad libitum water and a liquid diet for 72 hours. Dietary treatment was started on the fourth postoperative day. Rats were monitored for 14 days as to food (pair fed) and fluid intake and weight.
Enterocyte Proliferation and Apoptosis
Crypt cell proliferation was assessed by evaluation of 5-bromodeoxyuridine (5-BrdU) incorporation. Standard BrdU labeling reagent (Zymed Laboratories, San Francisco, CA) was injected intraperitoneally at a concentration of 1 mL/100 g body weight 2 hours before sacrifice. Tissue slices were stained with a biotinylated monoclonal anti-BrdU antibody system provided in a kit form (Zymed Laboratories). An index of proliferation was determined as the ratio of crypt cells staining positively for BrdU per 10 crypts.
Additional 5-mL-thick sections were prepared to establish the degree of enterocyte apoptosis. Immunohistochemistry for caspase-3 (caspase-3 cleaved concentrated polyclonal antibody; dilution 1:100; Biocare Medical, Walnut Creek, CA) was performed for identification of apoptotic cells using a combination of the streptovidin–biotin–peroxidase method and microwave antigen retrieval on formalin-fixed, paraffin-embedded tissues according to the manufacturer's protocols. In crypts, cell apoptosis was expressed as the total number of apoptotic cells per 100 crypts. Apoptosis along the villi was differentiated between the lower one-third of the villi (lateral villi) and upper one-third of the villi (villi tips). A qualified pathologist blinded as to the source of intestinal tissue performed all of the measurements.
Immunohistochemistry for TGF-β2 Receptor Expression
Immunohistochemistry for TGF-β2 receptor (TGF-β2 polyclonal antibody; dilution 1:100, sc-400; Santa Cruz, CA) was performed to identify TGF-β2 receptor expression, using a combination of streptovidin–biotin–peroxidase method according to manufacturers’ protocols. The paraffin-embedded sections were dewaxed and rehydrated with xylene and graded alcohol. Tissue sections were microwave pretreated in 10 mmol/L EDTA buffer and incubated with an endogenous peroxidase (3%) in methanol for 10 min. After incubation with blocking solution at room temperature for 10 min, the sections were then incubated with TGF-β2 receptor concentrated polyclonal antibody (dilution 1:50) for 60 min and second human-absorbed, biotinylated, affinity-purified antibody for 20 min. TGF-β2 receptor positive color development was obtained by incubating the sections with 3,3′-deoxyaminobenzidine (DAB) substrate (Zymed Laboratories). TGF-β2 receptor expression and enterocyte apoptosis along the villi was differentiated between the lower one-third of the villi (lateral villi), upper one-third of the villi (villi tips), and crypt compartment. A qualified pathologist blinded to the source of intestinal tissue performed all of the measurements.
Tissue was homogenized in RIPA lysis buffer containing 50 mmol/L Tris-HCl (pH 7.4), 150 mmol/L NaCl, 1% NP-40, 2 mmol/L EDTA, supplemented with a cocktail of protease and phosphatase inhibitors. Protein concentrations were determined by Bradford reagent according to the manufacturer's instructions. Samples containing an equal amount of total protein (15 μg) were resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis under reducing conditions. After electrophoresis, proteins were transferred to a polyvinylidene difluoride membrane and probed with various primary antibodies to anti-Bcl-2 antibody (1:1000 dilution, sc-7382), anti-bax antibody (1:200 dilution, sc-493), anti-phospho-extracellular signal-regulated kinase (ERK) antibody (1:2500 dilution, sc-7383), anti-β-catenin antibody (1:1000 dilution, sc-7199), anti-TGF-β2 receptor antibody (1:1000 dilution, sc-400), anti-ERK2 antibody (1:1000 dilution, sc-56899), and anti-β-actin antibody (1:1000 dilution). Horseradish peroxidase-conjugated secondary antibody was purchased from Jackson ImmunoResearch Laboratories (West Grove, PA) and an enhanced chemiluminescent substrate from Biological Industries (Kibbutz Beth HaEmek, Israel). The optical density of the specific protein bands was quantified by using a densitometer (Vilber Lourmat, Lyon, France).
The data are expressed as mean ± standard error of the mean. A 1-way ANOVA for comparison, followed by Tukeys test for pairwise comparison was used for statistical analysis. Prism software was used (GraphPad Software, San Diego, CA); and statistical significance was defined as P < 0.05.
Effect of TGF-β2 on CaCo-2 Cell Viability and Apoptosis
Since postresected bowel is characterized by high epithelial cell proliferative activity, in an effort to assess the biological effects of TGF-β2, we first determined the effects of TGF-β2 on viability/cytotoxicity and apoptosis of CaCo-2 cells. As expected, incubation with TGF-β2 resulted in a significant decrease in viability of CaCo-2 cells compared with medium only (Fig. 1). Treatment of CaCo-2 cells with increasing concentrations of TGF-β2 (0.5 ng/mL) did not significantly change cell viability compared with low concentration of TGF-β2 (0.1 ng/mL). Results of FACS analysis showed that TGF-β2 increased significantly apoptosis of CaCo-2 cells in a dose-dependent fashion with the maximal effect observed at a concentration of 0.5 ng/mL.
TGF-β Receptor Expression and Cellular Proliferation and Apoptosis in Crypts
SBS rats (group C) demonstrated a significant increase in enterocyte proliferation in both jejunum (37%, P < 0.05) and ileum (19%, P < 0.05) and a concomitant increase in cell apoptosis in jejunum (2.5-fold increase, P < 0.05) and ileum (4-fold increase, P < 0.05) compared with sham animals (group A) (Fig. 2). Massive small bowel resection resulted in a strong increase in the TGF-β2 receptor immunoreactivity of apical and basolateral membranes of enterocytes within the crypts compared with control animals. This increase in receptor expression coincided with a decrease in crypt cell proliferation in jejunum (25%, P < 0.05) and ileum (18%, P < 0.05) and by a concomitant increase in crypt cell apoptosis in jejunum (48%, P < 0.05) and ileum (3-fold increase, P < 0.05) following exposure to TGF-β2-enriched diet (group D) (vs SBS untreated animals [group C]).
TGF-β Receptor Expression and Enterocyte Apoptosis in Villi
SBS rats demonstrated a significant increase in cell apoptotic rates in villus tips (2.5-fold increase, P < 0.05) compared with sham rats (group A) (Fig. 3). In contrast to the crypt compartment, SBS rats (group B) demonstrated a lower TGF-β2 receptor immunoreactivity in villus tips, compared with sham rats. This decrease in TGF-β2 receptor expression coincided with decrease in cell apoptosis in villus tips (2-fold decrease, P < 0.05) following TGF-β2 administration (group C). In contrast to the crypts and villus tips, TGF-β receptor expression remained unchanged in lateral villi in resected rats compared with sham rats. In relation to TGF-β2r expression, cell apoptosis remain unchanged in this compartment following TGF-β2 administration.
Treatment of sham animals with TGF-β2-supplemented diet (group B) resulted in a significant decrease in p-ERK and increase in bax protein expression compared with sham rats (group A), which correlated with decreased cell proliferation and increased cell apoptosis. Short bowel syndrome rats (group B) showed an increased intensity of the bands of bax, p-ERK, and β-catenin compared with sham animals, which was closely correlated with increased cell proliferation and cell apoptosis after bowel resection. Treatment of SBS animals with TGF-β2-supplemented diet (group D) led to a significant decrease (vs SBS rats) in bax, p-ERK, and β-catenin, which was in agreement with decreased cell proliferation and cell apoptosis in this group. Total TGF-β receptor expression (villus and crypt) was downregulated in SBS rats compared with control animals but was upregulated following dietary TGF-β2 administration (Fig. 4).
In the intestinal mucosa, numerous cytokines have been shown to affect epithelial cell differentiation and proliferation through epithelial–mesenchymal and epithelial–immune cell interaction (15). The TGF-β family includes TGF-β1, -2, and -3 and several peptides exhibiting various degrees of homology to these prototypic members. TGF-β is usually secreted usually as a biologically inactive complex; the physiological processes regulating its bioactivation from the secreted latent form are poor understood. TGF-β has been found to bind to specific cell surface molecules (7 different type I receptors and 4 different type II (9,10). After ligand stimulation, the activated distinct set of type I and type II receptors transduces the signal by phosphorylation of receptor-regulated Smad molecules, which form a complex with Smad 4 and translocate to the nucleus to regulate transcription. The TGF-β family appears to play a key regulatory function in a diverse spectrum of biological processes including modulation of proliferative activity of virtually all of the mammalian cell populations, cellular differentiation, embryological development of many tissues, and formation of extracellular matrix.
TGF-β2, a multifunctional polypeptide (cytokine) present in human and bovine milk, plays a critical role in the development of tolerance, the prevention of autoimmunity, and in anti-inflammatory responses. TGF-β2 is a potent inhibitor of intestinal epithelial cell (IEC) growth and stimulates IEC differentiation (12). Several experiments have shown that TGF-β2 exerts its maximal effect when activated within milk protein environment. Therefore, TGF-β2-enriched formula (but not pure TGF-β2) was chosen for in vivo experiment. The amount of TGF-β2 in the TGF-β2-enriched diet was approximately 2 μg/g of protein and negligible in the normal chow.
We have shown that TGF-β2-enriched diet inhibits intestinal regrowth after massive small bowel resection in a rat (13); however, the mechanisms of this negative effect remain poorly understood. In particular, it is not clear where this factor acts along the crypt–villus axis and whether it acts on different locations along this axis in correlation with the TGF-β2 receptor.
The purpose of the present study was to investigate the effects of TGF-β2 on intestinal mucosal homeostasis in conjunction with TGF-β2 receptor expression along the villus–crypt axis following bowel resection in rats. Cell proliferation and apoptosis were measured in crypts of the remnant ileum to characterize enterocyte turnover, and enterocyte apoptosis was determined in villus tips and lateral villi.
Because postresected bowel is characterized by a high epithelial cell proliferative activity, we first determined the effect of TGF-β2 on viability and apoptosis of CaCo-2 cells. We have demonstrated that incubation of CaCo-2 cells with TGF-β2 resulted in a significant decrease in viability and increased apoptosis of CaCo-2 cells compared with medium only, and that treatment of CaCo-2 cells with increasing concentrations of TGF-β2 (0.5 ng/mL) did not significantly change cell viability compared with a low concentration of TGF-β2 (0.1 ng/mL); however, the proapoptotic effect of TGF-β2 was dose dependent with the maximal effect observed at a concentration of 0.5 ng/mL. The antiproliferating and pro-apoptotic effects of TGF-β on epithelial cells have been described in many experiments (16). Biasi et al (17) have demonstrated that CaCo-2 cells are sensitive to the growth-inhibitory effects of TGF-β, which induced a marked enhancement of apoptosis through c-Jun N-terminal kinase and Smad4 activities.
Next, we investigated whether the effect of TGF-β2 on enterocyte turnover (proliferation and apoptosis) is correlated with TGF-β2 receptor expression along the villus–crypt axis after bowel resection in a rat. Similar to our previous experiments (13), in the current study, massive small bowel resection resulted in a significant increase in the enterocyte proliferation rate in both jejunum and ileum, which was accompanied by the augmented p-ERK protein levels as well as an increased β-catenin expression, which suggests elevated intestinal stem cell activity. Evaluation of enterocyte apoptosis in the present study has shown increased cell apoptosis in SBS rats compared with sham animals and was accompanied by enhanced bax protein levels. The present observation is consistent with our previous reports and data from other investigators (18).
Next, we investigated the effects of TGF-β2-enriched diet (Modulen) on intestinal mucosal homeostasis in resected animals. The three active ingredients in Modulen, TGF-β2, glutamine, and short-chain fatty acids, combine to provide excess energy sources to intestinal cells under catabolic conditions and to provide antioxidant activity and cytokines to facilitate cell regeneration. The stimulating effects of glutamine and short-chain fatty acids on intestinal regrowth after massive small bowel resection have been previously described (1,4). Despite the presence of these pro-adaptive agents, the exposure of resected animals to TGF-β2-enriched diet in our experiment inhibited intestinal regrowth. This suggests that the inhibitory effect of TGF-β2 on intestinal mucosal homeostasis exceeds the pro-adaptive effects of glutamine and short-chain fatty acids. Exposure of SBS animals to the TGF-β2-enriched diet resulted in an impaired enterocyte proliferation. TGF-β have been found to inhibit proliferation of all of the epithelial cell proliferation of all of the epithelial cell populations through prolongation of the G I phase (19). TGF-β also overrides the effects of direct mitogens such as EGF, TGF-α, and IGF including their effect on intestinal epithelium (7,20). Decreased p-ERK and β-catenin protein levels were in agreement with decreased cell proliferation. These data are consistent with previous studies suggesting an interaction between the TGF-β and Wnt/β-catenin pathways (21). It is believed that TGF-β binding of TGF-β type II receptors, which colocalize with adherent junction complexes in PTC, results in dissociation of β-catenin from the adherent junction complex and an increase in the availability of cytoplasmic β-catenin, which binds to the TGF-β1-signaling molecules, the R-Smads (22).
In the present study, we investigated the effects of TGF-β2 on enterocyte proliferation and apoptosis in conjunction with TGF-β2 receptor expression along the villus–crypt axis. We have shown that the antiproliferative and pro-apoptotic effect of TGF-β2 on enterocyte turnover is strongly correlated with TGF-β2 receptor expression along the villus–crypt axis. In the crypt compartment, a significant increase in TGF-β2 receptor expression after bowel resection coincided with decreased cell proliferation and increased cell apoptosis following TGF-β2 administration. The stimulating effect of TGF-β2 on cell apoptosis in the crypt was consistent with pro-apoptotic effects of TGF-β2 in CaCo-2 cells in an in vitro experiment. In villus tips, SBS rats demonstrated lower receptor immunoreactivity compared with sham animals. Because TGF-β exerts proapoptotic effects, this decrease in TGF-β2 receptor expression coincides with decreased cell apoptosis in villus tips following TGF-β2 administration. Growing evidence suggests that intestinal epithelium, like other rapidly renewing tissues, may process a feedback control whereby cell division in the precursor compartment (the crypts) is regulated by the number of cells in the functional compartment (the villus). It should be emphasized that treatment with TGF-β2 decreases significantly cell proliferation and increases cell apoptosis in crypts, but may inhibit cell death pathways in villus tips by this feedback control.
In summary, in a rat model of SBS, exposure to a TGF-β2-enriched diet inhibits enterocyte turnover. The antiproliferative and pro-apoptotic effects of TGF-β2 are correlated with TGF-β2 receptor expression along the villus–crypt axis.
1. Booth IW, Lander AD. Short bowel syndrome
. Bailliere's Clin Gastroenterol
2. Coran AG, Spivak D, Teitelbaum DH. An analysis of the morbidity and mortality of short bowel syndrome
in the pediatric age group. Eur J Pediatr Surg
3. Sukhotnik I, Coran AG, Kramer A, et al. Advances in the research of the short bowel syndrome
: un updated review. Pediatr Surg Int
4. O’Brien DP, Nelson LA, Huang FS, et al. Intestinal adaptation: structure, function, and regulation. Semin Pediatr Surg
5. Martin GR, Wallace LE, Hartmann B, et al. Nutrient-stimulated GLP-2 release and crypt cell proliferation in experimental short bowel syndrome
. Am J Physiol Gastrointest Liver Physiol
6. De Santa Barbara P, van den Brink GR, Roberts DJ. Development and differentiation of the intestinal epithelium. Cell Mol Life Sci
7. Sukhotnik I, Krausz MM, Siplovich L, et al. Peptide growth factors and intestinal adaptation in short bowel syndrome
. Isr Med Assoc J
8. Goumans MJ, Mummery C. Functional analysis of the TGFb receptor/Smad pathway through gene ablation in mice. Int J Dev Biol
9. Massague J. TGF-β signal transduction. Ann Rev Biochem
10. Ten Dijke P, Yamashita H, Ichijo H, et al. Characterization of type I receptors for transforming growth factor-β and activin. Science
11. Marek A, Brodzicki J, Berek A, et al. TGF-β (transforming growth factor β) in chronic inflammatory conditions: a new diagnostic and prognostic marker? Med Sci Monit
12. Oz HS, Ray M, Chen TS, et al. Efficacy of a transforming growth factor beta 2 containing nutritional support formula in a murine model of inflammatory bowel disease. J Am Coll Nutr
13. Sukhotnik I, Mogilner JG, Ben Lulu S, et al. Nutritional supplementation with transforming growth factor-beta (TGF-β) inhibits intestinal adaptation after massive small bowel resection in a rat. Pediatr Surg Int
14. Nakayama GR, Caton MC, Nova MP, et al. Assessment of the Alamar Blue assay for cellular growth and viability in vitro. J Immunol Methods
15. Schaeffer C, Diab-Assef M, Plateroti M, et al. Cytokine gene expression during postnatal small intestinal development: regulation by glucocorticoids. Gut
16. Vizio B, Poli G, Chiarpotto E, et al. 4-Hydroxynonenal and TGF-beta1 concur in inducing antiproliferative effects on the CaCo-2 human colon adenocarcinoma cell line. Biofactors
17. Biasi F, Vizio B, Mascia C, et al. c-Jun N-terminal kinase upregulation as a key event in the proapoptotic interaction between transforming growth factor-beta1 and 4-hydroxynonenal in colon mucosa. Free Radic Biol Med
18. Stern LE, Huang F, Kemp CJ, et al. Bax is required for increased enterocyte apoptosis
after massive small bowel resection. Surgery
19. Ciacci C, Lind SE, Podolsky DK. Transforming growth factor beta regulation of migration in wounded rat intestinal epithelial monolayers. Gastroenterology
20. Kojima H, Hidaka H, Matsumura K, et al. Converted regulation of early enterocyte differentiation by insulin-like growth factor I, insulin, and transforming growth factor-beta 1. Proc Assoc Am Phys
21. Nishita M, Hashimoto MK, Ogata S, et al. Interaction between Wnt and TGF-beta signaling pathways during formation of Spemann's organizer. Nature
22. Zhang M, Lee CD, Luo DD, et al. Polarity of response to transforming growth factor-beta1 in proximal tubular epithelial cells is regulated by beta-catenin. J Biol Chem