Glucagonlike Peptide 2 Analogue Teduglutide: Stimulation of Proliferation but Reduction of Differentiation in Human Caco-2 Intestinal Epithelial Cells

Importance  Short bowel syndrome occurs when a shortened intestine cannot absorb sufficient nutrients or fluids. Teduglutide is a recombinant analogue of human glucagonlike peptide 2 that reduces dependence on parenteral nutrition in patients with short bowel syndrome by promoting enterocytic proliferation, increasing the absorptive surface area. However, enterocyte function depends not only on the number of cells that are present but also on differentiated features that facilitate nutrient absorption and digestion.

Objective  To test the hypothesis that teduglutide impairs human intestinal epithelial differentiation.

Design and Setting  We investigated the effects of teduglutide in the modulation of proliferation and differentiation in human Caco-2 intestinal epithelial cells at a basic science laboratory. This was an in vitro study using Caco-2 cells, a human-derived intestinal epithelial cell line commonly used to model enterocytic biology.

Exposure  Cells were exposed to teduglutide or vehicle control.

Main Outcomes and Measures  We analyzed the cell cycle by bromodeoxyuridine incorporation or propidium iodide staining and flow cytometry and measured cell proliferation by 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium (MTS) assay. We used quantitative reverse transcription–polymerase chain reaction to assay the expression of the enterocytic differentiation markers villin, sucrase-isomaltase, glucose transporter 2 (GLUT2), and dipeptidyl peptidase 4 (DPP-4), as well as that of the putative differentiation signals schlafen 12 (SLFN12) and caudal-related homeobox intestine-specific transcription factor (Cdx2). Villin promoter activity was measured by a luciferase-based assay.

Results  The MTS assay demonstrated that teduglutide increased cell numbers by a mean (SD) of 10% (2%) over untreated controls at a maximal 500nM (n = 6, P < .05). Teduglutide increased bromodeoxyuridine-positive cells vs untreated controls by a mean (SD) of 19.4% (2.3%) vs 12.0% (0.8%) (n = 6, P < .05) and increased the S-phase fraction by flow cytometric analysis. Teduglutide reduced the mean (SD) expression of villin by 29% (6%), Cdx2 by 31% (10%), DPP-4 by 15% (6%), GLUT2 by 40% (11%), SLFN12 by 61% (14%), and sucrase-isomaltase by 28% (8%) (n = 6, P < .05 for all).

Conclusions and Relevance  Teduglutide increased Caco-2 proliferation but tended to inhibit intestinal epithelial differentiation. The effects of mitogenic stimulation with teduglutide in patients with short bowel syndrome might be greater if the more numerous teduglutide-treated cells could be stimulated toward a more fully differentiated phenotype.

Small intestine failure occurs when the mucosa becomes atrophic after prolonged starvation or when 1 or more bowel resections decrease the amount of small intestinal surface area available to interact with luminal nutrients. Critically ill patients who have not received enteral feeding may exhibit mucosal barrier failure early, leading to bacterial translocation and a septic response.¹ After recovery from their acute events, such patients may have difficulty readapting to enteral nutrition because the atrophied mucosa is unable to handle the digestive load. After massive small-bowel resection, patients experience more pronounced difficulties. Although adaptation to short bowel syndrome (SBS) occurs,² its capability is limited, and many patients require permanent total parenteral nutrition or small-bowel transplantation, each of which has well-recognized attendant morbidities and a substantial long-term mortality rate.^3,4

We attempt to palliate such patients by manipulating their diets in composition and frequency to make it more readily digestible, adding antiperistaltic agents to slow motility and increase the dwell time of the nutrients within the gut, as well as treating bacterial overgrowth where appropriate. However, the main modalities of treatment for SBS focus on the stimulation of proliferation. Supplementation of growth hormone, glutamine, and enteral nutrition has been effective in promoting intestinal adaptation in selected patients with SBS.^5-7 It has also been reported that glutamine acts better in combination with growth hormone in animal investigations.⁸ Recently, glucagonlike peptide 2 (GLP-2) and the GLP-2 analogue teduglutide (ALX-0600) have been reported to promote intestinal growth in patients with SBS.^9,10 Teduglutide is believed to be more biologically active than native GLP-2 in stimulating intestinal epithelial proliferation because teduglutide is resistant to GLP-2 degradation by dipeptidyl peptidase 4 (DPP-4),^9,11 but its effect on the regulation of intestinal epithelial differentiation has not been studied.

In the present study, we investigated the effects of teduglutide in the modulation of proliferation and differentiation in human Caco-2 intestinal epithelial cells. We hypothesized that teduglutide impairs human intestinal epithelial differentiation. To test this hypothesis, we compared the expression of villin, DPP-4, sucrose-isomaltase, glucose transporter 2 (GLUT2, also known as SLC2A2), and caudal-related homeobox intestine-specific transcription factor (Cdx2), as well as the putative differentiation marker schlafen 12 (SLFN12), in the presence or absence of teduglutide. We also assessed the effects of teduglutide on Caco-2 cell proliferation to verify biological activity in this model.

Dulbecco modified Eagle medium was obtained from Sigma-Aldrich. Penicillin-streptomycin and 0.5% trypsin–EDTA were from Gibco. The 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium (MTS) inner salt was purchased from Promega Corporation. Teduglutide (CPZ1435 HD-33) was obtained from Creative Peptides Inc. Human epidermal growth factor and other common laboratory reagents were obtained from Sigma-Aldrich.

We studied Caco-2 brush border enzyme intestinal epithelial cells, a subclone of the original Caco-2 cell line, selected for their ability to differentiate in culture toward an enterocytic phenotype as indicated by the formation of an apical brush border and the expression of brush border enzymes.¹² We maintained these cells at 37°C with 8% carbon dioxide in Dulbecco modified Eagle medium with 4500 mg/L of d-glucose, 4mM glutamine, 1mM sodium pyruvate, 100 U/mL of penicillin, 100 μg/mL of streptomycin, 10 μg/mL of transferrin, 10mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (pH 7.4), and 3.7 g/L of sodium bicarbonate, supplemented with 10% fetal bovine serum. All studies were performed on the cells within 15 passages.

Proliferation was assessed using a colorimetric MTS assay as described by the manufacturer (Promega Corporation). Briefly, Caco-2 cells (10 000 cells per 96-well plate) were plated in cell culture medium for 24 hours. On the next day, the cells were incubated with 0 to 1000nM teduglutide in serum-free medium for 72 hours. The experiment was terminated by adding 20 μL of 1 mg/mL of MTS solution to each of 96 wells. After 30 minutes of incubation in the dark at 37°C, the absorbance of each well was measured at 490 nm using a microplate reader (Molecular Devices, LLC).

Bromodeoxyuridine (BrdU) incorporation was assessed using a fluorescein isothiocyanate conjugated BrdU flow kit as described by the manufacturer (BD Biosciences). Briefly, Caco-2 cells (10 000 cells per 96-well plate) were plated in cell culture medium for 24 hours. On the next day, the cells were incubated with 500nM teduglutide and human epidermal growth factor in serum-free medium for 72 hours. Before termination of experiments, the cells were incubated with 10μM BrdU for 6 hours before flow analysis.

Caco-2 cells (10 000 cells per 96-well plate) were plated in cell culture medium for 24 hours. The next day, the cells were incubated with 500nM teduglutide in serum-free medium for 72 hours. The cells were harvested and washed twice with ice-cold phosphate-buffered saline containing 50% fetal bovine serum and then fixed with 70% ethanol for 1 hour and incubated in 0.35 mL of phosphate-buffered saline containing 50 µL/mL of propidium iodide (Sigma-Aldrich), 66 U/mL of RNase (Invitrogen), and 10% Triton X-100 (Sigma-Aldrich) on ice for 60 minutes. The DNA content analysis was performed by a FACScan with CellQuest software (BD Biosciences).

The RNA was extracted using the RNeasy kit (Qiagen). The complementary DNA (cDNA) generated was amplified using SYBR Green Real-time PCR Master Mix on an Applied Biosystems 7500 Real-time PCR System (Invitrogen). Expression levels were determined from the threshold cycle values using the method of 2^−ΔΔCt with 18S expression as the reference control gene. Human 18S primers used were 5′-CGCCGGTCCAAGAATTTCACCTCT-3′ (upstream) and 5′-CCCTCGATGCTCTTAGCTGAGTGT-3′ (downstream). Human villin primers used were 5′-TGCTATCTATGGTGTGGGAAGG-3′ (upstream) and 5′-TCCTGTAGTCTCTTGGTGTTGG-3′ (downstream). Other markers used were the following: SLFN12 (forward 5′-ATCTGGGCTTGCAAGAGAAC-3′ and reverse 5′-TTTTTGCCAGCTTCTGCTTT-3′), Cdx2 (forward 5′-CAACCTGGACTTCCTGTCAT-3′ and reverse 5′-CACAGACCAACAACCCAAAC-3′), DPP-4 (forward 5′-TTTGGGGCTGGTCATATGGAGGG-3′ and reverse 5′-ACTCCCACCGGGATACAGGGG-3′), GLUT2 (forward 5′-AGCTGCATTCAGCAATTGGACCTG-3′ and reverse 5′-ATGTGAACAGGGTAAAGGCCAGGA-3′), and sucrase-isomaltase (forward 5′-AAACCTACATGTGGGTGGTGGTCA-3′ and reverse 5′-AACAGAGAACCCTGTGCCATCTGA-3′). The cycle conditions for the polymerase chain reaction were 1 cycle of 5 minutes at 95°C and 40 cycles of 15 seconds at 95°C, 30 seconds at the annealing temperature (60°C), and 30 seconds at 60°C for extension.

All experiments were performed independently at least 3 times, with similar results. Data sets were analyzed using paired or unpaired t tests with Bonferroni correction as appropriate. Statistical significance was set at P < .05.

Teduglutide significantly increased cell numbers at higher concentrations (250-1000nM) compared with cell numbers in untreated control cell populations (Figure 1A). We further confirmed the effect of 500nM teduglutide on cell proliferation with BrdU incorporation, followed by flow analysis. The cells incubated with 500nM teduglutide exhibited a greater mean (SD) proportion of BrdU-positive cells than untreated control cells (21.1% [1.5%] vs 12.0% [0.8%], n = 6, P < .05) (Figure 1B). The cells incubated with a positive control epidermal growth factor (50 ng/mL) also exhibited a greater mean (SD) proportion of BrdU-positive cells than untreated control cells (26.7% [3.7%] vs 12.0% [0.8%], n = 6, P < .05) (Figure 1B).

We further performed cell cycle analysis using propidium iodide staining, followed by flow analyses. Teduglutide also increased the proportion of diploid cells in S phase compared with the S-phase fraction of untreated control cell populations (Figure 2A and B).

Human Caco-2 intestinal epithelial cells were treated with 500nM teduglutide in serum-free medium for 72 hours, and quantitative reverse transcription–polymerase chain reaction was performed with cDNA amplified from RNA. Teduglutide reduced the mean (SD) expression of differentiation marker transcripts: villin was reduced by 29% (6%), DPP-4 by 15% (6%), sucrase-isomaltase by 28% (8%), and GLUT2 by 40% (11%) (n = 6, P < .05 for all) (Figure 3).

Human Caco-2 intestinal epithelial cells were treated with 500nM teduglutide in serum-free medium for 72 hours, and quantitative reverse transcription–polymerase chain reaction was performed with cDNA amplified from RNA. We measured the effect of teduglutide on the expression of 2 intracellular signals (Cdx2 and SLFN12) that might influence enterocytic proliferation and differentiation. Teduglutide reduced the mean (SD) expression of the Cdx2 by 31% (10%) and of SLFN12 by 61% (14%) (n = 6, P < .05 for both) (Figure 4).

Small intestinal mucosal function is likely to be determined by the total absorptive surface area available to interact with luminal contents and by the phenotype of the enterocytes that line the lumen. Although some nutrient transport is passive, most digestive functions require active transport proteins or digestive enzymes that are increasingly expressed as the enterocyte matures. This study suggests that mitogenic stimuli such as teduglutide, although increasing the number of enterocytes available and the absorptive surface area, may not proportionately increase the amount of mucosal protein available for these energy-dependent critical digestive tasks.

There are many different markers of the mature enterocytic phenotype. Enzymes of the brush border membrane and transport proteins are characteristic features of the differentiated enterocytes.¹³ For this study, we selected villin, DPP-4, sucrose-isomaltase, and GLUT2. Villin is a calcium-regulated actin-binding protein of the microvillus core of the brush border, expressed in crypts and villi of the small and large intestines.^13,14 Expressed in the intestinal brush border of mature enterocytes, DPP-4 and its expression are controlled by diet composition.^15,16 Sucrose-isomaltase is a glucosidase enzyme expressed in the brush border of mature enterocytes and is increasingly expressed during developmental maturation of the intestine.^17,18 The levels of the fructose and glucose transporter GLUT2 are regulated by components of diet such as fructose and fat,¹⁹ and increased expression of GLUT2 and the hexose transporter sucrose-isomaltase has been reported after polyamine-induced differentiation in vivo.²⁰ Both GLUT2 and sucrose-isomaltase are commonly used markers of enterocytic differentiation.^13-16,20

The mechanism of the apparent inhibition of differentiation by teduglutide is not clear. It is possible that more rapid proliferation does not allow adequate time for differentiation. Alternatively, GLP-2 stimulation not only may initiate mitogenic signals but also might directly inhibit some of the intracellular signals that trigger differentiation. A member of the caudal-related homeobox gene family of transcription factors, Cdx2 is involved in enterocyte lineage specification. The function of Cdx2 within the cell is to induce differentiation and inhibit proliferation at the level of gene transcription.²¹ It activates many intestine-specific genes such as sucrase-isomaltase and lactase-phlorizin hydrolase.²² Schlafen 12 is a member of the schlafen superfamily that is expressed in humans. Rat schlafen 3 is orthologous to human SLFN12 and has recently been implicated in intestinal epithelial differentiation.^23,24 Therefore, the teduglutide-induced decreases in Cdx2 or SLFN12 could contribute to diminished differentiation in teduglutide-treated intestinal epithelial cells. This question awaits further study.

It would be important to confirm these studies in vivo because there are obvious differences between cell lines in culture and the intestinal epithelium in an intact human. Unfortunately, the pharmaceutical company that manufactures teduglutide declined to make human biopsy samples from preclinical testing available for external study. Nevertheless, investigators have commonly used human Caco-2 intestinal epithelial cells,^12,25-28 so these findings suggest the possibility that overdriving cell proliferation may result in a population of enterocytes that is on average less mature and less differentiated than at baseline. These results do not mean that teduglutide or other mitogenic agents, such as growth hormone, should not be used in SBS. Indeed, some evidence suggests that these agents can help to facilitate weaning from total parenteral nutrition in selected subgroups of patients with SBS. It has been reported that treatment with teduglutide reduced dependency on parenteral nutrition for some patients with SBS and intestinal failure.^29-31 However, many of the patients who benefited by demonstrating reduced parenteral fluid dependency in these studies still required some parenteral fluids and total parenteral nutrition, while mitogenic stimuli alone seem insufficient to benefit patients with extreme short gut syndromes. In addition, the effects of growth hormone or teduglutide regress rapidly if the drugs are stopped because of the rapid progression of migration from crypt to villous tip and then cell loss in vivo as mature enterocytes are shed into the lumen. Therefore, others have raised concerns about the possibility of neoplastic transformation in response to such long-term mitogenic stimuli.³² These results suggest that it may be useful to focus at least some future efforts on the promotion of enterocytic differentiation as an alternative to or a synergistic treatment with mitogenic interventions. Several stimuli have been reported to promote such differentiation, including transforming growth factor β,^33,34 gastrin-releasing peptide and its receptor,³⁵ bombesin,³⁶ calcium and vitamin D₃,³⁷ Notch pathway,³⁸ and sodium butyrate.³⁹

It seems possible that the physical stimulation of repetitive deformation that occurs during peristalsis or villous motility may facilitate enterocyte differentiation.^23,40,41 Early low-level feeding, which stimulates such mechanical activity by the gut, may be trophic for the gut mucosa even in the presence of adequate parenteral nutrition. This is well established in the critical care literature but is perhaps not applied universally.

There are other phenotypic characteristics of effective enterocytic differentiation that we have not studied, and in vivo investigations might yield different results, particularly in the aberrant neuroendocrine environment that is likely to characterize patients with SBS. Nevertheless, these results raise a caution that purely mitogenic stimuli may have suboptimal effects on the small intestinal epithelium. Combining a mitogenic stimuli, such as teduglutide, with a differentiating agent might improve weaning from total parenteral nutrition in patients with extreme SBS.

Accepted for Publication: May 13, 2013.

Corresponding Author: Marc D. Basson, MD, PhD, MBA, Department of Surgery, College of Human Medicine, Michigan State University, 1200 E Michigan Ave, Ste 655, Lansing, MI 48912 (marc.basson@hc.msu.edu).

Published Online: September 25, 2013. doi:10.1001/jamasurg.2013.3731.

Author Contributions: Drs Chaturvedi and Basson had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Basson.

Acquisition of data: All authors.

Analysis and interpretation of data: All authors.

Drafting of the manuscript: All authors.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Chaturvedi.

Administrative, technical, and material support: Chaturvedi.

Study supervision: All authors.

Conflict of Interest Disclosures: None reported.

Funding/Support: This work was supported in part by grant 1 R56 DK096137-01 from the National Institutes of Health (Dr Basson).

Role of the Sponsor: The National Institutes of Health had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Previous Presentation: Presented at the 37th Annual Surgical Symposium of the Association of VA Surgeons; April 23, 2013; Milwaukee, Wisconsin.