- Email: [email protected]
Glucagon like peptide-2 induces intestinal restitution through VEGF release from subepithelial myofibroblasts
Available online at www.sciencedirect.com
European Journal of Pharmacology 578 (2008) 279 – 285 www.elsevier.com/locate/ejphar
Glucagon like peptide-2 induces intestinal restitution through VEGF release from subepithelial myofibroblasts Kerem Bulut a,⁎, Christian Pennartz a , Peter Felderbauer a , Juris J. Meier a , Matthias Banasch a , Daniel Bulut b , Frank Schmitz a , Wolfgang E. Schmidt a , Peter Hoffmann c a b
Department of Medicine I, St. Josef-Hospital, Ruhr-University Bochum, Germany Department of Medicine II, St. Josef-Hospital, Ruhr-University Bochum, Germany c Department of Medicine, Kliniken-Essen Mitte, Essen, Germany
Received 5 January 2007; received in revised form 21 August 2007; accepted 24 August 2007 Available online 15 September 2007
Abstract Glucagon like peptide-2 (GLP-2) exerts intestinotrophic actions, but the underlying mechanisms are still a matter of debate. Recent studies demonstrated the expression of the GLP-2 receptor on fibroblasts located in the subepithelial tissue, where it might induce the release of growth factors such as keratinocyte growth factor (KGF) or vascular endothelial growth factor (VEGF). Therefore, in the present studies we sought to elucidate the downstream mechanisms involved in improved intestinal adaptation by GLP-2. Human colonic fibroblasts (CCD-18Co), human colonic cancer cells (Caco-2 cells) and rat ileum IEC-18 cells were used. GLP-2 receptor mRNA expression was determined using real time RTPCR. Conditioned media from CCD-18Co cells were obtained following incubation with GLP-2 (50–250 nM) for 24 h. Cell viability was assessed by a 3-[4, 5-dimethylthiazol-2-yl]-2, 5-diphenyl-tetrazolium bromide (MTT)-assay, and wound healing was determined with an established migration-assay. Transforming Growth Factor beta (TGF-β), VEGF and KGF mRNA levels were determined by RT-PCR. Protein levels of VEGF and TGF-β in CCD-18Co cells following GLP-2 stimulation were determined using ELISA. Neutralizing TGF-β and VEGF-A antibodies were utilized to assess the role of TGF-β and VEGF-A in the process of wound healing. GLP-2 receptor expression was detected in CCD-18Co cells. Conditioned media from CCD-18Co cells dose-dependently induced proliferation in Caco-2 cells, but not in IEC-18 cells. Conditioned media also enhanced cell migration in IEC-18 cells (P b 0.01), while migration was even inhibited in Caco-2 cells (P b 0.0012). GLP2 significantly stimulated mRNA expression of VEGF and TGF-β, but not of KGF in CCD-18Co. The migratory effects of GLP-2 were completely abolished in the presence of TGF-β and VEGF-A antibodies. GLP-2 exerts differential effects on the epithelium of the small intestine and the colon. Thus, in small intestinal cells GLP-2 stimulates wound repair, whereas no such effects were observed in colonic cells. The mechanisms underlying GLP-2 induced intestinal wound repair seem to involve the secretion of VEGF and, subsequently, TGF-β from subepithelial fibroblasts, whereas KGF appeared to be less important. © 2007 Elsevier B.V. All rights reserved. Keywords: Glucagon like peptide-2; Mucosal injury; VEGF; Subepithelial myofibroblast; Intestinal wound healing; Intestinal cell proliferation
1. Introduction Glucagon like peptide-2 (GLP-2) is a gut hormone that is cosecreted along with GLP-1 from enteroendocrine L-cells (Drucker et al., 1996; Drucker, 2001). A number of recent studies have demonstrated that GLP-2 induces epithelial cell ⁎ Corresponding author. Department of Medicine I, St. Josef-Hospital, RuhrUniversity of Bochum Gudrunstr. 56, D-44791 Bochum, Germany. Tel.: +49 234 509 2332/509 1 (central line); fax: +49 234 509 2309. E-mail address: [email protected] (K. Bulut). 0014-2999/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.ejphar.2007.08.044
proliferation both in vitro and in vivo (Drucker et al., 1996; Jasleen et al., 2002; Ramsanahie et al., 2002). Furthermore, GLP-2 promotes nutrient absorption via expansion of mucosal epithelium by stimulation of crypt cell proliferation and inhibition of apoptosis in the small intestine (Jeppesen et al., 2001; Jasleen et al., 2002). Interestingly, patients with inflammatory bowel disease exhibit increased serum levels of circulating GLP-2 indicating a potentially important role of this peptide. Based on these findings, GLP-2 has been proposed as a new therapeutic strategy for patients with intestinal malabsorption (Scott et al., 1998; Xiao et al., 2000; Jeppesen et al., 2001).
280
K. Bulut et al. / European Journal of Pharmacology 578 (2008) 279–285
However, while the intestinotrophic effects of GLP-2 have uniformly been demonstrated in vitro, in animal studies, and in humans, the underlying mechanisms are yet poorly understood (Jeppesen et al., 2001). In particular, it is rather unclear, by which mechanism GLP-2 improves intestinal wound healing and intestinal repair, given the lack of GLP-2 receptor expression on intestinal epithelial cells (Prasad et al., 2001; Jasleen et al., 2002; Ramsanahie et al., 2002). Mucosal wound healing after superficial injury requires initial migration of epithelial cells adjacent to the wound area across the mucosal defect, a process termed as restitution. Restitution begins within minutes after mucosal injury and appears to be independent of proliferation (Dignass and Podolsky, 1993). Recently, Orskov and colleagues proposed that the local release of keratinocyte growth factor (KGF) from subepithelial myofibroblasts might be responsible for the induction of intestinal cell proliferation by GLP-2 (Orskov et al., 2005). In addition to KGF, a number of other factors are known to be involved in intestinal wound repair. In particular, the role of Transforming Growth Factor beta (TGF-β) in the early phase of wound repair is well documented. Along these lines, TGF-β inhibits proliferation of intestinal epithelial cells and promotes cell migration into defect areas (Moses et al., 1987; Ciacci et al., 1993; Dignass and Podolsky, 1993; Zhu et al., 2005). Additionally, TGFβ has been shown to induce intestinal repair properties both in vitro and in vivo by Beck et al. (2003). Furthermore, recent studies from our group have demonstrated that the GLP-2 effects on intestinal epithelial cells repair are TGF-β dependent (Bulut et al., 2004). Vascular endothelial growth factor (VEGF) is another important mediator of intestinal wound healing. There is strong evidence, that VEGF is a crucial factor for intestinal adaptation and wound healing after small bowel resection shown by Parvadia et al. (in press). We have previously shown in an in vitro wound healing model, that VEGF promotes intestinal cell migration and wound healing in intestinal cell lines (Bulut et al., 2006). However, in addition to its beneficial role in the physiological regulation of tissue repair and wound healing, VEGF may also promote tumor angiogenesis and metastases growth (Ferrer et al., 1999; Cucina et al., 2003). Given the recent reports by Thulesen et al. in rodents that GLP-2 might also promote the progression of colonic cancer, it is important to identify the underlying mechanisms (Thulesen et al., 2004). In fact, if GLP-2 possesses tumor-promoting properties, this would clearly limit its therapeutic usefulness. Therefore, in the present studies, we sought to address the following questions: (1) Does GLP-2 exerts differential effects on wound healing in intestinal cell lines originating from the ileum and colon? (2) Are these effects mediated through subepithelial fibroblasts, and, if so (3) which growth factors are involved in the induction of intestinal cell repair by GLP-2?
2. Materials and methods 2.1. Materials 2.1.1. Cell lines and cell cultures Human colonic fibroblasts (CCD-18Co), human colonic cancer cells (Caco-2) and rat ileal cells (IEC-18) were purchased
from the American Type Culture Collection (ATCC), Rockville, Md., USA. All cells were routinely grown in Dulbecco's modified Eagle's mixture (DMEM) containing 5% (IEC-18) or 10% (Caco-2, CCD-18Co) fetal calf serum. Cells were kept at under 5% CO2 atmosphere and 37 °C. 3-[4, 5-dimethylthiazol-2-yl]-2, 5-diphenyl-tetrazolium bromide (MTT) for the assessment of epithelial cell proliferation by colorimetric analysis and human TGF-β1 were obtained from Sigma Chemical CO, St. Louis, MO, USA. GLP-2 was purchased from Poly Peptide Laboratories, Wolfenbüttel, Germany. 2.1.2. Effect of conditioned media on epithelial cell proliferation All experiments were performed as described previously (Bulut et al., 2004). IEC-18 cells were seeded into 24-well plates (105 cells/well) in the presence of DMEM containing 5% FCS. Caco-2 cells were first cultured for 24 h in DMEM containing 10% FCS and then seeded (5 × 104 cells/well) into 24-well plates. Culture media were replaced with serumdeprived media (containing 0.1% FCS in DMEM) 24 h before each study. Proliferation of epithelial cells was assessed following GLP-2 was added to media of CCD-18Co cells in various concentrations (50 nM–250 nM) which have been incubated for 24 h. The conditioned media, which represents the supernatant media of the CCD-18co cells after 24 h stimulation, was collected and used as stimulus subsequently to assess proliferation or likewise migration. Only the CCD18co cells (= subepithelial myofibroblasts) have been stimulated with GLP-2 to obtain conditioned media, which was subsequently used in Caco-2 and IEC-18 cells. Proliferation was measured using the MTT-test. In brief, 2 mg/ml of MTT was added to the media followed by an incubation period of 4 h. Cells were washed and consecutively dissolved with dimethyl sulfoxide (DMSO). Optical density of the dye was recorded at 550 nm (reference wavelength, 690 nm) using a microplate reader (ELx800G, BIO-TEK Instruments, Inc., Vermont, USA). Experiments for all substances were performed at least in quadruplicate. 2.1.3. In vitro wound assays Wound assays were performed as previously described by Dignass et al. (Dignass and Podolsky, 1993). Confluent monolayers of IEC-18 and Caco-2 cells in 100 × 15 mm Petri Dishes (Falcon™, Becton Dickinson Labware, NJ, USA) were wounded with a razor blade; two wounds approximately 10– 15 mm across the dish were made, separated by about 1.5 cm. Afterwards, cells were washed with PBS and subsequently cultured for 24 h in fresh serum-deprived medium (0.1% FCS). Conditioned media or saline control was added to the media to study the effect of a GLP-2 stimulation of colonic fibroblasts. Monoclonal TGF-β1 and VEGF-A antibodies were purchased from R&D Systems Inc., Minneapolis, USA. Then, a computerbased microscopy imaging system (Axiovision™, Carl-Zeiss, Munich, Germany) was used for the determination of wound healing at 0 h with a Carl-Zeiss™, Axiovert 25 microscope at 400-fold magnification. 24 h later, identification of migration was assessed by quantification of the number of cells observed across the wound, compared to the same frame at 0 h. Several
K. Bulut et al. / European Journal of Pharmacology 578 (2008) 279–285
281
(Berlin, Germany) and Light-Cycler™ (Roche Diagnostics, Germany). 2.2.1. Detection of VEGF-A protein with an enzyme-linked immunoabsorbent assay (ELISA) following stimulation of CCD-18Co with GLP-2 VEGF-A concentration was measured in the media after stimulation of CCD-18Co cells with GLP-2 using a commercial human s-VEGF-1 ELISA-Kit purchased from Bender MedSystems, Vienna, Austria according to the manufacturer's manual. In brief, cells were incubated with GLP-2 (50 nM and 250 nM) for 24 h. Afterwards the conditioned media was taken and the concentration of VEGF was detected using the ELISA-Kit. The detection limit for this ELISA is b8.0 pg/ml.
Fig. 1. GLP-2R mRNA expression in CCD-18Co cells.
wound areas (at least 10) per plate were investigated to quantify migration.
2.2.2. Detection of TGF-β1 protein with an enzyme-linked immunoabsorbent assay (ELISA) following stimulation of CCD-18Co with GLP-2 TGF-β1 protein concentration was measured in the media after stimulation of CCD-18Co cells with GLP-2 using a
2.1.4. Inhibition of GLP-2 induced migration in intestinal monolayers with TGF-β1 and VEGF-A antibodies Following wounding of the monolayers of IEC-18 and Caco-2, GLP-2-stimulated conditioned media was added. Migration was subsequently evaluated in wounded monolayers by microscopic quantification of cells as described above. To assess the role of TGF-β and VEGF-A in the process of migration and intestinal wound healing neutralizing monoclonal antibodies against TGF-β and VEGF-A were added to the GLP-2 conditioned media and subsequently migration was quantified 24 h later as described above. Both antibodies were purchased from R&D Systems Inc., Minneapolis, USA. 2.2. Assessment of GLP-2 induced VEGF-A, KGF and TGF-β1 mRNA expression in CCD-18Co cells using real time PCR The mRNA expression levels of TGF-β1 following incubation with different doses of GLP-2 (50–250 nM) in the exposed epithelial cells were evaluated using real time (RT) PCR. Primers flanking the designed coding region in the TGF-β1 mRNA were designed according to the nucleotide sequences published in Genbank: (# NM000660). (forward primer: 5′ GGTACCTGAACCCGTGTTGCT, reverse primer: 5′ TGTTGCTGTATTTCTGGTACAGCTC) synthesized by TIB MOLBIOL (Berlin, Germany) and Light-Cycler™ (Roche Diagnostics, Germany). The primers flanking the coding region in the KGF mRNA were designed according to the nucleotide sequences published in Genbank: (# NMm60828). (forward primer: 5′ GAAATCAGGACAGTGGCA, reverse primer: 5′ ACAGGAATCCCCTTTTG) and the primers flanking the coding region in the VEGF mRNA were designed according to the nucleotide sequences published in Genbank: (# NMm32977). (forward primer: 5′ GCACCCATGGCAGAAGG, reverse primer: 5′ CTCGATTGGATGGCAGTAGCT), synthesized by TIB MOLBIOL
Fig. 2. Assessment of proliferation in the small and large intestinal cell lines. The proliferation assessment shows a significant and dose dependent increase for the colonic Caco-2 cells with a maximum effect at 250 nm GLP-2 stimulated medium. However, the small intestinal IEC-18 cells were not significantly stimulated by GLP-2 conditioned fibroblast medium. Control was performed using DMEM with containing 0.5% fetal calf serum without supplementation of GLP-2. P b 0.001.
282
K. Bulut et al. / European Journal of Pharmacology 578 (2008) 279–285
and Wilcoxon-Test. Data were considered significant at a P value b 0.05. 3. Results 3.1. Expression of the GLP-2 receptor in CCD-18Co
Fig. 3. Differential effects of conditioned medium from GLP-2 stimulated subepithelial myofibroblasts in small and large bowel epithelial cells. While a decrease of migration in large bowel cells (Caco-2) was observed, there was a significantly increased migration in small bowel cells (IEC-18). Neutralizing antibodies for VEGF-A or TGF-β significantly inhibited the GLP-2 induced migration, indicating that VEGF and TGF-β are responsible for the observed migration. Control was performed with 0.5% fetal calf serum containing medium without any stimulus. ⁎⁎P b 0.01 GLP-2 stimulated vs. control, # P b 0.001 GLP-2 stimulated vs. anti-VEGF/TGF-β.
commercial human TGF-β1 ELISA-Kit purchased from Ray Bio™, Ray Biotech Inc., USA, according to the manufacturer's manual. CCD-18co cells were incubated with GLP-2 (50 nM, 100 nM and 250 nM) for 4 h. Afterwards the conditioned media was taken and the concentration of TGF-β1 was detected using the ELISA-Kit. The detection limit for this ELISA is b80 pg/ml. 2.3. Detection of GLP-2R mRNA expression in CCD-18Co The mRNA expression levels of GLP-2 Receptor in the investigated CCD-18Co cells were evaluated using real time RTPCR (Light-Cycler™). To confirm these results selected probes were transferred to an Ethidiumbromide stained Agarose-gel (2%).
The expression of the GLP-2 receptor mRNA on unstimulated CCD-18Co cells was evaluated using real time RT-PCR (Light-Cycler™). Fig. 1 demonstrates the presence of GLP-2 receptor mRNA with 194 base pairs (bp) in unstimulated CCD18Co cells. To confirm these results selected probes were transferred to an Ethidiumbromide stained Agarose-gel (2%). The Ethidiumbromide stained Agarose-gel confirms the presence of naturally occurring GLP-2 receptor mRNA expression in CCD-18Co subepithelial myofibroblasts. 3.2. Epithelial cell proliferation Conditioned medium from GLP-2-stimulated subepithelial myofibroblasts dose-dependently increased proliferation in Caco2 cells after 72 h of incubation (P b 0.01). Proliferation increased up to 33% at a concentration of 100 nm GLP-2 and increased up to 35% at a concentration of 250 nm GLP-2 compared to the unstimulated control group. In contrast, proliferation of IEC-18 cells was unaffected by 48 h of incubation with GLP-2 conditioned medium, independent of the used concentration of stimulation. The different time points of evaluation between IEC-18 and Caco-2 (48 h vs. 72 h) are caused by the different proliferation speed of this two cell lines (Fig. 2). 3.3. Epithelial cell migration
Data are presented as mean ± standard error of the mean (S.E. M). Differences between the groups were tested with ANOVA
Conditioned medium from GLP-2-stimulated subepithelial myofibroblasts significantly promoted the migration of IEC-18 cells across the wound edge (P b 0.01). Concomitant administration of anti-VEGF-A and anti-TGF-β completely abolished these effects (P b 0.0001 vs. GLP-2 treatment alone) as shown in Fig. 3. In contrast, cell migration was even dose-dependently inhibited by GLP-2 stimulated conditioned media in Caco-2 cells
Fig. 4. Assessment of VEGF-A protein in the medium after stimulation with GLP-2 of CCD-18Co cells. The detection level of the used ELISA is given as b8 pg/ml. The unstimulated CCD-18Co control doesn't exhibit any VEGF production. ⁎⁎P b 0.01 vs. control (control was performed with 0.5% fetal calf serum containing medium without stimulus).
Fig. 5. Assessment of TGF-β1 protein in the medium after stimulation with GLP-2 of CCD-18Co cells. The detection level of the used ELISA is given as b8 pg/ml. The maximum of TGF-β1 secretion was detected at a concentration of 100 nM GLP-2. (Control was performed with 0.5% fetal calf serum containing medium without stimulus). ⁎⁎P b 0.01 vs. control.
2.4. Statistical analysis
K. Bulut et al. / European Journal of Pharmacology 578 (2008) 279–285
(P = 0.0012), and addition of anti-VEGF antibodies and antiTGF-beta antibodies led to a further reduction in cell migration of Caco-2 cells. Both cell lines exhibited significant effects at 100 nM of GLP-2. Supplementation of anti-VEGF antibodies or anti-TGF-beta antibodies to the control group, didn't show any significant difference in migration rates compared to control group (data not shown). 3.4. Effects of VEGF protein release in GLP-2 stimulated CCD-18Co cells After stimulation of CCD-18Co with different concentrations of GLP-2 (50 and 250 nM), the VEGF protein concentrations in the medium were determined by enzymelinked immunoabsorbent assay (ELISA) (Fig. 4). At the end of incubation, the viability of CCD-18co cells was measured to normalize the assessed amount of protein. No significant difference could be detected after incubation. The cell viability was 98.9% for control group, 98.3% for 50 nM GLP-2 and 99.1% for 250 nM GLP-2. The ELISA-Results indicate a
283
significant and dose dependent induction of VEGF release from subepithelial myofibroblasts by GLP-2, consistent with the results of the mRNA detection (Fig. 6). 3.5. Effect of GLP-2 stimulation in CCD-18Co on TGF-β1 protein release After stimulation of CCD-18Co with different concentrations of GLP-2 (50 and 100 nM), the TGF-β1 protein concentrations in the medium were determined by ELISA (Fig. 5). At the end of incubation, the viability of CCD-18co cells was measured to normalize the assessed amount of protein. No significant difference could be detected after incubation. The ELISA-Results revealed a significant and dose dependent induction of TGF-β1 release from subepithelial myofibroblasts stimulated by GLP-2, consistent with the results of the mRNA detection (Fig. 6). 3.6. GLP-2 effects on the mRNA expression of KGF, VEGF-m and TGF-β in CCD-18Co cells The mRNA expression levels of KGF, VEGF and TGF-β1 in CCD-18Co cells exposed to different concentrations of GLP-2 (50, 100 and 250 nM) were evaluated using real time RT-PCR (Light-Cycler®). The melting curve analysis revealed a dose dependent increase of TGF-β and VEGF mRNA transcription after GLP-2 stimulation compared to controls after 4 h (P b 0.01). In contrast, KGF mRNA expression was unchanged after GLP-2 stimulation at both concentrations (Fig. 6). 4. Discussion
Fig. 6. GLP-2 effects on the mRNA expression of KGF, VEGF and TGF-β in CCD-18Co cells. RT-PCR analysis of mRNA expression levels of KGF, VEGF and TGF-β1 in CCD-18Co cells exposed to different concentrations of GLP-2 (50 and 250 nM). Control was with 0.5% fetal calf serum containing medium without stimulus). ⁎⁎P b 0.01 vs. control.
In the present studies we sought to further clarify the downstream mechanisms involved in improved intestinal adaptation by GLP-2. We report that GLP-2 stimulated the release of growth factors, such as TGF-β and VEGF, from subepithelial myofibroblasts, and that these effects are likely to be involved in the induction of intestinal wound healing and epithelial restitution by GLP-2. Our present results are consistent with recent studies by Orskov and colleagues, who demonstrated GLP-2 receptor expression primarily on subepithelial myofibroblasts in rat, mouse, marmoset and human small and large intestine. The development of the myofibroblast cell line CCD18Co, that naturally expresses the GLP-2 receptor, by Orskov and colleagues also allowed us to further study the mechanisms involved in the intestinotrophic effects of GLP-2. Using a combination of real time RT-PCR and ELISA, we found that GLP-2 treatment significantly stimulated the release of VEGF-A and TGF-β1 protein from CCD18Co cells, whereas KGF appeared to be less important in this experimental setting. Another interesting finding from the present studies is that GLP-2 exerts differential effects on cell lines originating from the ileum (IEC-18) and from the colon (Caco-2). Thus, in IEC18 cells, migration was significantly enhanced by GLP-2conditioned medium, whereas proliferation was rather unaffected. In contrast, in Caco-2 cells GLP-2-conditioned medium
284
K. Bulut et al. / European Journal of Pharmacology 578 (2008) 279–285
dose-dependently promoted cell proliferation, but even led to an inhibition of cell migration. This may indicate that the intestinotrophic effects of GLP-2 in the small intestine are primarily mediated by enhanced cell migration, whereas in the colon, the proliferative effects might predominate, in line with previous studies indicating differential effects of GLP-2 in different areas on the intestine (Orskov et al., 2005). We have previously shown, that the effect of GLP-2 induced proliferation in the colon was about 60% of the maximum inducible effect of TGF-α, which was used as positive control for proliferation (Bulut et al., 2004). Alternatively, the differential effects of GLP-2-conditioned medium observed in these two cell lines might be in part due to their different origins (tumor cell line vs. non-malignant epithelial cell line) or species differences (human vs. rat). The observed induction of VEGF-A and TGF-β1 secretion from subepithelial myofibroblasts by GLP-2 provides a novel mechanistic explanation of its proliferative and anti-apoptotic effects observed in previous studies in vitro and in vivo. In fact, there has been a debate as to the potential link between GLP-2 action and improved intestinal adaptation, since the GLP-2 receptor is not expressed on intestinal epithelial cells. Given the expression of GLP-2 receptors on enteroendocrine cells, it has been suggested that GLP-2 might exert its actions through autocrine, paracrine and endocrine mechanisms via yet unknown mediators (Drucker, 2001). The present data showing increased secretion of growth factors from myofibroblasts as well as previous publications showing GLP-2 receptor expression mainly in the subepithelial layer tend to support an alternative model, in which GLP-2 primarily acts on subepithelial myofibroblasts to release secondary mediators (VEGF-A and TGF-β1), which in turn stimulate epithelial cell proliferation or migration depending on their localization in the gut. Our present findings differ from a recent study by Orskov and colleagues who suggested an involvement of KGF in the growth promoting effects of GLP-2. Their observation suggested a co-localisation between the GLP-2 receptor and KGF. Their in vivo results revealed in immunoneutralisation experiments, that neutralisation of KGF abolished the intestinotrophic effects on GLP-2 (Orskov et al., 2005). In contrast, Sams and colleagues failed to observe a regulative role for KGF in fetal human colonic epithelial cells stimulated by GLP-2, similar to our present results in subepithelial myofibroblasts (Sams et al., 2006).We also couldn't see any KGF linked effects, induced by GLP-2. While the reasons underlying the differences between these studies cannot be clarified with certainty, one possible explanation is that KGF is known to exert proliferative actions on the intestinal epithelium on a basal level but probably this proliferative action is unlikely to be regulated by GLP-2. This could explain why treatment with KGF antisera diminished intestinal cell growth during GLP-2 treatment even without a direct stimulatory effect of GLP-2 on KGF release. While the induction of growth factor release by GLP-2 underscores its potential beneficial effects in the treatment of patients with intestinal malabsorption as well as after intestinal injury, e.g. induced by radio-chemotherapy, it also raise
concerns as to the potential induction of malignant cell growth during chronic GLP-2 administration. Thus, both VEGF-A and TGF-β are well known to promote the proliferation of tumor cells and to induce neoangiogenesis and metastatic spreading (Leung et al., 1989; Alon et al., 1995; Ferrara, 1996). This would also explain the findings by Thulesen and colleagues, who reported that chronic GLP-2 administration promoted the growth of colonic neoplasms in mice (Thulesen et al., 2004). Therefore, it seems important to rule out any potential carcinogenic effects of GLP-2 before this compound or its analogues are being introduced for the treatment of patients with intestinal disorders. In conclusion, the present studies demonstrate that GLP-2 induces the release of VEGF and TGF-β from subepithelial myofibroblasts. These effects are likely to explain the intestinotrophic and reparative actions of GLP-2 shown previously in vitro and in vivo. Acknowledgements This study was supported by the FoRUM-Grant (Grant No.: F 502-2006). The excellent technical assistance of Ilka Werner and Rainer Lebert is greatly acknowledged. References Alon, T., Hemo, I., Itin, A., Pe`er, J., Stone, J., Keshet, E., 1995. Vascular endothelial growth factor acts as a survival factor for newly formed retinal vessels and has implications for retinopathy of prematurity. Nat. Med. 1, 1024–1028. Beck, P.L., Rosenberg, I., Xavier, R.J., Koh, T., Wong, J.F., Podolsky, D.K., 2003. Transforming growth factor-beta mediates intestinal healing and susceptibility to injury in vitro and in vivo through epithelial cells. Am. J. Pathol. 162, 597–608. Bulut, K., Meier, J.J., Ansorge, N., Felderbauer, P., Schmitz, F., Hoffmann, P., Schmidt, W.E., Gallwitz, B., 2004. Glucagon-like peptide 2 improves intestinal wound healing through induction of epithelial cell migration in vitro — evidence for a TGF-beta mediated effect. Reg. Peptides 121, 137–143. Bulut, K., Pennartz, C., Felderbauer, P., Ansorge, N., Banasch, M., Schmitz, F., Schmidt, W.E., Hoffmann, P., 2006. Vascular endothelial growth factor (VEGF 164) ameliorates intestinal epithelial injury in vitro in IEC-18 and Caco-2 monolayers via induction of TGF-beta release from epithelial cells. Scand. J. Gastroenterol. 41, 1–6. Ciacci, C., Lind, S., Podolsky, D.K., 1993. Transforming growth factor regulation of migration in wounded rat intestinal epithelial monolayers. Gastroenterology 105, 93–101. Cucina, A., Borelli, V., Randone, B., Coluccia, P., Sapienza, P., Cavallaro, A., 2003. Vascular endothelial growth factor increases the migration of smooth muscle cells through the mediation of growth factors released by endothelial cells. J. Surg. Res. 109, 16–23. Dignass, A.U., Podolsky, D., 1993. Cytokine modulation of intestinal epithelial cell restitution: central role of transforming growth factor. Gastroenterology 105, 1323–1332. Drucker, D.J., 2001. Minireview, the Glucagon-like peptides. Endocrinology 142, 521–527. Drucker, D.J., Ehrlich, P., Asa, S.L., Brubaker, P.L., 1996. Induction of intestinal epithelial proliferation by glucagon-like peptide 2. Proc. Natl. Acad. Sci. 23, 7911–7916. Ferrara, N., 1996. Vascular endothelial growth factor. Eur. J. Cancer. 32, 2413–2422. Ferrer, F.A., Miller, L., Lindquist, R., et al., 1999. Expression of vascular endothelial growth factor receptors in human prostate cancer. Urology 54, 567–572.
K. Bulut et al. / European Journal of Pharmacology 578 (2008) 279–285 Jasleen, J., Ashley, S., Shimoda, N., Zinner, M.J., Whang, E.E., 2002. Glucagon-like peptide 2 stimulates intestinal epithelial proliferation in vitro. Dig. Dis. Sci. 47, 1135–1140. Jeppesen, P.B., Hartmann, B., Thulesen, J., Graff, J., Lohmann, J., Hansen, B.S., Tofteng, F., Paulsen, S.S., Madsen, J.L., Holst, J.J., Mortensen, P.B., 2001. Glucagon-like peptide 2 improves nutrient adsorption and nutritional status in short-bowel patients with no colon. Gastroenterology 120, 806–815. Leung, D.W., Cachianes, G., Kuang, W.J., Goeddel, D.V., Ferrara, N., 1989. Vascular endothelial growth factor is a secreted angiogenic mitogen. Science 246, 1306–1309. Moses, H.L., Coffey, R.J., Leof, E.B., Lyons, R.M., Keski-Oka, J., 1987. Transforming growth factor regulation of cell proliferation. J. Cell Physiol. 5, 1–7. Orskov, C., Hartmann, B., Poulsen, S.S., Thulesen, J., Hare, K.J., Holst, J.J., 2005. GLP-2 stimulates colonic growth via KGF, released by subepithelial myofibroblasts with GLP-2 receptors. Regul. Pept. 124, 05–12. Parvadia, J.K., Kesnawi, S.G., Vaikunth, S., Maldonado, A.R., Marwan, A., Stehr, W., Uzvolgyi, E., Warner, B.W., Yamano, S., Taichman, N., Crombleholme, T.M., in press. Role of vascular endothelial growth factor (VEGF) in small bowel adaptation after resection: the adaptive response is angiogenesis dependent. Am. J. Physiol. Gastrointest. Liver Physiol. (Electronic publication ahead of print).
285
Prasad, R., Alavi, K., Schwartz, M.Z., 2001. GLP-2alpha accelerates recovery of mucosal absorptive function after intestinal ischemia/reperfusion. J. Pediatr. Surg. 36, 570–572. Ramsanahie, A.P., Perez, A., Duensing, A.U., Zinner, M.J., Ashley, S.W., Whang, E.E., 2002. Glucagon-like peptide 2 enhances intestinal epithelial restitution. J. Surg. Res. 107, 44–49. Sams, A., Hastrup, S., Andersen, M., Thim, L., 2006. Naturally occuring glucagon-like peptide-2 (GLP-2) receptors in human intestinal cell lines. Eur. J. Pharmacol. 532, 18–23. Scott, R.B., Kirk, D., MacNaughton, W.K., Meddings, J.B., 1998. GLP-2 augments the adaptive response to massive intestinal resection in rat. Am. J. Physiol. Endocrinol. Metabol. 275, G911–G921. Thulesen, J., Hartmann, B., Hare, K.J., Kissow, H., Orskov, C., Holst, J.J., Poulsen, S.S., 2004. Glucagon-like peptide 2 (GLP-2) accelerates the growth of colonic neoplasms in mice. Gut 53, 1145–1150. Xiao, Q., Boushey, R., Cino, M., Drucker, D.J., Brubaker, P.L., 2000. Circulating levels of glukagon-like peptide-2 in human subjects with inflammatory bowel disease. Am. J. Physiol. Regulatory Integrative Comp. Physiol. 278, R1057–R1063. Zhu, C., Ying, D., Zhou, D., Mi, J., Zhang, W., Chang, Q., Li, L., 2005. Expression of TGF-beta 1 in smooth muscle cells regulates endothelial progenitor cells migration and differentiation. J. Surg. Res. 12, 151–156.
European Journal of Pharmacology 578 (2008) 279 – 285 www.elsevier.com/locate/ejphar
Glucagon like peptide-2 induces intestinal restitution through VEGF release from subepithelial myofibroblasts Kerem Bulut a,⁎, Christian Pennartz a , Peter Felderbauer a , Juris J. Meier a , Matthias Banasch a , Daniel Bulut b , Frank Schmitz a , Wolfgang E. Schmidt a , Peter Hoffmann c a b
Department of Medicine I, St. Josef-Hospital, Ruhr-University Bochum, Germany Department of Medicine II, St. Josef-Hospital, Ruhr-University Bochum, Germany c Department of Medicine, Kliniken-Essen Mitte, Essen, Germany
Received 5 January 2007; received in revised form 21 August 2007; accepted 24 August 2007 Available online 15 September 2007
Abstract Glucagon like peptide-2 (GLP-2) exerts intestinotrophic actions, but the underlying mechanisms are still a matter of debate. Recent studies demonstrated the expression of the GLP-2 receptor on fibroblasts located in the subepithelial tissue, where it might induce the release of growth factors such as keratinocyte growth factor (KGF) or vascular endothelial growth factor (VEGF). Therefore, in the present studies we sought to elucidate the downstream mechanisms involved in improved intestinal adaptation by GLP-2. Human colonic fibroblasts (CCD-18Co), human colonic cancer cells (Caco-2 cells) and rat ileum IEC-18 cells were used. GLP-2 receptor mRNA expression was determined using real time RTPCR. Conditioned media from CCD-18Co cells were obtained following incubation with GLP-2 (50–250 nM) for 24 h. Cell viability was assessed by a 3-[4, 5-dimethylthiazol-2-yl]-2, 5-diphenyl-tetrazolium bromide (MTT)-assay, and wound healing was determined with an established migration-assay. Transforming Growth Factor beta (TGF-β), VEGF and KGF mRNA levels were determined by RT-PCR. Protein levels of VEGF and TGF-β in CCD-18Co cells following GLP-2 stimulation were determined using ELISA. Neutralizing TGF-β and VEGF-A antibodies were utilized to assess the role of TGF-β and VEGF-A in the process of wound healing. GLP-2 receptor expression was detected in CCD-18Co cells. Conditioned media from CCD-18Co cells dose-dependently induced proliferation in Caco-2 cells, but not in IEC-18 cells. Conditioned media also enhanced cell migration in IEC-18 cells (P b 0.01), while migration was even inhibited in Caco-2 cells (P b 0.0012). GLP2 significantly stimulated mRNA expression of VEGF and TGF-β, but not of KGF in CCD-18Co. The migratory effects of GLP-2 were completely abolished in the presence of TGF-β and VEGF-A antibodies. GLP-2 exerts differential effects on the epithelium of the small intestine and the colon. Thus, in small intestinal cells GLP-2 stimulates wound repair, whereas no such effects were observed in colonic cells. The mechanisms underlying GLP-2 induced intestinal wound repair seem to involve the secretion of VEGF and, subsequently, TGF-β from subepithelial fibroblasts, whereas KGF appeared to be less important. © 2007 Elsevier B.V. All rights reserved. Keywords: Glucagon like peptide-2; Mucosal injury; VEGF; Subepithelial myofibroblast; Intestinal wound healing; Intestinal cell proliferation
1. Introduction Glucagon like peptide-2 (GLP-2) is a gut hormone that is cosecreted along with GLP-1 from enteroendocrine L-cells (Drucker et al., 1996; Drucker, 2001). A number of recent studies have demonstrated that GLP-2 induces epithelial cell ⁎ Corresponding author. Department of Medicine I, St. Josef-Hospital, RuhrUniversity of Bochum Gudrunstr. 56, D-44791 Bochum, Germany. Tel.: +49 234 509 2332/509 1 (central line); fax: +49 234 509 2309. E-mail address: [email protected] (K. Bulut). 0014-2999/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.ejphar.2007.08.044
proliferation both in vitro and in vivo (Drucker et al., 1996; Jasleen et al., 2002; Ramsanahie et al., 2002). Furthermore, GLP-2 promotes nutrient absorption via expansion of mucosal epithelium by stimulation of crypt cell proliferation and inhibition of apoptosis in the small intestine (Jeppesen et al., 2001; Jasleen et al., 2002). Interestingly, patients with inflammatory bowel disease exhibit increased serum levels of circulating GLP-2 indicating a potentially important role of this peptide. Based on these findings, GLP-2 has been proposed as a new therapeutic strategy for patients with intestinal malabsorption (Scott et al., 1998; Xiao et al., 2000; Jeppesen et al., 2001).
280
K. Bulut et al. / European Journal of Pharmacology 578 (2008) 279–285
However, while the intestinotrophic effects of GLP-2 have uniformly been demonstrated in vitro, in animal studies, and in humans, the underlying mechanisms are yet poorly understood (Jeppesen et al., 2001). In particular, it is rather unclear, by which mechanism GLP-2 improves intestinal wound healing and intestinal repair, given the lack of GLP-2 receptor expression on intestinal epithelial cells (Prasad et al., 2001; Jasleen et al., 2002; Ramsanahie et al., 2002). Mucosal wound healing after superficial injury requires initial migration of epithelial cells adjacent to the wound area across the mucosal defect, a process termed as restitution. Restitution begins within minutes after mucosal injury and appears to be independent of proliferation (Dignass and Podolsky, 1993). Recently, Orskov and colleagues proposed that the local release of keratinocyte growth factor (KGF) from subepithelial myofibroblasts might be responsible for the induction of intestinal cell proliferation by GLP-2 (Orskov et al., 2005). In addition to KGF, a number of other factors are known to be involved in intestinal wound repair. In particular, the role of Transforming Growth Factor beta (TGF-β) in the early phase of wound repair is well documented. Along these lines, TGF-β inhibits proliferation of intestinal epithelial cells and promotes cell migration into defect areas (Moses et al., 1987; Ciacci et al., 1993; Dignass and Podolsky, 1993; Zhu et al., 2005). Additionally, TGFβ has been shown to induce intestinal repair properties both in vitro and in vivo by Beck et al. (2003). Furthermore, recent studies from our group have demonstrated that the GLP-2 effects on intestinal epithelial cells repair are TGF-β dependent (Bulut et al., 2004). Vascular endothelial growth factor (VEGF) is another important mediator of intestinal wound healing. There is strong evidence, that VEGF is a crucial factor for intestinal adaptation and wound healing after small bowel resection shown by Parvadia et al. (in press). We have previously shown in an in vitro wound healing model, that VEGF promotes intestinal cell migration and wound healing in intestinal cell lines (Bulut et al., 2006). However, in addition to its beneficial role in the physiological regulation of tissue repair and wound healing, VEGF may also promote tumor angiogenesis and metastases growth (Ferrer et al., 1999; Cucina et al., 2003). Given the recent reports by Thulesen et al. in rodents that GLP-2 might also promote the progression of colonic cancer, it is important to identify the underlying mechanisms (Thulesen et al., 2004). In fact, if GLP-2 possesses tumor-promoting properties, this would clearly limit its therapeutic usefulness. Therefore, in the present studies, we sought to address the following questions: (1) Does GLP-2 exerts differential effects on wound healing in intestinal cell lines originating from the ileum and colon? (2) Are these effects mediated through subepithelial fibroblasts, and, if so (3) which growth factors are involved in the induction of intestinal cell repair by GLP-2?
2. Materials and methods 2.1. Materials 2.1.1. Cell lines and cell cultures Human colonic fibroblasts (CCD-18Co), human colonic cancer cells (Caco-2) and rat ileal cells (IEC-18) were purchased
from the American Type Culture Collection (ATCC), Rockville, Md., USA. All cells were routinely grown in Dulbecco's modified Eagle's mixture (DMEM) containing 5% (IEC-18) or 10% (Caco-2, CCD-18Co) fetal calf serum. Cells were kept at under 5% CO2 atmosphere and 37 °C. 3-[4, 5-dimethylthiazol-2-yl]-2, 5-diphenyl-tetrazolium bromide (MTT) for the assessment of epithelial cell proliferation by colorimetric analysis and human TGF-β1 were obtained from Sigma Chemical CO, St. Louis, MO, USA. GLP-2 was purchased from Poly Peptide Laboratories, Wolfenbüttel, Germany. 2.1.2. Effect of conditioned media on epithelial cell proliferation All experiments were performed as described previously (Bulut et al., 2004). IEC-18 cells were seeded into 24-well plates (105 cells/well) in the presence of DMEM containing 5% FCS. Caco-2 cells were first cultured for 24 h in DMEM containing 10% FCS and then seeded (5 × 104 cells/well) into 24-well plates. Culture media were replaced with serumdeprived media (containing 0.1% FCS in DMEM) 24 h before each study. Proliferation of epithelial cells was assessed following GLP-2 was added to media of CCD-18Co cells in various concentrations (50 nM–250 nM) which have been incubated for 24 h. The conditioned media, which represents the supernatant media of the CCD-18co cells after 24 h stimulation, was collected and used as stimulus subsequently to assess proliferation or likewise migration. Only the CCD18co cells (= subepithelial myofibroblasts) have been stimulated with GLP-2 to obtain conditioned media, which was subsequently used in Caco-2 and IEC-18 cells. Proliferation was measured using the MTT-test. In brief, 2 mg/ml of MTT was added to the media followed by an incubation period of 4 h. Cells were washed and consecutively dissolved with dimethyl sulfoxide (DMSO). Optical density of the dye was recorded at 550 nm (reference wavelength, 690 nm) using a microplate reader (ELx800G, BIO-TEK Instruments, Inc., Vermont, USA). Experiments for all substances were performed at least in quadruplicate. 2.1.3. In vitro wound assays Wound assays were performed as previously described by Dignass et al. (Dignass and Podolsky, 1993). Confluent monolayers of IEC-18 and Caco-2 cells in 100 × 15 mm Petri Dishes (Falcon™, Becton Dickinson Labware, NJ, USA) were wounded with a razor blade; two wounds approximately 10– 15 mm across the dish were made, separated by about 1.5 cm. Afterwards, cells were washed with PBS and subsequently cultured for 24 h in fresh serum-deprived medium (0.1% FCS). Conditioned media or saline control was added to the media to study the effect of a GLP-2 stimulation of colonic fibroblasts. Monoclonal TGF-β1 and VEGF-A antibodies were purchased from R&D Systems Inc., Minneapolis, USA. Then, a computerbased microscopy imaging system (Axiovision™, Carl-Zeiss, Munich, Germany) was used for the determination of wound healing at 0 h with a Carl-Zeiss™, Axiovert 25 microscope at 400-fold magnification. 24 h later, identification of migration was assessed by quantification of the number of cells observed across the wound, compared to the same frame at 0 h. Several
K. Bulut et al. / European Journal of Pharmacology 578 (2008) 279–285
281
(Berlin, Germany) and Light-Cycler™ (Roche Diagnostics, Germany). 2.2.1. Detection of VEGF-A protein with an enzyme-linked immunoabsorbent assay (ELISA) following stimulation of CCD-18Co with GLP-2 VEGF-A concentration was measured in the media after stimulation of CCD-18Co cells with GLP-2 using a commercial human s-VEGF-1 ELISA-Kit purchased from Bender MedSystems, Vienna, Austria according to the manufacturer's manual. In brief, cells were incubated with GLP-2 (50 nM and 250 nM) for 24 h. Afterwards the conditioned media was taken and the concentration of VEGF was detected using the ELISA-Kit. The detection limit for this ELISA is b8.0 pg/ml.
Fig. 1. GLP-2R mRNA expression in CCD-18Co cells.
wound areas (at least 10) per plate were investigated to quantify migration.
2.2.2. Detection of TGF-β1 protein with an enzyme-linked immunoabsorbent assay (ELISA) following stimulation of CCD-18Co with GLP-2 TGF-β1 protein concentration was measured in the media after stimulation of CCD-18Co cells with GLP-2 using a
2.1.4. Inhibition of GLP-2 induced migration in intestinal monolayers with TGF-β1 and VEGF-A antibodies Following wounding of the monolayers of IEC-18 and Caco-2, GLP-2-stimulated conditioned media was added. Migration was subsequently evaluated in wounded monolayers by microscopic quantification of cells as described above. To assess the role of TGF-β and VEGF-A in the process of migration and intestinal wound healing neutralizing monoclonal antibodies against TGF-β and VEGF-A were added to the GLP-2 conditioned media and subsequently migration was quantified 24 h later as described above. Both antibodies were purchased from R&D Systems Inc., Minneapolis, USA. 2.2. Assessment of GLP-2 induced VEGF-A, KGF and TGF-β1 mRNA expression in CCD-18Co cells using real time PCR The mRNA expression levels of TGF-β1 following incubation with different doses of GLP-2 (50–250 nM) in the exposed epithelial cells were evaluated using real time (RT) PCR. Primers flanking the designed coding region in the TGF-β1 mRNA were designed according to the nucleotide sequences published in Genbank: (# NM000660). (forward primer: 5′ GGTACCTGAACCCGTGTTGCT, reverse primer: 5′ TGTTGCTGTATTTCTGGTACAGCTC) synthesized by TIB MOLBIOL (Berlin, Germany) and Light-Cycler™ (Roche Diagnostics, Germany). The primers flanking the coding region in the KGF mRNA were designed according to the nucleotide sequences published in Genbank: (# NMm60828). (forward primer: 5′ GAAATCAGGACAGTGGCA, reverse primer: 5′ ACAGGAATCCCCTTTTG) and the primers flanking the coding region in the VEGF mRNA were designed according to the nucleotide sequences published in Genbank: (# NMm32977). (forward primer: 5′ GCACCCATGGCAGAAGG, reverse primer: 5′ CTCGATTGGATGGCAGTAGCT), synthesized by TIB MOLBIOL
Fig. 2. Assessment of proliferation in the small and large intestinal cell lines. The proliferation assessment shows a significant and dose dependent increase for the colonic Caco-2 cells with a maximum effect at 250 nm GLP-2 stimulated medium. However, the small intestinal IEC-18 cells were not significantly stimulated by GLP-2 conditioned fibroblast medium. Control was performed using DMEM with containing 0.5% fetal calf serum without supplementation of GLP-2. P b 0.001.
282
K. Bulut et al. / European Journal of Pharmacology 578 (2008) 279–285
and Wilcoxon-Test. Data were considered significant at a P value b 0.05. 3. Results 3.1. Expression of the GLP-2 receptor in CCD-18Co
Fig. 3. Differential effects of conditioned medium from GLP-2 stimulated subepithelial myofibroblasts in small and large bowel epithelial cells. While a decrease of migration in large bowel cells (Caco-2) was observed, there was a significantly increased migration in small bowel cells (IEC-18). Neutralizing antibodies for VEGF-A or TGF-β significantly inhibited the GLP-2 induced migration, indicating that VEGF and TGF-β are responsible for the observed migration. Control was performed with 0.5% fetal calf serum containing medium without any stimulus. ⁎⁎P b 0.01 GLP-2 stimulated vs. control, # P b 0.001 GLP-2 stimulated vs. anti-VEGF/TGF-β.
commercial human TGF-β1 ELISA-Kit purchased from Ray Bio™, Ray Biotech Inc., USA, according to the manufacturer's manual. CCD-18co cells were incubated with GLP-2 (50 nM, 100 nM and 250 nM) for 4 h. Afterwards the conditioned media was taken and the concentration of TGF-β1 was detected using the ELISA-Kit. The detection limit for this ELISA is b80 pg/ml. 2.3. Detection of GLP-2R mRNA expression in CCD-18Co The mRNA expression levels of GLP-2 Receptor in the investigated CCD-18Co cells were evaluated using real time RTPCR (Light-Cycler™). To confirm these results selected probes were transferred to an Ethidiumbromide stained Agarose-gel (2%).
The expression of the GLP-2 receptor mRNA on unstimulated CCD-18Co cells was evaluated using real time RT-PCR (Light-Cycler™). Fig. 1 demonstrates the presence of GLP-2 receptor mRNA with 194 base pairs (bp) in unstimulated CCD18Co cells. To confirm these results selected probes were transferred to an Ethidiumbromide stained Agarose-gel (2%). The Ethidiumbromide stained Agarose-gel confirms the presence of naturally occurring GLP-2 receptor mRNA expression in CCD-18Co subepithelial myofibroblasts. 3.2. Epithelial cell proliferation Conditioned medium from GLP-2-stimulated subepithelial myofibroblasts dose-dependently increased proliferation in Caco2 cells after 72 h of incubation (P b 0.01). Proliferation increased up to 33% at a concentration of 100 nm GLP-2 and increased up to 35% at a concentration of 250 nm GLP-2 compared to the unstimulated control group. In contrast, proliferation of IEC-18 cells was unaffected by 48 h of incubation with GLP-2 conditioned medium, independent of the used concentration of stimulation. The different time points of evaluation between IEC-18 and Caco-2 (48 h vs. 72 h) are caused by the different proliferation speed of this two cell lines (Fig. 2). 3.3. Epithelial cell migration
Data are presented as mean ± standard error of the mean (S.E. M). Differences between the groups were tested with ANOVA
Conditioned medium from GLP-2-stimulated subepithelial myofibroblasts significantly promoted the migration of IEC-18 cells across the wound edge (P b 0.01). Concomitant administration of anti-VEGF-A and anti-TGF-β completely abolished these effects (P b 0.0001 vs. GLP-2 treatment alone) as shown in Fig. 3. In contrast, cell migration was even dose-dependently inhibited by GLP-2 stimulated conditioned media in Caco-2 cells
Fig. 4. Assessment of VEGF-A protein in the medium after stimulation with GLP-2 of CCD-18Co cells. The detection level of the used ELISA is given as b8 pg/ml. The unstimulated CCD-18Co control doesn't exhibit any VEGF production. ⁎⁎P b 0.01 vs. control (control was performed with 0.5% fetal calf serum containing medium without stimulus).
Fig. 5. Assessment of TGF-β1 protein in the medium after stimulation with GLP-2 of CCD-18Co cells. The detection level of the used ELISA is given as b8 pg/ml. The maximum of TGF-β1 secretion was detected at a concentration of 100 nM GLP-2. (Control was performed with 0.5% fetal calf serum containing medium without stimulus). ⁎⁎P b 0.01 vs. control.
2.4. Statistical analysis
K. Bulut et al. / European Journal of Pharmacology 578 (2008) 279–285
(P = 0.0012), and addition of anti-VEGF antibodies and antiTGF-beta antibodies led to a further reduction in cell migration of Caco-2 cells. Both cell lines exhibited significant effects at 100 nM of GLP-2. Supplementation of anti-VEGF antibodies or anti-TGF-beta antibodies to the control group, didn't show any significant difference in migration rates compared to control group (data not shown). 3.4. Effects of VEGF protein release in GLP-2 stimulated CCD-18Co cells After stimulation of CCD-18Co with different concentrations of GLP-2 (50 and 250 nM), the VEGF protein concentrations in the medium were determined by enzymelinked immunoabsorbent assay (ELISA) (Fig. 4). At the end of incubation, the viability of CCD-18co cells was measured to normalize the assessed amount of protein. No significant difference could be detected after incubation. The cell viability was 98.9% for control group, 98.3% for 50 nM GLP-2 and 99.1% for 250 nM GLP-2. The ELISA-Results indicate a
283
significant and dose dependent induction of VEGF release from subepithelial myofibroblasts by GLP-2, consistent with the results of the mRNA detection (Fig. 6). 3.5. Effect of GLP-2 stimulation in CCD-18Co on TGF-β1 protein release After stimulation of CCD-18Co with different concentrations of GLP-2 (50 and 100 nM), the TGF-β1 protein concentrations in the medium were determined by ELISA (Fig. 5). At the end of incubation, the viability of CCD-18co cells was measured to normalize the assessed amount of protein. No significant difference could be detected after incubation. The ELISA-Results revealed a significant and dose dependent induction of TGF-β1 release from subepithelial myofibroblasts stimulated by GLP-2, consistent with the results of the mRNA detection (Fig. 6). 3.6. GLP-2 effects on the mRNA expression of KGF, VEGF-m and TGF-β in CCD-18Co cells The mRNA expression levels of KGF, VEGF and TGF-β1 in CCD-18Co cells exposed to different concentrations of GLP-2 (50, 100 and 250 nM) were evaluated using real time RT-PCR (Light-Cycler®). The melting curve analysis revealed a dose dependent increase of TGF-β and VEGF mRNA transcription after GLP-2 stimulation compared to controls after 4 h (P b 0.01). In contrast, KGF mRNA expression was unchanged after GLP-2 stimulation at both concentrations (Fig. 6). 4. Discussion
Fig. 6. GLP-2 effects on the mRNA expression of KGF, VEGF and TGF-β in CCD-18Co cells. RT-PCR analysis of mRNA expression levels of KGF, VEGF and TGF-β1 in CCD-18Co cells exposed to different concentrations of GLP-2 (50 and 250 nM). Control was with 0.5% fetal calf serum containing medium without stimulus). ⁎⁎P b 0.01 vs. control.
In the present studies we sought to further clarify the downstream mechanisms involved in improved intestinal adaptation by GLP-2. We report that GLP-2 stimulated the release of growth factors, such as TGF-β and VEGF, from subepithelial myofibroblasts, and that these effects are likely to be involved in the induction of intestinal wound healing and epithelial restitution by GLP-2. Our present results are consistent with recent studies by Orskov and colleagues, who demonstrated GLP-2 receptor expression primarily on subepithelial myofibroblasts in rat, mouse, marmoset and human small and large intestine. The development of the myofibroblast cell line CCD18Co, that naturally expresses the GLP-2 receptor, by Orskov and colleagues also allowed us to further study the mechanisms involved in the intestinotrophic effects of GLP-2. Using a combination of real time RT-PCR and ELISA, we found that GLP-2 treatment significantly stimulated the release of VEGF-A and TGF-β1 protein from CCD18Co cells, whereas KGF appeared to be less important in this experimental setting. Another interesting finding from the present studies is that GLP-2 exerts differential effects on cell lines originating from the ileum (IEC-18) and from the colon (Caco-2). Thus, in IEC18 cells, migration was significantly enhanced by GLP-2conditioned medium, whereas proliferation was rather unaffected. In contrast, in Caco-2 cells GLP-2-conditioned medium
284
K. Bulut et al. / European Journal of Pharmacology 578 (2008) 279–285
dose-dependently promoted cell proliferation, but even led to an inhibition of cell migration. This may indicate that the intestinotrophic effects of GLP-2 in the small intestine are primarily mediated by enhanced cell migration, whereas in the colon, the proliferative effects might predominate, in line with previous studies indicating differential effects of GLP-2 in different areas on the intestine (Orskov et al., 2005). We have previously shown, that the effect of GLP-2 induced proliferation in the colon was about 60% of the maximum inducible effect of TGF-α, which was used as positive control for proliferation (Bulut et al., 2004). Alternatively, the differential effects of GLP-2-conditioned medium observed in these two cell lines might be in part due to their different origins (tumor cell line vs. non-malignant epithelial cell line) or species differences (human vs. rat). The observed induction of VEGF-A and TGF-β1 secretion from subepithelial myofibroblasts by GLP-2 provides a novel mechanistic explanation of its proliferative and anti-apoptotic effects observed in previous studies in vitro and in vivo. In fact, there has been a debate as to the potential link between GLP-2 action and improved intestinal adaptation, since the GLP-2 receptor is not expressed on intestinal epithelial cells. Given the expression of GLP-2 receptors on enteroendocrine cells, it has been suggested that GLP-2 might exert its actions through autocrine, paracrine and endocrine mechanisms via yet unknown mediators (Drucker, 2001). The present data showing increased secretion of growth factors from myofibroblasts as well as previous publications showing GLP-2 receptor expression mainly in the subepithelial layer tend to support an alternative model, in which GLP-2 primarily acts on subepithelial myofibroblasts to release secondary mediators (VEGF-A and TGF-β1), which in turn stimulate epithelial cell proliferation or migration depending on their localization in the gut. Our present findings differ from a recent study by Orskov and colleagues who suggested an involvement of KGF in the growth promoting effects of GLP-2. Their observation suggested a co-localisation between the GLP-2 receptor and KGF. Their in vivo results revealed in immunoneutralisation experiments, that neutralisation of KGF abolished the intestinotrophic effects on GLP-2 (Orskov et al., 2005). In contrast, Sams and colleagues failed to observe a regulative role for KGF in fetal human colonic epithelial cells stimulated by GLP-2, similar to our present results in subepithelial myofibroblasts (Sams et al., 2006).We also couldn't see any KGF linked effects, induced by GLP-2. While the reasons underlying the differences between these studies cannot be clarified with certainty, one possible explanation is that KGF is known to exert proliferative actions on the intestinal epithelium on a basal level but probably this proliferative action is unlikely to be regulated by GLP-2. This could explain why treatment with KGF antisera diminished intestinal cell growth during GLP-2 treatment even without a direct stimulatory effect of GLP-2 on KGF release. While the induction of growth factor release by GLP-2 underscores its potential beneficial effects in the treatment of patients with intestinal malabsorption as well as after intestinal injury, e.g. induced by radio-chemotherapy, it also raise
concerns as to the potential induction of malignant cell growth during chronic GLP-2 administration. Thus, both VEGF-A and TGF-β are well known to promote the proliferation of tumor cells and to induce neoangiogenesis and metastatic spreading (Leung et al., 1989; Alon et al., 1995; Ferrara, 1996). This would also explain the findings by Thulesen and colleagues, who reported that chronic GLP-2 administration promoted the growth of colonic neoplasms in mice (Thulesen et al., 2004). Therefore, it seems important to rule out any potential carcinogenic effects of GLP-2 before this compound or its analogues are being introduced for the treatment of patients with intestinal disorders. In conclusion, the present studies demonstrate that GLP-2 induces the release of VEGF and TGF-β from subepithelial myofibroblasts. These effects are likely to explain the intestinotrophic and reparative actions of GLP-2 shown previously in vitro and in vivo. Acknowledgements This study was supported by the FoRUM-Grant (Grant No.: F 502-2006). The excellent technical assistance of Ilka Werner and Rainer Lebert is greatly acknowledged. References Alon, T., Hemo, I., Itin, A., Pe`er, J., Stone, J., Keshet, E., 1995. Vascular endothelial growth factor acts as a survival factor for newly formed retinal vessels and has implications for retinopathy of prematurity. Nat. Med. 1, 1024–1028. Beck, P.L., Rosenberg, I., Xavier, R.J., Koh, T., Wong, J.F., Podolsky, D.K., 2003. Transforming growth factor-beta mediates intestinal healing and susceptibility to injury in vitro and in vivo through epithelial cells. Am. J. Pathol. 162, 597–608. Bulut, K., Meier, J.J., Ansorge, N., Felderbauer, P., Schmitz, F., Hoffmann, P., Schmidt, W.E., Gallwitz, B., 2004. Glucagon-like peptide 2 improves intestinal wound healing through induction of epithelial cell migration in vitro — evidence for a TGF-beta mediated effect. Reg. Peptides 121, 137–143. Bulut, K., Pennartz, C., Felderbauer, P., Ansorge, N., Banasch, M., Schmitz, F., Schmidt, W.E., Hoffmann, P., 2006. Vascular endothelial growth factor (VEGF 164) ameliorates intestinal epithelial injury in vitro in IEC-18 and Caco-2 monolayers via induction of TGF-beta release from epithelial cells. Scand. J. Gastroenterol. 41, 1–6. Ciacci, C., Lind, S., Podolsky, D.K., 1993. Transforming growth factor regulation of migration in wounded rat intestinal epithelial monolayers. Gastroenterology 105, 93–101. Cucina, A., Borelli, V., Randone, B., Coluccia, P., Sapienza, P., Cavallaro, A., 2003. Vascular endothelial growth factor increases the migration of smooth muscle cells through the mediation of growth factors released by endothelial cells. J. Surg. Res. 109, 16–23. Dignass, A.U., Podolsky, D., 1993. Cytokine modulation of intestinal epithelial cell restitution: central role of transforming growth factor. Gastroenterology 105, 1323–1332. Drucker, D.J., 2001. Minireview, the Glucagon-like peptides. Endocrinology 142, 521–527. Drucker, D.J., Ehrlich, P., Asa, S.L., Brubaker, P.L., 1996. Induction of intestinal epithelial proliferation by glucagon-like peptide 2. Proc. Natl. Acad. Sci. 23, 7911–7916. Ferrara, N., 1996. Vascular endothelial growth factor. Eur. J. Cancer. 32, 2413–2422. Ferrer, F.A., Miller, L., Lindquist, R., et al., 1999. Expression of vascular endothelial growth factor receptors in human prostate cancer. Urology 54, 567–572.
K. Bulut et al. / European Journal of Pharmacology 578 (2008) 279–285 Jasleen, J., Ashley, S., Shimoda, N., Zinner, M.J., Whang, E.E., 2002. Glucagon-like peptide 2 stimulates intestinal epithelial proliferation in vitro. Dig. Dis. Sci. 47, 1135–1140. Jeppesen, P.B., Hartmann, B., Thulesen, J., Graff, J., Lohmann, J., Hansen, B.S., Tofteng, F., Paulsen, S.S., Madsen, J.L., Holst, J.J., Mortensen, P.B., 2001. Glucagon-like peptide 2 improves nutrient adsorption and nutritional status in short-bowel patients with no colon. Gastroenterology 120, 806–815. Leung, D.W., Cachianes, G., Kuang, W.J., Goeddel, D.V., Ferrara, N., 1989. Vascular endothelial growth factor is a secreted angiogenic mitogen. Science 246, 1306–1309. Moses, H.L., Coffey, R.J., Leof, E.B., Lyons, R.M., Keski-Oka, J., 1987. Transforming growth factor regulation of cell proliferation. J. Cell Physiol. 5, 1–7. Orskov, C., Hartmann, B., Poulsen, S.S., Thulesen, J., Hare, K.J., Holst, J.J., 2005. GLP-2 stimulates colonic growth via KGF, released by subepithelial myofibroblasts with GLP-2 receptors. Regul. Pept. 124, 05–12. Parvadia, J.K., Kesnawi, S.G., Vaikunth, S., Maldonado, A.R., Marwan, A., Stehr, W., Uzvolgyi, E., Warner, B.W., Yamano, S., Taichman, N., Crombleholme, T.M., in press. Role of vascular endothelial growth factor (VEGF) in small bowel adaptation after resection: the adaptive response is angiogenesis dependent. Am. J. Physiol. Gastrointest. Liver Physiol. (Electronic publication ahead of print).
285
Prasad, R., Alavi, K., Schwartz, M.Z., 2001. GLP-2alpha accelerates recovery of mucosal absorptive function after intestinal ischemia/reperfusion. J. Pediatr. Surg. 36, 570–572. Ramsanahie, A.P., Perez, A., Duensing, A.U., Zinner, M.J., Ashley, S.W., Whang, E.E., 2002. Glucagon-like peptide 2 enhances intestinal epithelial restitution. J. Surg. Res. 107, 44–49. Sams, A., Hastrup, S., Andersen, M., Thim, L., 2006. Naturally occuring glucagon-like peptide-2 (GLP-2) receptors in human intestinal cell lines. Eur. J. Pharmacol. 532, 18–23. Scott, R.B., Kirk, D., MacNaughton, W.K., Meddings, J.B., 1998. GLP-2 augments the adaptive response to massive intestinal resection in rat. Am. J. Physiol. Endocrinol. Metabol. 275, G911–G921. Thulesen, J., Hartmann, B., Hare, K.J., Kissow, H., Orskov, C., Holst, J.J., Poulsen, S.S., 2004. Glucagon-like peptide 2 (GLP-2) accelerates the growth of colonic neoplasms in mice. Gut 53, 1145–1150. Xiao, Q., Boushey, R., Cino, M., Drucker, D.J., Brubaker, P.L., 2000. Circulating levels of glukagon-like peptide-2 in human subjects with inflammatory bowel disease. Am. J. Physiol. Regulatory Integrative Comp. Physiol. 278, R1057–R1063. Zhu, C., Ying, D., Zhou, D., Mi, J., Zhang, W., Chang, Q., Li, L., 2005. Expression of TGF-beta 1 in smooth muscle cells regulates endothelial progenitor cells migration and differentiation. J. Surg. Res. 12, 151–156.