Dig Dis Sci DOI 10.1007/s10620-014-3257-5
ORIGINAL ARTICLE
Inhibition of Endothelial Slit2/Robo1 Signaling by Thalidomide Restrains Angiogenesis by Blocking the PI3K/Akt Pathway Yinan Li • Sengwang Fu • Haiying Chen • Qian Feng • Yunjie Gao • Hanbing Xue • Zhizheng Ge • Jingyuan Fang • Shudong Xiao
Received: 3 November 2013 / Accepted: 16 June 2014 Ó Springer Science+Business Media New York 2014
Abstract Background Thalidomide is effective in the treatment of angiodysplasia. The mechanisms underlying its activity may be associated with inhibition of angiogenic factors. It was recently shown that Slit2/Robo1 signaling plays a role in angiogenesis. Purpose The aim of this study was to explore the expression and effects of Robo1 and Slit2 in angiodysplasia and to identify the possible therapeutic mechanisms of thalidomide. Method Slit2 and Robo1 expression were analyzed in tissue samples and human umbilical vein endothelial cells (HUVECs) treated with thalidomide using a combination of laboratory assays that were able to detect functional activity. Results Slit2, Robo1 and vascular endothelial growth factor (VEGF) were strongly expressed in five angiodysplasia lesions out of seven cases, while expression was low in one out of seven normal tissues. Exposure of HUVECs to recombinant N-Slit2 resulted in an increase in VEGF
Y. Li S. Fu H. Chen Q. Feng Y. Gao H. Xue Z. Ge (&) J. Fang S. Xiao Shanghai Institution of Digestive Disease, Renji Hospital, School of Medicine, Shanghai Jiao Tong University, 145 Middle Shandong Rd. GI Division, Shanghai 200001, China e-mail: [email protected]; [email protected] Y. Li S. Fu H. Chen Q. Feng Y. Gao H. Xue Z. Ge J. Fang S. Xiao Key Laboratory of Gastroenterology and Hepatology, Ministry of Health, Shanghai Jiao Tong University, Shanghai, China Y. Li S. Fu H. Chen Q. Feng Y. Gao H. Xue Z. Ge J. Fang S. Xiao State Key Laboratory of Oncogene and Related Genes, Shanghai Jiao Tong University, Shanghai, China
levels and stimulated proliferation, migration and tube formation. These effects were blocked by an inhibitor of PI3K and thalidomide. Conclusions Robo1 and Slit2 may have important roles in the formation of gastrointestinal vascular malformation. High concentrations of Slit2 increased the levels of VEGF in HUVECs via signaling through the PI3K/Akt pathway— an effect that could be inhibited by thalidomide. Keywords Gastrointestinal Vascular malformation Angiogenesis Thalidomide Slit2 Robo1 VEGF
Introduction Gastrointestinal vascular malformation (GIVM) is the main cause for obscure gastrointestinal bleeding (OGIB) [1]. In previous studies, gastrointestinal microvascular endothelial cells secreted many different growth factors, and vascular endothelial growth factor (VEGF) played an important role in GIVM [2]. A variety of different stimuli are involved in VEGF expression including certain members of the Slit–Robo family [3, 4]. The Robo family is comprised of four known members (Robo1–4), whereas the Slit family has three members (Slit1–3). The full-length Slit2 protein is a 200-kDa secreted ligand, which is cleaved into two smaller fragments, a 140-kDa N-terminal product (N-Slit2) and a 50–60-kDa C-terminal product (C-Slit2). N-Slit2 binds to the cell membrane, whereas C-Slit2 is diffusible [5]. Although Slit2 was initially characterized as a repulsive guidance cue for neuronal axons [6], it is also involved in the development of several other organ systems and especially in angiogenesis [7, 8]. Accumulating evidence implicates axon guidance molecules in angiogenesis [9,
123
Dig Dis Sci
10]. Slit2 has been widely identified in various human diseases associated with angiogenesis, such as cancer [11, 12], eclampsia [13] and proliferative retinopathy [14]. The interaction between Slit2 and Robo1 induces tumor angiogenesis and lymphangiogenesis [15]. These data substantiate the significance of the Slit2/Robo1 signaling system in vascular development. Thalidomide has been recognized as a potent teratogen, which induces limb defects. Studies with animal models have provided evidence for the angiostatic activity in the appearance of embryopathy. Due to its anti-angiogenic properties, thalidomide is now being used for a broad spectrum of pharmacological and immunological effects [16, 17]. Thalidomide displays anti-angiogenesis activity in GIVM [18] and hereditary hemorrhagic telangiectasia (HHT) [19]. However, the molecular mechanisms underlying this effect have not yet been fully elucidated. The vasculature in GIVM lesions is predominantly composed of vascular endothelial cells with sparse smooth muscle cell lining [20]. We therefore set out to determine the antiangiogenesis effect of thalidomide on human umbilical vein endothelial cells (HUVECs) and to investigate its potential inhibitory effect on Akt activation. In a previous study, we confirmed the presence of VEGF expression in the vascular malformation regions [21], and we indicated that VEGF might play a cellular role in the formation of vascular malformation. In this study, we found that the endothelial cells in vascular malformation tissue stained positively for Slit2, implicating an autocrine/ paracrine role for endothelial cells in regulating angiogenesis function. We also investigated the effect of thalidomide on Robo1, Slit2 and VEGF expression and on the migration, proliferation and vascular formation capacity of HUVECs. Using recombinant human N-terminal Slit2 protein (N-Slit2) and LY294002 (an inhibitor of phosphoinositide 3-kinase, PI3K), we were able to demonstrate that Slit/Robo signaling regulates the expression of VEGF in endothelial cells via the PI3K/AKT pathway.
Materials and Methods Tissue Samples The Ethics Committee of the Shanghai JiaoTong University School of Medicine approved this study protocol, and informed consent was obtained from all patients diagnosed by endoscopic examination. The studied patients were two males and five females, with ages ranging from 41 to 78 years. These patients have suffered from an average of 4 (range 2–8) episodes of bleeding in a year before biopsy. From these patients, a biopsy specimen of the angiodysplastic lesion and another normal tissue specimen outside
123
the angiodysplastic lesion were collected. The Institutional Review Board of the Renji Hospital approved the use of human tissues and the performance of all described experiments. Immunohistochemistry Tissues were snap-frozen, and 6-lm sections were cut, airdried and fixed in 4 % paraformaldehyde (PFA) for 20 min. Sections were subsequently washed with PBS and blocked with 10 % normal goat serum for 1 h at 37 °C. Appropriately diluted anti-Slit2 polyclonal antibody (Abcam, Cambridge, UK) was applied to the tissue sections at 4 °C overnight and incubated for 1 h at 37 °C. After being washed, the slides were stained with 40 ,60 -diamino-2phenylindole (DAPI). For each of the immunostaining procedures, negative controls included omission of the primary antibody and use of an irrelevant polyclonal or isotype-matched monoclonal primary antibody. Cell Culture and Reagents Primary HUVECs were obtained from Lifeline (Frederick, MD, USA) and were cultured in the VascuLife VEGF Medium Complete Kit (LL-0003, Lifeline, Frederick, MD, USA) at 37 °C under 5 % CO2 and 95 % humidity. HUVECs were plated in 6-well culture dishes and were used for experiments at 80–90 % confluence. The cells were incubated in fresh, serum-free medium for 24 h before use in experiments. Thalidomide (Sigma, St. Louis, MO) was added to the medium at the concentration of 0, 0.2, 0.4 or 0.8 mM. To assess PI3K/Akt activity, the pharmacological inhibitor LY294002 (Sigma, St. Louis, MO) was added to the culture medium at a final concentration of 10 lM. Where indicated, cells were treated with N-Slit2 protein (PeproTech, Rocky Hill, NJ) at a final concentration of 100 ng/ml. RNA Isolation Total RNA was isolated using Trizol reagent (Invitrogen, Germany) according to the manufacturer’s instructions. After being washed with 75 % ethanol, the final RNA extracts were eluted in 20 ll of distilled diethyl pyrocarbonate-treated water. The concentration and purity of RNA were measured using a spectrophotometer. All RNA preparations had an OD260/OD280 ratio of 1.9–2.0. Real-Time Polymerase Chain Reaction Robo1, Slit2 and VEGF quantitative real-time polymerase chain reaction (PCR) detections12 (Table 1) were normalized to b-actin expression, and expression was calculated using the following equation: fold change = 2-DDct.
Dig Dis Sci
Western Blot Analysis Cells were washed three times in ice-cold phosphate-buffered saline (PBS, 4 °C, pH 7.4), and proteins were prepared at room temperature using protein extraction and protease inhibitor kits. Cell lysates were cleared by centrifugation at 12,0009g at 4 °C. The supernatant was collected, and the protein content of each lysate was measured using a bicinchoninic acid (BCA) Protein Assay Kit, according to the manufacturer’s instructions. The standard and experimental samples were separated by 10 % SDSPAGE and transferred to polyvinylidene difluoride membranes. Primary antibodies that were used included antiRobo1 (1:1,000; Abcam, Cambridge, UK), anti-Slit2 (1:1,000; Abcam, Cambridge, UK), anti-VEGF (1:1,000; Abcam, Cambridge, UK), anti-pAkt (Ser473; 1:1,000; Abcam, Cambridge, UK), anti-Akt (1:1,000; Abcam, Cambridge, UK) and anti-GAPDH 1:5,000; (Abcam, Cambridge, UK). Membranes were washed and incubated with peroxidase-conjugated secondary antibodies (1:5,000, Dianova, Germany), and proteins were visualized using enhanced chemiluminescence Western blotting detection reagents according to the manufacturer’s recommendations. The intensity of each band was normalized to each GAPDH internal control. All immunoblots were repeated at least three times. Enzyme-Linked Immunosorbent Assay Cell supernatants were analyzed for VEGF levels using commercially available enzyme-linked immunosorbent assay (ELISA) kits (R&D Biosystems, USA). Conditioned media were collected after 24 h of incubation with either thalidomide, LY294002 or N-Slit2, centrifuged and stored at -80 °C until analysis. Measurements were conducted according to the manufacturer’s instructions, and all samples were assayed in triplicate. Flow Cytometry Cells were detached using ethylenediaminetetraacetic acid (EDTA) and washed in ice-cold PBS (4 °C), and the cell cycle distribution (saponin/PI staining) was analyzed by fluorescence-activated cell sorting (FACS). Three samples were used per experiment, and each experiment was repeated.
500–600 lm. After scratching, the wells were washed with PBS twice and then incubated with fresh medium containing the different drugs. Migration was evaluated by measuring the wound width at 0, 12 and 24 h after scratching. Images were taken with a microscope video system, and at three places in each wound, the width was measured and averaged. Tube Formation Assay The tube formation assay was performed using a 48-well plate with 100 ll of Matrigel basement membrane matrix (BD Bioscience, Bedford, MA) per well and solidified at 37 °C for 30 min. HUVECs (2 9 104) were cultured on these gels under different conditions. After 24 h of incubation, Matrigel-induced morphological changes in HUVECs and their tubular networks were photographed at 1009 magnification for digital assessment using Image-Pro Plus 6.0. Proliferation Assay HUVECs (5 9 103 cells) were seeded and grown in each well of a 96-well microtiter plate and incubated overnight in culture medium (200 ll). Cells were starved without FCS overnight at 80–90 % confluence and then treated with recombinant N-Slit2, thalidomide or LY294002 at different concentrations. Cells without any treatment were used as controls. After 24 h of culture, optical density (OD) was measured using the CCK8 assay kit (Dojindo, Kumamoto, Japan) according to the manufacturer’s instructions and using microplate computer software (BioRad Laboratories). Statistical Evaluation All data were evaluated for normal distribution of the data. Statistical differences (p \ 0.05) were evaluated using oneway analysis of variance (ANOVA) followed by the leastsignificant difference (LSD) test for multiple comparisons. All data are presented as mean ± standard deviation (SD).
Results Slit2, Robo1, and VEGF Are Highly Expressed in Vascular Malformation Tissues
Scratch Wound Assay HUVECs were seeded in a 6-well tissue culture plate at a density of 2 9 105/well. After the cells had reached 90 % confluency, a scratch wound was made in the middle of the plate by a 100-ll pipette tip, generating a wound width of
We assembled a cohort of vascular malformation and normal gastrointestinal mucosal tissues from seven patients and immunohistochemically assayed for expression of Slit2. Detection with the Slit2 antibodies demonstrated that Slit2 was overexpressed in the dilated and twisted capillary
123
Dig Dis Sci Fig. 1 Slit2, Robo1 and VEGF are highly expressed in vascular malformation tissues. a Vascular malformation (VM) lesions (A1) and normal tissues (A2) were stained with anti-Slit2 antibody; A3 negative control staining of angiodysplasia tissue (scale bar 100 lm). b Immunoblot analysis for Slit2, Robo1 and VEGF expression in vascular malformation lesions and normal tissues.**p \ 0.01
A
B VM
Normal
Robo1 Slit2 VEGF
GAPDH
endothelial cells in the mucosal and submucosal layers of angiodysplasia lesions (Fig. 1A1), in 6 out of 7 of the patients’ tissues. On the contrary, Slit2 was only expressed slightly in the normal mucous membranes and vascular endothelial cells of the same person (Fig. 1A2). For assessment of Slit2, Robo1 and VEGF expression, we also analyzed the protein level in seven vascular malformation tissues and matched normal tissue lysates, by Western blot. Slit2, Robo1 and VEGF were overexpressed in comparison with normal levels by factors of 3.84 ± 0.83, 2.14 ± 0.45 and 2.90 ± 0.82, respectively (p \ 0.01; Fig. 1b). Thalidomide Inhibits Robo1, Slit2, and VEGF mRNA and Protein Expression in HUVECs We initially examined expression levels of Robo1, Slit2 and VEGF in HUVECs after treatment with thalidomide. Exposure of HUVECs to thalidomide (0, 0.2, 0.4 or 0.8 mM) decreased Robo1, Slit2 and VEGF mRNA levels, as measured by real-time qPCR (Fig. 2a). Thalidomide, at a concentration of 0.8 mM, induced the strongest inhibition of VEGF, Slit2 and Robo1 mRNA expression levels (mean inhibitions and SDs: 0.36 ± 0.04, 0.39 ± 0.08 and 0.32 ± 0.04, respectively), by approximately 60–70 % as compared to the control group (p \ 0.01). It was
123
subsequently tested whether Slit2, Robo1 and VEGF protein levels are also regulated by thalidomide. For Robo1, this was evaluated by Western blot analysis using total cell lysates from HUVECs cultured with various concentrations of thalidomide for 24 h. Robo1 protein expression, in response to 0.8 mM thalidomide, was lower than in the control group (0.53 ± 0.05, Fig. 2b, p \ 0.01). The secretion of Slit2 and VEGF protein by HUVECs was measured by ELISA and was also suppressed by thalidomide. Cells treated with thalidomide, at a concentration of 0.8 mM, induced the highest level of VEGF suppression (Fig. 2c, d, p \ 0.01).
Effects of Thalidomide and Recombinant N-Slit2 Protein on the Phosphorylation of Akt in HUVECs Next, we examined the effects of thalidomide and N-Slit2 on the PI3K/Akt signaling cascade in endothelial cells. Exposure of HUVECs to N-Slit2 (100 ng/ml) induced Akt phosphorylation as demonstrated by Western blot analysis. Treatment of HUVECs with thalidomide significantly inhibited Akt phosphorylation, which was similar to the effect of LY294002, a specific PI3K/Akt inhibitor. Thalidomide was also able to fully counteract the inducing activity of the N-Slit2 protein (Fig. 6).
Dig Dis Sci
A
Table 1 Real-time qPCR primer design Gene b-actin
Oligonucleotide primers (50 –30 )
Size (bp)
F: CTTAGTTGCGTTACACCCTT
144
R: CCTTCACCGTTCCAGTTT Slit2
F: CACCTCGTACAGCCGCACTT
104
R: TGTGGACCGCTGAGGAGCAA Robo1
F: GGAAGAAGACGAAGCCGACAT
107
R: TCTCCAGGTCCCCAACACTG
B
Tha
0
0.2
0.4
0.8
(mM)
Robo1
VEGF
F: AGGGAAGAGGAGGAGATG
148
R: GCTGGGTTTGTCGGTGTT
GAPDH
C
D
Fig. 2 Thalidomide inhibits Robo1, Slit2, and VEGF mRNA and protein expression in HUVECs. a HUVECs were treated with an increased dose of thalidomide (Tha) for 24 h. The mRNA levels of Slit2, Robo1 and VEGF were determined by real-time qPCR and normalized with GAPDH using absolute quantification. b HUVECs were treated with increased doses of thalidomide, and Robo1 protein levels were measured by immunoblot analysis. c, d HUVECs were treated with increased doses of thalidomide, and VEGF and Slit2 protein levels were measured using ELISA. The data represent mean values ± standard deviations (error bars) from three independent experiments. **p \ 0.01
Thalidomide Can Reverse the Promoting Effect of Recombinant N-Slit2 Protein on Migration, Proliferation, and Tube Formation of HUVECs To further elucidate the biological effects of Slit2 and thalidomide on migration, proliferation and tube formation of HUVECs, we performed scratch wound assays, proliferation assays and tube formation assays. Wound closure was clearly inhibited by exposure of cells to thalidomide or LY294002, as compared to the control group (Fig. 3a). In contrast, recombinant N-Slit2 significantly promoted the
Fig. 3 Effects of thalidomide and recombinant N-Slit2 protein on the phosphorylation of Akt in HUVEC. a Photograph of the immunoblot analysis for pAkt and Akt expression in HUVECs. b Expression levels of pAkt protein in HUVECs, *p \ 0.05, **p \ 0.01. The data represent mean values ± SDs (error bars) from three independent experiments. NC negative control, S Slit2 (100 ng/ml), T Thalidomide (0.8 mM), L LY294002 (10 lM)
migration capacity of HUVECs. The increasing effect of N-Slit2 on HUVECs could be reversed by co-incubation with both thalidomide and LY294002 (Fig. 3a). Besides, treatment of N-Slit2 promoted the proliferation of HUVEC which can also be reversed by thalidomide or LY294002 (Fig. 3b). Similarly, increased formation of tubular networks in the Matrigel was observed when HUVECs were incubated with N-Slit2. A gradual abrogation of the network formation occurred in the presence of thalidomide (Fig. 3c). Taken together, N-Slit2 promoted the migration, proliferation and tube formation activities of HUVECs, while thalidomide was able to suppress these activities.
123
Dig Dis Sci
of VEGF was increased by exposure to recombinant N-Slit2 protein, while HUVECs cultured with thalidomide or LY294002 decreased in VEGF expression level (Fig. 5), as compared to controls.
Discussion
Fig. 4 Effects of thalidomide and recombinant N-Slit2 protein on the expression of VEGF in HUVECs. HUVECs were treated with 100 ng/ ml recombinant N-Slit2 protein (S), 0.8 mM thalidomide (T), 10 lM LY294002 (L) or a combination. VEGF protein levels were assessed using ELISA. The data represent mean values ± SDs (error bars) from three independent experiments.*p \ 0.05, **p \ 0.01
Effects of Thalidomide and Recombinant N-Slit2 Protein on Cell Cycle After incubation with recombinant N-Slit2, thalidomide and LY294002, HUVECs were analyzed using flow cytometry to detect alterations in the cell cycle distribution. The recombinant N-Slit2 protein resulted in a significant reduction of HUVECs in the G0-/G1-phase of cell cycle and promoted an accumulation of cells in the S-phase, as compared to controls (Fig. 4). Of the cells treated with recombinant N-Slit2 protein, 53.14 % were in the G1-phase and 32.50 % were in S-phase of cell cycle, compared to 60.46 % of the control group in G1-phase and 27.38 % in S-phase (p \ 0.05). Thalidomide showed a contrasting effect (64.90 vs. 60.46 % in G1-phase and 19.81 vs. 27.38 % in S-phase, p \ 0.05). Thalidomide inhibited the N-Slit2-induced accumulation of cells in S-phase (32.50 vs. 22.26 %, p \ 0.01). In addition, PI3K inhibitor LY294002 shows a similar effect as thalidomide. These results illustrated that the contrary effects between thalidomide and N-Slit2 on angiogenesis were cell cycle-dependent and associated with PI3K-Akt pathway. Interestingly, no apoptotic peak appears in the cell cycle histogram (data not shown). To determine whether N-SLIT2 or thalidomide affects cell apoptosis, we examined the expression of Bcl-2 protein. The results illustrated that N-SLIT2 treatment could moderately up-regulate the Bcl-2 expression in HUVECs, and the presence of thalidomide and LY294002 could reverse this effect (Fig. 4b). Thalidomide Can Reverse the Promoting Effect of Recombinant N-Slit2 Protein on VEGF Expression of HUVECs We evaluated whether VEGF levels might be affected in HUVECs treated with N-SLIT2 or thalidomide. Expression
123
The Slit–Robo signaling pathway was originally characterized as a regulator of neuron axonal growth; the pathway also regulates angiogenesis3,4,7. Because there is no animal model for developing regional malformations, studies on this disease have been limited. Therefore, studies in patients are of great value. Given the low incidence and the low number of patients that undergo endoscopic treatment, the data presented here are based on a small sample size. However, we demonstrated for the first time that Slit2 is overexpressed in the vascular malformation regions of human gastrointestinal mucosa in patients with GIVM. Slit2 was strongly expressed near the enlarged, twisted vessels of the vascular malformation region, while it was detected only slightly in the normal tissues of the same patients (Fig. 1). In a previous study, we found that VEGF is also expressed more in angiodysplasia lesions20. Our results suggest that Slit2 signaling may actively contribute to the progression of angiogenesis and in this way participate in the processes underlying angiodysplasia in GIVM patients. In this study, we demonstrated by real-time RT-PCR, immunoblotting and ELISA that thalidomide can inhibit the expression of Robo1, Slit2 and VEGF (Fig. 2). These results suggest that Robo1 and Slit2 may be involved in the anti-angiogenesis action of thalidomide. Many studies have established that growth factorinduced endothelial cell migration and subsequent tube formation are known to be PI3K-Akt dependent [22–24]. We studied the PI3K/Akt signaling pathway using LY294002, a well-known inhibitor of PI3K. We found that N-Slit2 can promote the phosphorylation of Akt (Fig. 3) and the expression of VEGF (Fig. 4). Meanwhile, LY294002 was found to be able to block the effects of N-Slit2 on VEGF expression and phosphorylation of Akt. Results indicated that Slit2 could promote the proliferation, migration and tube formation of HUVECs, while thalidomide could reverse these effects (Fig. 4). VEGF is generally considered to be a key angiogenic factor in the neovascularization that occurs during GIVM2. VEGF signaling is tightly regulated in endothelial cells. This regulation includes cross talk with other signaling pathways, such as Notch [25–27] and Ang-2 [28, 29]. Our data show further evidence for a functional interplay between the Slit2/Robo1 pathway and VEGF signaling in GIVM. Based on the results presented, it is clear that recombinant N-Slit2
Dig Dis Sci Fig. 5 Thalidomide can reverse the promoting effect of recombinant N-Slit2 protein on migration, proliferation and tube formation of HUVECs. a Influence of N-Slit2 and thalidomide on migration ability of HUVECs. b CCK-8 proliferation assay results after thalidomide and recombinant N-Slit2 protein exposure of HUVECs for 24 h. c Effects of thalidomide and recombinant N-Slit2 protein on tube formation of HUVECs. *p \ 0.05, **p \ 0.01. The data represent mean values ± SDs (error bars) from three independent experiments. NC negative control, S Slit2 (100 ng/ml), T thalidomide (0.8 mM), L: LY294002 (10 lM) S ? L: Slit2 (100 ng/ ml) and LY294002 (10 lM)
protein promoted VEGF secretion in HUVECs (Fig. 4). Additionally, we found that thalidomide, by its ability to suppress the expression of Slit2 and Robo1, could inhibit this process while presenting a possible treatment option for GIVM (Fig. 4).
Endothelial cell migration and subsequent tube formation play an important role during angiogenesis [30]. Our data demonstrate that activation of Akt in HUVECs by Slit2 was involved in proliferation, migration and tube formation of HUVECs (Fig. 5).
123
Dig Dis Sci
Fig. 6 Effects of thalidomide and recombinant N-Slit2 protein on cell cycle. HUVECs treated with 100 ng/ml recombinant N-Slit2 protein (S), 0.8 mM thalidomide (T), 10 lM LY294002 (L) or a combination. a Cell cycle distribution was assessed using flow cytometry. b Bcl-2 protein levels were measured using Western blot. The data represent mean values ± SDs (error bars) from three independent experiments. *p \ 0.05, **p \ 0.01
Usually, endothelial cells are quiescent and reside in the G0-phase. We found that N-Slit2 protein significantly increases HUVEC proliferation. In this study, additionally, we observed that recombinant N-Slit2 protein may play a role in many of the processes involved in the early events of angiogenesis, such as migration and tube formation of endothelial cells (Fig. 6). At the same time, we also found that recombinant N-Slit2 protein caused significant reductions in the subset of cells in G0-/G1-phase and an accumulation of cells in S-phase. In addition, N-Slit2 upregulated the expression level of the anti-apoptosis protein Bcl-2 (Fig. 6). Taken together, these results suggest that Slit2 is important in initiating the pathogenesis of GIVM and that all these effect can be inhibited by thalidomide. Our previous study showed that thalidomide suppresses hypoxia- or bFGF-induced VEGF expression at the protein level18. In the present study, we found that the expression levels of Slit2/Robo1 and VEGF at both mRNA and protein levels were suppressed by thalidomide. We also demonstrated that the inhibitory effect of thalidomide on angiogenesis is attributable to inhibition of signaling via Akt phosphorylation, leading to reduced angiogenic response of HUVECs. Since the N-Slit2 can promote Akt phosphorylation (Fig. 3), as mentioned above, the anti-angiogenesis effect of thalidomide may be associated with the Slit2/ Robo1 pathway. Thalidomide itself also had an effect on the proliferation, migration and tube formation on HUVECs. By taking our study mentioned above (Fig. 4) in consideration, we postulate that this can be the result of endogenous Slit2
123
suppression. However, from our present data, we cannot exclude a role of PTEN stabilization and consequent inhibition of PI3K/Akt signaling in this process [31]. Studies have been dedicated to determining the mechanisms of thalidomide teratogenicity. Thalidomide induces dysfunction of embryo limb vessels at critical stages in limb development, prior to the onset of malformations, suggesting that vascular malfunction is an upstream event in hampering limb development [32]. However, several thalidomide analogues such lenalidomide [33] and pomalidomide [34] also show a significant anti-angiogenesis effect and less teratogenicity. These analogues also indirectly inhibit endothelial cell proliferation by inhibiting the secretion of VEGF and inducing G0/G1 arrest [35]. In summary, our study suggests that Slit2–-Robo1 signaling plays a role in GIVM. Increasing concentrations of Slit2 increased VEGF expression in HUVECs. Thalidomide was able to block the expression of Robo1 and Slit2. The recombinant N-Slit2 protein also resulted in the increased proliferation, migration and capillary tube formation of HUVECs. In addition, it increased the number of HUVECs in S-phase. Treatment of HUVECs with recombinant N-Slit2 protein, increased VEGF mRNA expression and VEGF secretion, as well as phosphorylation of Akt, which can also be inhibited by thalidomide as well as LY294002. Although preliminary, all of these results suggest that Slit2–Robo1 signaling may be involved in GIVM. Nevertheless, further research is required to elucidate the precise role of Slit2-Robo1 signaling in the development of GIVM and to examine its potential role as a therapeutic target of thalidomide. Acknowledgments The research presented here was funded by the Shanghai Municipal Health Bureau Academic Discipline Project, Project Number: 20114002. Conflict of interest of interest.
The authors declare that they have no conflict
References 1. Regula J, Wronska E, Pachlewski J. Vascular lesions of the gastrointestinal tract. Best Pract Res Clin Gastroenterol. 2008;22:313–328. 2. Junquera F, Saperas E, de Torres I, Vidal MT, Malagelada JR. Increased expression of angiogenic factors in human colonic angiodysplasia. Am J Gastroenterol. 1999;94:1070–1076. 3. Paul JD, Coulombe KL, Toth PT, et al. SLIT3-ROBO4 activation promotes vascular network formation in human engineered tissue and angiogenesis in vivo. J Mol Cell Cardiol. 2013;64:124–131. 4. Guo SW, Zheng Y, Lu Y, et al. Slit2 overexpression results in increased microvessel density and lesion size in mice with induced endometriosis. Reprod Sci. 2013;20:285–298. 5. Brose K, Bland KS, Wang KH, et al. Slit proteins bind Robo receptors and have an evolutionarily conserved role in repulsive axon guidance. Cell. 1999;96:795–806.
Dig Dis Sci 6. Wong K, Park HT, Wu JY, Rao Y. Slit proteins: molecular guidance cues for cells ranging from neurons to leukocytes. Curr Opin Genet Dev. 2002;12:583–591. 7. Kramer SG, Kidd T, Simpson JH, Goodman CS. Switching repulsion to attraction: changing responses to slit during transition in mesoderm migration. Science. 2001;292:737–740. 8. Adams RH, Eichmann A. Axon guidance molecules in vascular patterning. Cold Spring Harb Perspect Biol. 2010;2:a001875. 9. Sheldon H, Andre M, Legg JA, et al. Active involvement of Robo1 and Robo4 in filopodia formation and endothelial cell motility mediated via WASP and other actin nucleation-promoting factors. FASEB J. 2009;23:513–522. 10. Urbich C, Ro¨ssig L, Kaluza D, et al. HDAC5 is a repressor of angiogenesis and determines the angiogenic gene expression pattern of endothelial cells. Blood. 2009;113:5669–5679. 11. Wang B, Xiao Y, Ding BB, et al. Induction of tumor angiogenesis by Slit-Robo signaling and inhibition of cancer growth by blocking Robo activity. Cancer Cell. 2003;4:19–29. 12. Zhou W, Yu W, Xie W, et al. The role of SLIT-ROBO signaling in proliferative diabeticretinopathy and retinal pigment epithelial cells. Molecular Vision. 2011;17:1526–1536. 13. Liao WX, Laurent LC, Agent S, Hodges J, Chen DB. Human placental expression of Slit/ROBO signaling cues: effects of preeclampsia and hypoxia. Biol Reprod. 2012;86:111. 14. Zhou W, Yu W, Xie W, Huang L, Xu Y, Li X. The role of SlitROBO signaling in proliferative diabetic retinopathy and retinal pigment epithelial cells. Mol Vis. 2011;17:1526–1536. 15. Yang XM, Han HX, Sui F, Dai YM, Chen M, Geng JG. Slit-Robo signaling mediates lymphangiogenesis and promotes tumor lymphatic metastasis. Biochem Biophys Res Commun. 2010;396: 571–577. 16. Bartlett JB, Dredge K, Dalgleish AG. The evolution of thalidomide and its IMiD derivatives as anticancer agents. Nat Rev Cancer. 2004;4:314–322. 17. Danese S, Sans M, Spencer DM, et al. Angiogenesis blockade as a new therapeutic approach to experimental colitis. Gut. 2007;56:855–862. 18. Ge ZZ, Chen HM, Gao YJ, et al. Long-term effects of thalidomide for recurrent gastrointestinal bleeding due to vascular malformation: an open-label, randomized, parallel controlled study. Gastroenterology. 2011;141:1629–1637. 19. Lebrin F, Srun S, Raymond K, et al. Thalidomide stimulates vessel maturation and reduces epistaxis in individuals with hereditary hemorrhagic telangiectasia. Nat Med. 2010;16:420–428. 20. Regula J, Wronska E, Pachlewski J. Vascular lesions of the gastrointestinal tract. Best Pract Res Clin Gastroenterol. 2008;22: 313–328. 21. Tan H, Chen H, Xu C, et al. Role of vascular endothelial growth factor in angiodysplasia: an interventional study with thalidomide. J Gastroenterol Hepatol. 2012;27:1094–1101.
22. Chavakis E, Dernbach E, Hermann C, Mondorf UF, Zeiher AM, Dimmeler S. Oxidized LDL inhibits vascular endothelial growth factor-induced endothelial cell migration by an inhibitory effect on the Akt/endothelial nitric oxide synthase pathway. Circulation. 2001;103:2102–2103. 23. Fujio Y, Walsh K. Akt mediates cytoprotection of endothelial cells by vascular endothelial growth factor in an anchoragedependent manner. J Biol Chem. 1999;274:16349–16354. 24. Nakamura JL, Karlsson A, Arvold ND, et al. PKB/Akt mediates radiosensitization by the signaling inhibitor LY294002 in human malignant gliomas. J Neurooncol. 2005;71:215–222. 25. Benedito R, Roca C, So¨rensen I, et al. The notch ligands Dll4 and Jagged1 have opposing effects on angiogenesis. Cell.. 2009;137:1124–1135. 26. Larrive´e B, Prahst C, Gordon E, et al. ALK1 signaling inhibits angiogenesis by cooperating with the Notch pathway. Dev Cell. 2012;22:489–500. 27. Hernandez SL, Banerjee D, Garcia A, et al. Notch and VEGF pathways play distinct but complementary roles in tumor angiogenesis. Vasc Cell. 2013;5:17. 28. Zhu Y, Lee C, Shen F, Du R, Young WL, Yang GY. Angiopoietin-2 facilitates vascular endothelial growth factor-induced angiogenesis in mature mouse brain. Stroke. 2005;36:1533–1537. 29. Loukovaara S, Robciuc A, Holopainen JM. Ang-2 upregulation correlates with increased levels of MMP-9, VEGF, EPO and TGFb1 in diabetic eyes undergoing vitrectomy. Acta Ophthalmol. 2013;91:513–519. 30. Saunders WB, Bohnsack BL, Faske JB, et al. Coregulation of vascular tube stabilization by endothelial cell TIMP-2 and pericyte TIMP-3. J Cell Biol. 2006;175:179–191. 31. Knobloch J, Schmitz I, Go¨tz K, Schulze-Osthoff K, Ru¨ther U. Thalidomide induces limb anomalies by PTEN stabilization, Akt suppression, and stimulation of caspase-dependent cell death. Mol Cell Biol. 2008;28:529–538. 32. Therapontos C, Erskine L, Gardner ER, Figg WD, Vargesson N. Thalidomide induces limb defects by preventing angiogenic outgrowth during early limb formation. Proc Natl Acad Sci USA. 2009;106:8573–8578. 33. Dredge K, Horsfall R, Robinson SP, et al. Orally administered lenalidomide (CC-5013) is anti-angiogenic in vivo and inhibits endothelial cell migration and Akt phosphorylation in vitro. Microvas Res. 2005;69:59–63. 34. Terpos E, Kanellias N, Christoulas D, Kastritis D, Dimopoulos MA. Pomalidomide: a novel drug to treat relapsed and refractory multiple myeloma. Onco Targets Ther. 2013;6:531–538. 35. Dredge K, Marriott JB, Macdonald CD, et al. Novel thalidomide analogues display anti-angiogenic activity independently of immunomodulatory effects. Br J Cancer. 2002;87:1166–1172.
123
ORIGINAL ARTICLE
Inhibition of Endothelial Slit2/Robo1 Signaling by Thalidomide Restrains Angiogenesis by Blocking the PI3K/Akt Pathway Yinan Li • Sengwang Fu • Haiying Chen • Qian Feng • Yunjie Gao • Hanbing Xue • Zhizheng Ge • Jingyuan Fang • Shudong Xiao
Received: 3 November 2013 / Accepted: 16 June 2014 Ó Springer Science+Business Media New York 2014
Abstract Background Thalidomide is effective in the treatment of angiodysplasia. The mechanisms underlying its activity may be associated with inhibition of angiogenic factors. It was recently shown that Slit2/Robo1 signaling plays a role in angiogenesis. Purpose The aim of this study was to explore the expression and effects of Robo1 and Slit2 in angiodysplasia and to identify the possible therapeutic mechanisms of thalidomide. Method Slit2 and Robo1 expression were analyzed in tissue samples and human umbilical vein endothelial cells (HUVECs) treated with thalidomide using a combination of laboratory assays that were able to detect functional activity. Results Slit2, Robo1 and vascular endothelial growth factor (VEGF) were strongly expressed in five angiodysplasia lesions out of seven cases, while expression was low in one out of seven normal tissues. Exposure of HUVECs to recombinant N-Slit2 resulted in an increase in VEGF
Y. Li S. Fu H. Chen Q. Feng Y. Gao H. Xue Z. Ge (&) J. Fang S. Xiao Shanghai Institution of Digestive Disease, Renji Hospital, School of Medicine, Shanghai Jiao Tong University, 145 Middle Shandong Rd. GI Division, Shanghai 200001, China e-mail: [email protected]; [email protected] Y. Li S. Fu H. Chen Q. Feng Y. Gao H. Xue Z. Ge J. Fang S. Xiao Key Laboratory of Gastroenterology and Hepatology, Ministry of Health, Shanghai Jiao Tong University, Shanghai, China Y. Li S. Fu H. Chen Q. Feng Y. Gao H. Xue Z. Ge J. Fang S. Xiao State Key Laboratory of Oncogene and Related Genes, Shanghai Jiao Tong University, Shanghai, China
levels and stimulated proliferation, migration and tube formation. These effects were blocked by an inhibitor of PI3K and thalidomide. Conclusions Robo1 and Slit2 may have important roles in the formation of gastrointestinal vascular malformation. High concentrations of Slit2 increased the levels of VEGF in HUVECs via signaling through the PI3K/Akt pathway— an effect that could be inhibited by thalidomide. Keywords Gastrointestinal Vascular malformation Angiogenesis Thalidomide Slit2 Robo1 VEGF
Introduction Gastrointestinal vascular malformation (GIVM) is the main cause for obscure gastrointestinal bleeding (OGIB) [1]. In previous studies, gastrointestinal microvascular endothelial cells secreted many different growth factors, and vascular endothelial growth factor (VEGF) played an important role in GIVM [2]. A variety of different stimuli are involved in VEGF expression including certain members of the Slit–Robo family [3, 4]. The Robo family is comprised of four known members (Robo1–4), whereas the Slit family has three members (Slit1–3). The full-length Slit2 protein is a 200-kDa secreted ligand, which is cleaved into two smaller fragments, a 140-kDa N-terminal product (N-Slit2) and a 50–60-kDa C-terminal product (C-Slit2). N-Slit2 binds to the cell membrane, whereas C-Slit2 is diffusible [5]. Although Slit2 was initially characterized as a repulsive guidance cue for neuronal axons [6], it is also involved in the development of several other organ systems and especially in angiogenesis [7, 8]. Accumulating evidence implicates axon guidance molecules in angiogenesis [9,
123
Dig Dis Sci
10]. Slit2 has been widely identified in various human diseases associated with angiogenesis, such as cancer [11, 12], eclampsia [13] and proliferative retinopathy [14]. The interaction between Slit2 and Robo1 induces tumor angiogenesis and lymphangiogenesis [15]. These data substantiate the significance of the Slit2/Robo1 signaling system in vascular development. Thalidomide has been recognized as a potent teratogen, which induces limb defects. Studies with animal models have provided evidence for the angiostatic activity in the appearance of embryopathy. Due to its anti-angiogenic properties, thalidomide is now being used for a broad spectrum of pharmacological and immunological effects [16, 17]. Thalidomide displays anti-angiogenesis activity in GIVM [18] and hereditary hemorrhagic telangiectasia (HHT) [19]. However, the molecular mechanisms underlying this effect have not yet been fully elucidated. The vasculature in GIVM lesions is predominantly composed of vascular endothelial cells with sparse smooth muscle cell lining [20]. We therefore set out to determine the antiangiogenesis effect of thalidomide on human umbilical vein endothelial cells (HUVECs) and to investigate its potential inhibitory effect on Akt activation. In a previous study, we confirmed the presence of VEGF expression in the vascular malformation regions [21], and we indicated that VEGF might play a cellular role in the formation of vascular malformation. In this study, we found that the endothelial cells in vascular malformation tissue stained positively for Slit2, implicating an autocrine/ paracrine role for endothelial cells in regulating angiogenesis function. We also investigated the effect of thalidomide on Robo1, Slit2 and VEGF expression and on the migration, proliferation and vascular formation capacity of HUVECs. Using recombinant human N-terminal Slit2 protein (N-Slit2) and LY294002 (an inhibitor of phosphoinositide 3-kinase, PI3K), we were able to demonstrate that Slit/Robo signaling regulates the expression of VEGF in endothelial cells via the PI3K/AKT pathway.
Materials and Methods Tissue Samples The Ethics Committee of the Shanghai JiaoTong University School of Medicine approved this study protocol, and informed consent was obtained from all patients diagnosed by endoscopic examination. The studied patients were two males and five females, with ages ranging from 41 to 78 years. These patients have suffered from an average of 4 (range 2–8) episodes of bleeding in a year before biopsy. From these patients, a biopsy specimen of the angiodysplastic lesion and another normal tissue specimen outside
123
the angiodysplastic lesion were collected. The Institutional Review Board of the Renji Hospital approved the use of human tissues and the performance of all described experiments. Immunohistochemistry Tissues were snap-frozen, and 6-lm sections were cut, airdried and fixed in 4 % paraformaldehyde (PFA) for 20 min. Sections were subsequently washed with PBS and blocked with 10 % normal goat serum for 1 h at 37 °C. Appropriately diluted anti-Slit2 polyclonal antibody (Abcam, Cambridge, UK) was applied to the tissue sections at 4 °C overnight and incubated for 1 h at 37 °C. After being washed, the slides were stained with 40 ,60 -diamino-2phenylindole (DAPI). For each of the immunostaining procedures, negative controls included omission of the primary antibody and use of an irrelevant polyclonal or isotype-matched monoclonal primary antibody. Cell Culture and Reagents Primary HUVECs were obtained from Lifeline (Frederick, MD, USA) and were cultured in the VascuLife VEGF Medium Complete Kit (LL-0003, Lifeline, Frederick, MD, USA) at 37 °C under 5 % CO2 and 95 % humidity. HUVECs were plated in 6-well culture dishes and were used for experiments at 80–90 % confluence. The cells were incubated in fresh, serum-free medium for 24 h before use in experiments. Thalidomide (Sigma, St. Louis, MO) was added to the medium at the concentration of 0, 0.2, 0.4 or 0.8 mM. To assess PI3K/Akt activity, the pharmacological inhibitor LY294002 (Sigma, St. Louis, MO) was added to the culture medium at a final concentration of 10 lM. Where indicated, cells were treated with N-Slit2 protein (PeproTech, Rocky Hill, NJ) at a final concentration of 100 ng/ml. RNA Isolation Total RNA was isolated using Trizol reagent (Invitrogen, Germany) according to the manufacturer’s instructions. After being washed with 75 % ethanol, the final RNA extracts were eluted in 20 ll of distilled diethyl pyrocarbonate-treated water. The concentration and purity of RNA were measured using a spectrophotometer. All RNA preparations had an OD260/OD280 ratio of 1.9–2.0. Real-Time Polymerase Chain Reaction Robo1, Slit2 and VEGF quantitative real-time polymerase chain reaction (PCR) detections12 (Table 1) were normalized to b-actin expression, and expression was calculated using the following equation: fold change = 2-DDct.
Dig Dis Sci
Western Blot Analysis Cells were washed three times in ice-cold phosphate-buffered saline (PBS, 4 °C, pH 7.4), and proteins were prepared at room temperature using protein extraction and protease inhibitor kits. Cell lysates were cleared by centrifugation at 12,0009g at 4 °C. The supernatant was collected, and the protein content of each lysate was measured using a bicinchoninic acid (BCA) Protein Assay Kit, according to the manufacturer’s instructions. The standard and experimental samples were separated by 10 % SDSPAGE and transferred to polyvinylidene difluoride membranes. Primary antibodies that were used included antiRobo1 (1:1,000; Abcam, Cambridge, UK), anti-Slit2 (1:1,000; Abcam, Cambridge, UK), anti-VEGF (1:1,000; Abcam, Cambridge, UK), anti-pAkt (Ser473; 1:1,000; Abcam, Cambridge, UK), anti-Akt (1:1,000; Abcam, Cambridge, UK) and anti-GAPDH 1:5,000; (Abcam, Cambridge, UK). Membranes were washed and incubated with peroxidase-conjugated secondary antibodies (1:5,000, Dianova, Germany), and proteins were visualized using enhanced chemiluminescence Western blotting detection reagents according to the manufacturer’s recommendations. The intensity of each band was normalized to each GAPDH internal control. All immunoblots were repeated at least three times. Enzyme-Linked Immunosorbent Assay Cell supernatants were analyzed for VEGF levels using commercially available enzyme-linked immunosorbent assay (ELISA) kits (R&D Biosystems, USA). Conditioned media were collected after 24 h of incubation with either thalidomide, LY294002 or N-Slit2, centrifuged and stored at -80 °C until analysis. Measurements were conducted according to the manufacturer’s instructions, and all samples were assayed in triplicate. Flow Cytometry Cells were detached using ethylenediaminetetraacetic acid (EDTA) and washed in ice-cold PBS (4 °C), and the cell cycle distribution (saponin/PI staining) was analyzed by fluorescence-activated cell sorting (FACS). Three samples were used per experiment, and each experiment was repeated.
500–600 lm. After scratching, the wells were washed with PBS twice and then incubated with fresh medium containing the different drugs. Migration was evaluated by measuring the wound width at 0, 12 and 24 h after scratching. Images were taken with a microscope video system, and at three places in each wound, the width was measured and averaged. Tube Formation Assay The tube formation assay was performed using a 48-well plate with 100 ll of Matrigel basement membrane matrix (BD Bioscience, Bedford, MA) per well and solidified at 37 °C for 30 min. HUVECs (2 9 104) were cultured on these gels under different conditions. After 24 h of incubation, Matrigel-induced morphological changes in HUVECs and their tubular networks were photographed at 1009 magnification for digital assessment using Image-Pro Plus 6.0. Proliferation Assay HUVECs (5 9 103 cells) were seeded and grown in each well of a 96-well microtiter plate and incubated overnight in culture medium (200 ll). Cells were starved without FCS overnight at 80–90 % confluence and then treated with recombinant N-Slit2, thalidomide or LY294002 at different concentrations. Cells without any treatment were used as controls. After 24 h of culture, optical density (OD) was measured using the CCK8 assay kit (Dojindo, Kumamoto, Japan) according to the manufacturer’s instructions and using microplate computer software (BioRad Laboratories). Statistical Evaluation All data were evaluated for normal distribution of the data. Statistical differences (p \ 0.05) were evaluated using oneway analysis of variance (ANOVA) followed by the leastsignificant difference (LSD) test for multiple comparisons. All data are presented as mean ± standard deviation (SD).
Results Slit2, Robo1, and VEGF Are Highly Expressed in Vascular Malformation Tissues
Scratch Wound Assay HUVECs were seeded in a 6-well tissue culture plate at a density of 2 9 105/well. After the cells had reached 90 % confluency, a scratch wound was made in the middle of the plate by a 100-ll pipette tip, generating a wound width of
We assembled a cohort of vascular malformation and normal gastrointestinal mucosal tissues from seven patients and immunohistochemically assayed for expression of Slit2. Detection with the Slit2 antibodies demonstrated that Slit2 was overexpressed in the dilated and twisted capillary
123
Dig Dis Sci Fig. 1 Slit2, Robo1 and VEGF are highly expressed in vascular malformation tissues. a Vascular malformation (VM) lesions (A1) and normal tissues (A2) were stained with anti-Slit2 antibody; A3 negative control staining of angiodysplasia tissue (scale bar 100 lm). b Immunoblot analysis for Slit2, Robo1 and VEGF expression in vascular malformation lesions and normal tissues.**p \ 0.01
A
B VM
Normal
Robo1 Slit2 VEGF
GAPDH
endothelial cells in the mucosal and submucosal layers of angiodysplasia lesions (Fig. 1A1), in 6 out of 7 of the patients’ tissues. On the contrary, Slit2 was only expressed slightly in the normal mucous membranes and vascular endothelial cells of the same person (Fig. 1A2). For assessment of Slit2, Robo1 and VEGF expression, we also analyzed the protein level in seven vascular malformation tissues and matched normal tissue lysates, by Western blot. Slit2, Robo1 and VEGF were overexpressed in comparison with normal levels by factors of 3.84 ± 0.83, 2.14 ± 0.45 and 2.90 ± 0.82, respectively (p \ 0.01; Fig. 1b). Thalidomide Inhibits Robo1, Slit2, and VEGF mRNA and Protein Expression in HUVECs We initially examined expression levels of Robo1, Slit2 and VEGF in HUVECs after treatment with thalidomide. Exposure of HUVECs to thalidomide (0, 0.2, 0.4 or 0.8 mM) decreased Robo1, Slit2 and VEGF mRNA levels, as measured by real-time qPCR (Fig. 2a). Thalidomide, at a concentration of 0.8 mM, induced the strongest inhibition of VEGF, Slit2 and Robo1 mRNA expression levels (mean inhibitions and SDs: 0.36 ± 0.04, 0.39 ± 0.08 and 0.32 ± 0.04, respectively), by approximately 60–70 % as compared to the control group (p \ 0.01). It was
123
subsequently tested whether Slit2, Robo1 and VEGF protein levels are also regulated by thalidomide. For Robo1, this was evaluated by Western blot analysis using total cell lysates from HUVECs cultured with various concentrations of thalidomide for 24 h. Robo1 protein expression, in response to 0.8 mM thalidomide, was lower than in the control group (0.53 ± 0.05, Fig. 2b, p \ 0.01). The secretion of Slit2 and VEGF protein by HUVECs was measured by ELISA and was also suppressed by thalidomide. Cells treated with thalidomide, at a concentration of 0.8 mM, induced the highest level of VEGF suppression (Fig. 2c, d, p \ 0.01).
Effects of Thalidomide and Recombinant N-Slit2 Protein on the Phosphorylation of Akt in HUVECs Next, we examined the effects of thalidomide and N-Slit2 on the PI3K/Akt signaling cascade in endothelial cells. Exposure of HUVECs to N-Slit2 (100 ng/ml) induced Akt phosphorylation as demonstrated by Western blot analysis. Treatment of HUVECs with thalidomide significantly inhibited Akt phosphorylation, which was similar to the effect of LY294002, a specific PI3K/Akt inhibitor. Thalidomide was also able to fully counteract the inducing activity of the N-Slit2 protein (Fig. 6).
Dig Dis Sci
A
Table 1 Real-time qPCR primer design Gene b-actin
Oligonucleotide primers (50 –30 )
Size (bp)
F: CTTAGTTGCGTTACACCCTT
144
R: CCTTCACCGTTCCAGTTT Slit2
F: CACCTCGTACAGCCGCACTT
104
R: TGTGGACCGCTGAGGAGCAA Robo1
F: GGAAGAAGACGAAGCCGACAT
107
R: TCTCCAGGTCCCCAACACTG
B
Tha
0
0.2
0.4
0.8
(mM)
Robo1
VEGF
F: AGGGAAGAGGAGGAGATG
148
R: GCTGGGTTTGTCGGTGTT
GAPDH
C
D
Fig. 2 Thalidomide inhibits Robo1, Slit2, and VEGF mRNA and protein expression in HUVECs. a HUVECs were treated with an increased dose of thalidomide (Tha) for 24 h. The mRNA levels of Slit2, Robo1 and VEGF were determined by real-time qPCR and normalized with GAPDH using absolute quantification. b HUVECs were treated with increased doses of thalidomide, and Robo1 protein levels were measured by immunoblot analysis. c, d HUVECs were treated with increased doses of thalidomide, and VEGF and Slit2 protein levels were measured using ELISA. The data represent mean values ± standard deviations (error bars) from three independent experiments. **p \ 0.01
Thalidomide Can Reverse the Promoting Effect of Recombinant N-Slit2 Protein on Migration, Proliferation, and Tube Formation of HUVECs To further elucidate the biological effects of Slit2 and thalidomide on migration, proliferation and tube formation of HUVECs, we performed scratch wound assays, proliferation assays and tube formation assays. Wound closure was clearly inhibited by exposure of cells to thalidomide or LY294002, as compared to the control group (Fig. 3a). In contrast, recombinant N-Slit2 significantly promoted the
Fig. 3 Effects of thalidomide and recombinant N-Slit2 protein on the phosphorylation of Akt in HUVEC. a Photograph of the immunoblot analysis for pAkt and Akt expression in HUVECs. b Expression levels of pAkt protein in HUVECs, *p \ 0.05, **p \ 0.01. The data represent mean values ± SDs (error bars) from three independent experiments. NC negative control, S Slit2 (100 ng/ml), T Thalidomide (0.8 mM), L LY294002 (10 lM)
migration capacity of HUVECs. The increasing effect of N-Slit2 on HUVECs could be reversed by co-incubation with both thalidomide and LY294002 (Fig. 3a). Besides, treatment of N-Slit2 promoted the proliferation of HUVEC which can also be reversed by thalidomide or LY294002 (Fig. 3b). Similarly, increased formation of tubular networks in the Matrigel was observed when HUVECs were incubated with N-Slit2. A gradual abrogation of the network formation occurred in the presence of thalidomide (Fig. 3c). Taken together, N-Slit2 promoted the migration, proliferation and tube formation activities of HUVECs, while thalidomide was able to suppress these activities.
123
Dig Dis Sci
of VEGF was increased by exposure to recombinant N-Slit2 protein, while HUVECs cultured with thalidomide or LY294002 decreased in VEGF expression level (Fig. 5), as compared to controls.
Discussion
Fig. 4 Effects of thalidomide and recombinant N-Slit2 protein on the expression of VEGF in HUVECs. HUVECs were treated with 100 ng/ ml recombinant N-Slit2 protein (S), 0.8 mM thalidomide (T), 10 lM LY294002 (L) or a combination. VEGF protein levels were assessed using ELISA. The data represent mean values ± SDs (error bars) from three independent experiments.*p \ 0.05, **p \ 0.01
Effects of Thalidomide and Recombinant N-Slit2 Protein on Cell Cycle After incubation with recombinant N-Slit2, thalidomide and LY294002, HUVECs were analyzed using flow cytometry to detect alterations in the cell cycle distribution. The recombinant N-Slit2 protein resulted in a significant reduction of HUVECs in the G0-/G1-phase of cell cycle and promoted an accumulation of cells in the S-phase, as compared to controls (Fig. 4). Of the cells treated with recombinant N-Slit2 protein, 53.14 % were in the G1-phase and 32.50 % were in S-phase of cell cycle, compared to 60.46 % of the control group in G1-phase and 27.38 % in S-phase (p \ 0.05). Thalidomide showed a contrasting effect (64.90 vs. 60.46 % in G1-phase and 19.81 vs. 27.38 % in S-phase, p \ 0.05). Thalidomide inhibited the N-Slit2-induced accumulation of cells in S-phase (32.50 vs. 22.26 %, p \ 0.01). In addition, PI3K inhibitor LY294002 shows a similar effect as thalidomide. These results illustrated that the contrary effects between thalidomide and N-Slit2 on angiogenesis were cell cycle-dependent and associated with PI3K-Akt pathway. Interestingly, no apoptotic peak appears in the cell cycle histogram (data not shown). To determine whether N-SLIT2 or thalidomide affects cell apoptosis, we examined the expression of Bcl-2 protein. The results illustrated that N-SLIT2 treatment could moderately up-regulate the Bcl-2 expression in HUVECs, and the presence of thalidomide and LY294002 could reverse this effect (Fig. 4b). Thalidomide Can Reverse the Promoting Effect of Recombinant N-Slit2 Protein on VEGF Expression of HUVECs We evaluated whether VEGF levels might be affected in HUVECs treated with N-SLIT2 or thalidomide. Expression
123
The Slit–Robo signaling pathway was originally characterized as a regulator of neuron axonal growth; the pathway also regulates angiogenesis3,4,7. Because there is no animal model for developing regional malformations, studies on this disease have been limited. Therefore, studies in patients are of great value. Given the low incidence and the low number of patients that undergo endoscopic treatment, the data presented here are based on a small sample size. However, we demonstrated for the first time that Slit2 is overexpressed in the vascular malformation regions of human gastrointestinal mucosa in patients with GIVM. Slit2 was strongly expressed near the enlarged, twisted vessels of the vascular malformation region, while it was detected only slightly in the normal tissues of the same patients (Fig. 1). In a previous study, we found that VEGF is also expressed more in angiodysplasia lesions20. Our results suggest that Slit2 signaling may actively contribute to the progression of angiogenesis and in this way participate in the processes underlying angiodysplasia in GIVM patients. In this study, we demonstrated by real-time RT-PCR, immunoblotting and ELISA that thalidomide can inhibit the expression of Robo1, Slit2 and VEGF (Fig. 2). These results suggest that Robo1 and Slit2 may be involved in the anti-angiogenesis action of thalidomide. Many studies have established that growth factorinduced endothelial cell migration and subsequent tube formation are known to be PI3K-Akt dependent [22–24]. We studied the PI3K/Akt signaling pathway using LY294002, a well-known inhibitor of PI3K. We found that N-Slit2 can promote the phosphorylation of Akt (Fig. 3) and the expression of VEGF (Fig. 4). Meanwhile, LY294002 was found to be able to block the effects of N-Slit2 on VEGF expression and phosphorylation of Akt. Results indicated that Slit2 could promote the proliferation, migration and tube formation of HUVECs, while thalidomide could reverse these effects (Fig. 4). VEGF is generally considered to be a key angiogenic factor in the neovascularization that occurs during GIVM2. VEGF signaling is tightly regulated in endothelial cells. This regulation includes cross talk with other signaling pathways, such as Notch [25–27] and Ang-2 [28, 29]. Our data show further evidence for a functional interplay between the Slit2/Robo1 pathway and VEGF signaling in GIVM. Based on the results presented, it is clear that recombinant N-Slit2
Dig Dis Sci Fig. 5 Thalidomide can reverse the promoting effect of recombinant N-Slit2 protein on migration, proliferation and tube formation of HUVECs. a Influence of N-Slit2 and thalidomide on migration ability of HUVECs. b CCK-8 proliferation assay results after thalidomide and recombinant N-Slit2 protein exposure of HUVECs for 24 h. c Effects of thalidomide and recombinant N-Slit2 protein on tube formation of HUVECs. *p \ 0.05, **p \ 0.01. The data represent mean values ± SDs (error bars) from three independent experiments. NC negative control, S Slit2 (100 ng/ml), T thalidomide (0.8 mM), L: LY294002 (10 lM) S ? L: Slit2 (100 ng/ ml) and LY294002 (10 lM)
protein promoted VEGF secretion in HUVECs (Fig. 4). Additionally, we found that thalidomide, by its ability to suppress the expression of Slit2 and Robo1, could inhibit this process while presenting a possible treatment option for GIVM (Fig. 4).
Endothelial cell migration and subsequent tube formation play an important role during angiogenesis [30]. Our data demonstrate that activation of Akt in HUVECs by Slit2 was involved in proliferation, migration and tube formation of HUVECs (Fig. 5).
123
Dig Dis Sci
Fig. 6 Effects of thalidomide and recombinant N-Slit2 protein on cell cycle. HUVECs treated with 100 ng/ml recombinant N-Slit2 protein (S), 0.8 mM thalidomide (T), 10 lM LY294002 (L) or a combination. a Cell cycle distribution was assessed using flow cytometry. b Bcl-2 protein levels were measured using Western blot. The data represent mean values ± SDs (error bars) from three independent experiments. *p \ 0.05, **p \ 0.01
Usually, endothelial cells are quiescent and reside in the G0-phase. We found that N-Slit2 protein significantly increases HUVEC proliferation. In this study, additionally, we observed that recombinant N-Slit2 protein may play a role in many of the processes involved in the early events of angiogenesis, such as migration and tube formation of endothelial cells (Fig. 6). At the same time, we also found that recombinant N-Slit2 protein caused significant reductions in the subset of cells in G0-/G1-phase and an accumulation of cells in S-phase. In addition, N-Slit2 upregulated the expression level of the anti-apoptosis protein Bcl-2 (Fig. 6). Taken together, these results suggest that Slit2 is important in initiating the pathogenesis of GIVM and that all these effect can be inhibited by thalidomide. Our previous study showed that thalidomide suppresses hypoxia- or bFGF-induced VEGF expression at the protein level18. In the present study, we found that the expression levels of Slit2/Robo1 and VEGF at both mRNA and protein levels were suppressed by thalidomide. We also demonstrated that the inhibitory effect of thalidomide on angiogenesis is attributable to inhibition of signaling via Akt phosphorylation, leading to reduced angiogenic response of HUVECs. Since the N-Slit2 can promote Akt phosphorylation (Fig. 3), as mentioned above, the anti-angiogenesis effect of thalidomide may be associated with the Slit2/ Robo1 pathway. Thalidomide itself also had an effect on the proliferation, migration and tube formation on HUVECs. By taking our study mentioned above (Fig. 4) in consideration, we postulate that this can be the result of endogenous Slit2
123
suppression. However, from our present data, we cannot exclude a role of PTEN stabilization and consequent inhibition of PI3K/Akt signaling in this process [31]. Studies have been dedicated to determining the mechanisms of thalidomide teratogenicity. Thalidomide induces dysfunction of embryo limb vessels at critical stages in limb development, prior to the onset of malformations, suggesting that vascular malfunction is an upstream event in hampering limb development [32]. However, several thalidomide analogues such lenalidomide [33] and pomalidomide [34] also show a significant anti-angiogenesis effect and less teratogenicity. These analogues also indirectly inhibit endothelial cell proliferation by inhibiting the secretion of VEGF and inducing G0/G1 arrest [35]. In summary, our study suggests that Slit2–-Robo1 signaling plays a role in GIVM. Increasing concentrations of Slit2 increased VEGF expression in HUVECs. Thalidomide was able to block the expression of Robo1 and Slit2. The recombinant N-Slit2 protein also resulted in the increased proliferation, migration and capillary tube formation of HUVECs. In addition, it increased the number of HUVECs in S-phase. Treatment of HUVECs with recombinant N-Slit2 protein, increased VEGF mRNA expression and VEGF secretion, as well as phosphorylation of Akt, which can also be inhibited by thalidomide as well as LY294002. Although preliminary, all of these results suggest that Slit2–Robo1 signaling may be involved in GIVM. Nevertheless, further research is required to elucidate the precise role of Slit2-Robo1 signaling in the development of GIVM and to examine its potential role as a therapeutic target of thalidomide. Acknowledgments The research presented here was funded by the Shanghai Municipal Health Bureau Academic Discipline Project, Project Number: 20114002. Conflict of interest of interest.
The authors declare that they have no conflict
References 1. Regula J, Wronska E, Pachlewski J. Vascular lesions of the gastrointestinal tract. Best Pract Res Clin Gastroenterol. 2008;22:313–328. 2. Junquera F, Saperas E, de Torres I, Vidal MT, Malagelada JR. Increased expression of angiogenic factors in human colonic angiodysplasia. Am J Gastroenterol. 1999;94:1070–1076. 3. Paul JD, Coulombe KL, Toth PT, et al. SLIT3-ROBO4 activation promotes vascular network formation in human engineered tissue and angiogenesis in vivo. J Mol Cell Cardiol. 2013;64:124–131. 4. Guo SW, Zheng Y, Lu Y, et al. Slit2 overexpression results in increased microvessel density and lesion size in mice with induced endometriosis. Reprod Sci. 2013;20:285–298. 5. Brose K, Bland KS, Wang KH, et al. Slit proteins bind Robo receptors and have an evolutionarily conserved role in repulsive axon guidance. Cell. 1999;96:795–806.
Dig Dis Sci 6. Wong K, Park HT, Wu JY, Rao Y. Slit proteins: molecular guidance cues for cells ranging from neurons to leukocytes. Curr Opin Genet Dev. 2002;12:583–591. 7. Kramer SG, Kidd T, Simpson JH, Goodman CS. Switching repulsion to attraction: changing responses to slit during transition in mesoderm migration. Science. 2001;292:737–740. 8. Adams RH, Eichmann A. Axon guidance molecules in vascular patterning. Cold Spring Harb Perspect Biol. 2010;2:a001875. 9. Sheldon H, Andre M, Legg JA, et al. Active involvement of Robo1 and Robo4 in filopodia formation and endothelial cell motility mediated via WASP and other actin nucleation-promoting factors. FASEB J. 2009;23:513–522. 10. Urbich C, Ro¨ssig L, Kaluza D, et al. HDAC5 is a repressor of angiogenesis and determines the angiogenic gene expression pattern of endothelial cells. Blood. 2009;113:5669–5679. 11. Wang B, Xiao Y, Ding BB, et al. Induction of tumor angiogenesis by Slit-Robo signaling and inhibition of cancer growth by blocking Robo activity. Cancer Cell. 2003;4:19–29. 12. Zhou W, Yu W, Xie W, et al. The role of SLIT-ROBO signaling in proliferative diabeticretinopathy and retinal pigment epithelial cells. Molecular Vision. 2011;17:1526–1536. 13. Liao WX, Laurent LC, Agent S, Hodges J, Chen DB. Human placental expression of Slit/ROBO signaling cues: effects of preeclampsia and hypoxia. Biol Reprod. 2012;86:111. 14. Zhou W, Yu W, Xie W, Huang L, Xu Y, Li X. The role of SlitROBO signaling in proliferative diabetic retinopathy and retinal pigment epithelial cells. Mol Vis. 2011;17:1526–1536. 15. Yang XM, Han HX, Sui F, Dai YM, Chen M, Geng JG. Slit-Robo signaling mediates lymphangiogenesis and promotes tumor lymphatic metastasis. Biochem Biophys Res Commun. 2010;396: 571–577. 16. Bartlett JB, Dredge K, Dalgleish AG. The evolution of thalidomide and its IMiD derivatives as anticancer agents. Nat Rev Cancer. 2004;4:314–322. 17. Danese S, Sans M, Spencer DM, et al. Angiogenesis blockade as a new therapeutic approach to experimental colitis. Gut. 2007;56:855–862. 18. Ge ZZ, Chen HM, Gao YJ, et al. Long-term effects of thalidomide for recurrent gastrointestinal bleeding due to vascular malformation: an open-label, randomized, parallel controlled study. Gastroenterology. 2011;141:1629–1637. 19. Lebrin F, Srun S, Raymond K, et al. Thalidomide stimulates vessel maturation and reduces epistaxis in individuals with hereditary hemorrhagic telangiectasia. Nat Med. 2010;16:420–428. 20. Regula J, Wronska E, Pachlewski J. Vascular lesions of the gastrointestinal tract. Best Pract Res Clin Gastroenterol. 2008;22: 313–328. 21. Tan H, Chen H, Xu C, et al. Role of vascular endothelial growth factor in angiodysplasia: an interventional study with thalidomide. J Gastroenterol Hepatol. 2012;27:1094–1101.
22. Chavakis E, Dernbach E, Hermann C, Mondorf UF, Zeiher AM, Dimmeler S. Oxidized LDL inhibits vascular endothelial growth factor-induced endothelial cell migration by an inhibitory effect on the Akt/endothelial nitric oxide synthase pathway. Circulation. 2001;103:2102–2103. 23. Fujio Y, Walsh K. Akt mediates cytoprotection of endothelial cells by vascular endothelial growth factor in an anchoragedependent manner. J Biol Chem. 1999;274:16349–16354. 24. Nakamura JL, Karlsson A, Arvold ND, et al. PKB/Akt mediates radiosensitization by the signaling inhibitor LY294002 in human malignant gliomas. J Neurooncol. 2005;71:215–222. 25. Benedito R, Roca C, So¨rensen I, et al. The notch ligands Dll4 and Jagged1 have opposing effects on angiogenesis. Cell.. 2009;137:1124–1135. 26. Larrive´e B, Prahst C, Gordon E, et al. ALK1 signaling inhibits angiogenesis by cooperating with the Notch pathway. Dev Cell. 2012;22:489–500. 27. Hernandez SL, Banerjee D, Garcia A, et al. Notch and VEGF pathways play distinct but complementary roles in tumor angiogenesis. Vasc Cell. 2013;5:17. 28. Zhu Y, Lee C, Shen F, Du R, Young WL, Yang GY. Angiopoietin-2 facilitates vascular endothelial growth factor-induced angiogenesis in mature mouse brain. Stroke. 2005;36:1533–1537. 29. Loukovaara S, Robciuc A, Holopainen JM. Ang-2 upregulation correlates with increased levels of MMP-9, VEGF, EPO and TGFb1 in diabetic eyes undergoing vitrectomy. Acta Ophthalmol. 2013;91:513–519. 30. Saunders WB, Bohnsack BL, Faske JB, et al. Coregulation of vascular tube stabilization by endothelial cell TIMP-2 and pericyte TIMP-3. J Cell Biol. 2006;175:179–191. 31. Knobloch J, Schmitz I, Go¨tz K, Schulze-Osthoff K, Ru¨ther U. Thalidomide induces limb anomalies by PTEN stabilization, Akt suppression, and stimulation of caspase-dependent cell death. Mol Cell Biol. 2008;28:529–538. 32. Therapontos C, Erskine L, Gardner ER, Figg WD, Vargesson N. Thalidomide induces limb defects by preventing angiogenic outgrowth during early limb formation. Proc Natl Acad Sci USA. 2009;106:8573–8578. 33. Dredge K, Horsfall R, Robinson SP, et al. Orally administered lenalidomide (CC-5013) is anti-angiogenic in vivo and inhibits endothelial cell migration and Akt phosphorylation in vitro. Microvas Res. 2005;69:59–63. 34. Terpos E, Kanellias N, Christoulas D, Kastritis D, Dimopoulos MA. Pomalidomide: a novel drug to treat relapsed and refractory multiple myeloma. Onco Targets Ther. 2013;6:531–538. 35. Dredge K, Marriott JB, Macdonald CD, et al. Novel thalidomide analogues display anti-angiogenic activity independently of immunomodulatory effects. Br J Cancer. 2002;87:1166–1172.
123
