Background
Antiangiogenesis is a promising approach to cancer therapy. As known, several antiangiogenic agents are currently under investigation in clinical trials. In contrast to those conventional therapies that kill tumor cells directly, angiogenesis inhibitors suppress tumor growth by blocking the formation of new blood vessels, which provide oxygen and nutrients for tumor growth. Endostatin (ES), a 20-kDa fragment cleaved from the collagen XVIII COOH terminus that inhibits endothelial cell proliferation and migration, is a well-known angiogenesis inhibitor, which shows antiangiogenesis and antitumor activities in several animal models. ES inhibits 65 different tumor types and modifies 12% of the human genome to down-regulate pathological angiogenesis [
1]. However, the mechanism and function of ES is still insufficient understanding. For anti-angiogenic activity, ES appears to be dependent on binding to E-selectin [
2]. Also, ES blocks activity of metalloproteinases 2, 9, and 13 [
3]. ES may down-regulate VEGF expression in tumor cells [
4]. IGF-II-mediated signaling and T-type Ca
2+ channels also involve the function of ES [
5,
6].
Shi
et al. identified that cell surface nucleolin on angiogenic blood vessels is a functional receptor for ES, and mediates the internalization and biological activities of ES [
7,
8]. Mechanism studies by Huang et al. show that vascular endothelial growth factor (VEGF) and extracellular matrix (ECM) synergistically induce the translocation of nucleolin from nucleus to cell surface [
9‐
11]. Previous studies show that ES specifically binds to neovascular endothelial cells through its interaction with the integrin receptors α5β1 and αVβ3, which has been implicated in tumor metastasis [
12,
13]. A more recent study shows that nucleolin and integrin α5β1 can form a co-receptor for ES
via UPAR on the endothelial cell membrane [
14]. ES labeled with a near-IR probe is shown to selectively accumulate in the tumor site [
15]. All these studies suggest that ES has a unique ability for targeted cancer therapy.
However, like many angiogenesis inhibitors, ES single administration didn’t achieve significant effects. The clinical development ended in the U.S. in 2003 due to limited efficacy and problems with protein formulation and application [
16]. Several studies reported the improved selectivity and efficacy of chimeric molecules comprised of toxins or other cytotoxic agents with targeting agents on tumor vasculature, such as vascular endothelial growth factor receptor-gelonin; Shiga-like toxin-vascular endothelial growth factor fusion protein and anti-TES-23 linked to neocarzinostatin [
17‐
19]. So the combination of the targeted and cytotoxic effects by engineering two independent molecules sounds to be a promising way for drug design.
Lidamycin (LDM), also called C-1027, is a member of chromoprotein family of antitumor antibiotics. The LDM molecule consists of an enediyne chromophore (AE) and a non-covalently bound apo-protein (LDP). It was shown that the AE exerts extremely potent cytotoxicity to cultured cancer cells, whereas the apo-protein LDP keeps the labile enediyne relatively stable. The non-covalently bound AE and LDP can be dissociated and re-associated. The activity of rebuilt molecule remains as potent as that of natural LDM. LDP, which is composed of 110 amino acid residues, showed specific binding capability to various human tumor tissues and displayed moderate cytotoxicity to Bel-7402 cells [
20,
21]. This specific binding capability and cytotoxicity of LDP implied its potential use as a targeting drug carrier in the design of new anticancer agents.
In order to combine the anti-angiogenic and cytotoxic functions of ES and LDM and to target both tumor endothelial cells and tumor cells, we designed two novel ES-based fusion proteins, ES-LDP and LDP-ES and their enediyne-energized analogs, and then detected their antitumor efficacies. Here we show that ES-LDP fusion proteins should possess targeting property of ES or LDP and moderate cytotoxicity effect of LDP in addition to antiangiogenesis activity of ES and the extremely potent cytotoxicity of the enediyne chromophore of LDM when they were assembled.
Methods
Cells and cell culture
HMEC cell line was maintained in endothelial-specific medium EBM-2 (Lonza, USA). The human lung carcinoma PG-BE1 was routinely grown in RPMI-1640 (HyClone,Beijing, China) supplemented with 10% fetal bovine serum (Gibco, USA), 100 U/mL penicillin, and 100 μg/mL streptomycin. The mouse breast cancer cell line 4T1 cells expressing the firefly luciferase gene (4T1-luc) were preserved in our laboratory. For stable expression, the cells were exposed to 500 μg/mL G418 (Gibco, USA). D-luciferin was purchased from Xenogen (Alameda, CA).
Construction of the expression vectors
Two fusion proteins named LDP-ES and ES-LDP were designed with an eight-amino acid-long linker (−GGGSGGSG-) between LDP and ES. Each ES-based fusion protein gene consists of the gene encoding LDP (110 amino acids; ref. 21), ES (184 amino acids; ref. 27), and the linker peptide. After two rounds of PCR and DNA cloning process, the resultant 909-bp fragment was digested by NdeI/XhoI and was inserted into pET30a expression vector to generate the expression plasmid. DNA sequencing analysis (Invitrogen Corp.) was used to verify that the gene was correct in sequence and had been cloned in the frame.
Wound healing assay
Cell migration was assessed in a wound-healing assay. HMEC or 4T1 Cells at 5x105 cells per well were cultured in 24-wells plate provided by the CytoSelect™ 24-Well Wound Healing Assay Kit and allowed to proliferate to form a confluent monolayer. The linear spacer inserted in the well was removed, which created a regular and defined “wound” within the cell monolayer. Wash wells with media to remove dead cells and debris. Wells were treated with different concentrations of ES, ES-LDP or LDP-ES and further cultured until the control wound was fully closed at 37°C. Cells were fixed and images were captured immediately at 40X magnification from light microscopy and cells that migrated to the scraped area were counted using Image-Pro Plus 6.0 software. Each experiment was performed twice, with triplicate samples.
Formation of capillary tube like structures by HMEC was assessed in Matrigel-based assay. Briefly, a 96-well plate coated with 60 μl of Matrigel per well was allowed to solidify at 37°C for 1 h. Cells (1.5 × 104 in 100 μl medium) were added on each well and 100 μl of medium containing different concentrations of ES, ES-LDP or LDP-ES were added and incubated for different periods of time. Each treatment was performed in triplicate. The enclosed networks of tubes were photographed under microscope. The total tube lengths and numbers of the tube structure of each photograph were measured using Image-Pro Plus 6.0 software.
Immunohistochemistry in tissue microarray
Multiple arrays of formalin-fixed, paraffin-embedded lung tumors and normal lung tissue were obtained from U.S. Biomax, Inc. (Xi’an, China). The microarray of product number BC041115a contains 110 lung tumors and unmatched normal lung tissue (10 cases/type). The normal controls were derived from the same organ but not from the same patient. The array dot diameter was 0.6 mm. All immunohistochemical studies were performed on paraffin-embedded sections as previously described [
21]. For Cetuximab controls, the tissue sections were stained in the same manner except that the detection antibody was replaced with poly-HRP-anti-human IgG against Cetuximab. The positive percentage of each protein could be calculated according to the staining intensity by reference to the Herceptest™ interpretation manual.
Additionally, we analyzed the cases by Image-Pro Plus 6.0 software, using the method introduced by Xavier et al. [
22,
23]. Briefly, the measurement parameters included density mean, area sum, and integrated optical density (IOD). The optical density was calibrated and the area of interest was set through: hue, 0 ~ 30; saturation, 0 ~ 255; intensity, 0 ~ 255, then the image was converted to gray scale image, and the values were counted. The time required to perform the analysis process can be greatly reduced by using macro of pathology. To avoid artificial effect, cells in areas with necrosis, poor morphology, or in the margins of sections were not taken into account. The IOD were log transformed and mainly performed statistical analysis.
Preparation of enediyne-energized ES-LDP and LDP-ES
The active enediyne chromophore (AE) of LDM was separated by using C4 column (GE Healthcare) with a 22% acetonitrile in 0.05% trifluoroactic acid mobile phase. The AE-containing solution was added to ES-LDP/PBS (10 mmol/L; pH7.4) or LDP-ES/PBS, respectively, with the molecular ratio of 4:1, and was incubated at 4°C for12 h while rocking. Free AE was removed by using a Sephadex G-75 column (GE Healthcare). Assembled enediyne-energized fusion proteins named LDP-ES-AE and ES-LDP-AE were confirmed by reverse-phase HPLC using a Vydac C4 300A column (Grace). Absorbance at 340 nm was measured.
Cell cytotoxicity assay by cell counting kit-8
Cells were seeded at 1 × 104 per well in 96-well plates and incubated in 37°C for 24 h and then exposed to different concentrations of LDM or energized fusion proteins (ES-LDP-AE, LDP-ES-AE) for 48 h. On the day of measuring the growth rate of cells, 100 μL of spent medium was replaced with an equal volume of fresh medium containing 10% CCK-8 (WST-8, Dojindo Laboratories, Tokyo, Japan). Cells were incubated at 37°C for 1 h, and cell number was assessed by measuring the absorbance at 450 nm on a microplate reader (Thermo). Three independent experiments were carried out. The IC50 represented the drug concentration resulting in 50% growth inhibition.
Tumor models
The syngeneic murine 4T1-luc breast cancer model and human lung carcinoma PG-BE1 xenograft model have been used. The BALB/c female mice and female athymic nude mice (BALB/c, nu/nu) were purchased from the Institute for Experimental Animals, Chinese Academy of Medical Sciences & Peking Union Medical College. The study protocols were in accordance with the regulations of Good Laboratory Practice for non-clinical laboratory studies of drugs issued by the National Scientific and Technologic Committee of People’s Republic of China. The treatment and use of animals during the study was approved by the Animal Ethics Committee of the Institute of Medicinal Biotechnology, Chinese Academy of Medical Sciences & Peking Union Medical College (permission number: c1-2011-1121).
Exponentially growing human lung carcinoma PG-BE1 cells were implanted into the 16–18-week-old female athymic nude mice by the subcutaneous injection of 10 × 106 cells on the right flank. After 3 weeks, the tumors were aseptically dissected and pieces of tumor tissue (2 mm3 in size) were transplanted s.c. separately by a trocar into athymic mice. When tumors reached about 100 mm3 in size, the mice were randomized into groups (n = 6 per group) and treated with ES,LDM,ES-based fusion proteins (ES-LDP, LDP-ES) and energized fusion proteins (ES-LDP-AE, LDP-ES-AE), respectively, at different doses and time intervals. Tumor growth was measured with a caliper, and tumor volumes were calculated with the following formula: V = 0.5a × b2, where a and b are the long and the perpendicular short diameters of the tumor, respectively. Typically, studies were terminated when tumors in the control animals reached an average size of 2000 mm3. Percentage of inhibition of tumor growth was calculated as 100 × {1-[(tumor volumefinal-tumor volumeinitial for the treated group)/(tumor volumefinal-tumor volumeinitial for the vehicle-treated group)]}.
We studied the lung metastasis of tumors using an i.v. injection model. BALB/c female mice were injected with 2 × 105 murine 4T1-luc breast cancer cells in 0.2 mL PBS solution via the lateral tail vein. Three days later after tumor cell injection, mice were randomly assigned to three groups and treated with ES or ES-LDP respectively. Seven days after the first treatment, all mice were injected again at same doses. After 17 days, Mice were anesthetized with isoflurane and i.p. injected with luciferase substrate D-luciferin (150 mg/kg). The animals were placed onto the warmed stage inside the camera box (IVIS-Imaging System, Xenogen) to observe tumor growth. Then, the lungs were immediately removed, weighed and fixed in 10% buffered formalin for counting of pulmonary metastatic nodules. The metastatic nodules of 4T1 tumor in lung were counted by direct visualization using a stereomicroscope. The total number of metastases per lung section was counted and averaged among the animals.
In vivo fluorescence imaging
When tumors reached about 200 mm3 in size in human PG-BE1 xenograft model, three hundred micrograms of DyLight 680-labeled ES-LDP or LDP-ES were injected i.v. (n = 3). The mice were placed under anesthesia by inhalation of isoflurane and the images were observed with the Xenogen Ivis 200 system and recorded by built-in camera (Caliper Life Sciences).
Statistical analysis
All of the data were presented as the mean ± SD for at least three independent experiments. Statistical analysis was performed with SPSS software (version 17.0). The significant differences between any of two groups were evaluated by One-way ANOVA. Statistical significance was defined as P < 0.05.
Discussion
ES, an angiogenesis inhibitor having been tested in multiple clinical trials, selectively targets endothelial cells in neovascularization and suppresses tumor growth. However, like other angiogenesis inhibitors, such as bevacizumab, sunitinib and sorafenib, ES could help patients to survive longer when given in combination with chemotherapy, but not when given alone [
26]. To enhance the therapeutic efficacy of ES, several ES derivatives with different modifications have been designed, which include ES-cytosine deaminase protein, prolactin antagonist-ES, anti-HER2 IgG3-ES, ZBP-ES (Endostar), Fc-ES, and cell-permeable ES protein (HM
73ES) [
27‐
32].
The ES-cytosine deaminase protein, which converts a non-cytotoxic prodrug 5-fluorocytosine (5-FC) to the cytotoxic antitumor drug 5-fluorouracil (5-FU) in the local tumor area, significantly inhibited the growth of endothelial cells and preferentially induced tumor cell apoptosis [
28]. The prolactin antagonist-ES fusion protein is a bifunctional protein, which inhibits both breast cancer cell proliferation and endothelial cell proliferation, exhibiting greater tumor inhibitory effects than prolactin antagonist and ES treated individually or in combination [
29]. Targeting of ES using anti-HER2 antibody and human ES fusion protein could improve antitumor activity of either anti-HER2 antibody and/or ES and provides the versatile approach that could be applied to other tumor targets with alternative antibody specificities [
30]. ZBP-ES, engineered by adding 9 extra amino acid residues MGGSHHHHH to the N-terminus of ES, showed increased thermodynamic stability and biological activity than the wild type ES [
27], and was approved as anti-cancer drug in China. As reported, Fc-ES is a superior molecule to the original clinical ES. Due to its long half-life, the amount of protein required is substantially reduced compared with the clinically tested ES [
31]. HM
73ES exhibited enhanced tissue penetration and suppressed the growth of human tumor xenografts to a significantly greater extent than unmodified ES by adding a macromolecule transduction domain (MTD). Those results suggest another important mechanism to explain the enhanced activity of ZBP-ES and ES lacking the MTD sequence [
32].
Recent studies indicated that LDP itself, the apoprotein of LDM, shows binding capability to a spectrum of human tissues, and notably that the binding capability correlates with the overexpression of EGFR and HER2 on the tumor tissue microarray [
21]. In addition, LDP displayed moderate cytotoxicity to human hepatoma Bel-7402 cells with an IC50 value of 7.05 × 10
-5 mol/l and it exerted tumor suppression on hepatoma H22 in Kunming mice [
20]. The functional receptor of ES nucleolin was found to be specifically expressed on the surface of angiogenic blood vessels in tumor tissues, which endows ES low toxicity and tumor-specific distribution [
7]. Therefore, in the present study, we constructed and prepared two ES-based fusion proteins ES-LDP and LDP-ES. Our results indicate that ES-LDP and LDP-ES disrupted the formation of endothelial tubule structures with the potency similar to that of ES. In addition, ES-based fusion proteins, especially ES-LDP, demonstrated much stronger inhibition of HMEC migration than ES. Furthermore, ES-LDP displayed high efficacy in PG-BE1 xenografts. This may be explained by the reason that N-terminal loop of ES around the zinc-binding site was involved in activity [
33], and that the N-terminal integrity is essential for the biological functions of ES [
27]. So it appears that the anti-tumor activity of ES could be enhanced by integrating with LDP and a free N-terminus of ES in the fusion protein is preferred. But its mechanism should be studied further, and there is more work to be done.
On the other hand, angiogenesis is involved in the development of distant metastasis. Thus by targeting angiogenesis, ES directly suppresses not only the growth of primary tumors but also metastasis. The present study has shown that ES-LDP or LDP-ES could markedly inhibit 4T1 cells migration in wound healing assay, with ES-LDP exhibiting a more potent activity. Therefore, we examined the effects of ES-LDP on the lung metastasis of 4T1-luc tumors. ES-LDP treatment significantly reduced the number of lung surface metastasis and lung weight gain in tumor-bearing animals. Therefore, ES-LDP could be explored as a novel therapeutic molecule in controlling metastasis of cancer.
LDM is considered as a highly potent “warhead” molecule for the construction of antibody-based tumor targeting drugs. At present, several LDM-containing energized fusion proteins have been manufactured with the two-step procedure in our laboratory, such as bispecific enediyne-energized fusion protein (Ec-LDP-Hr-AE) and tandem scFv-based enediyne-energized fusion protein (dFv-LDP-AE) [
25,
34]. Both of them possessed highly potent cytotoxicity to cancer cells and significant inhibitory efficacy
in vivo. Ec-LDP-Hr-AE was more potent and selective in its cytotoxicity against different carcinoma cell lines
in vitro and significantly inhibited the growth of SK-OV-3 xenografts in nude mouse model [
34]. dFv-LDP-AE displayed extremely potent cytotoxicity to kinds of cancer cells, especially the lung cancer cell lines, and greatly increased the antitumor efficacy with lung carcinoma PG-BE1 xenograft in nude mice [
25].
We adopted a strategy to energize the ES-based fusion proteins, ES-LDP and LDP-ES, with LDM enediyne chromophore to prepare ES-based and enediyne-energized fusion proteins (ES-LDP-AE and LDP-ES-AE). They all displayed potent antitumor activities against a variety of tumor cell lines with IC50 values ranged from 10
-9 M to 10
-10 M. Though the IC50 values had ten-fold difference, the IC50 value of ES-LDP-AE was always less than that of LDP-ES-AE. This difference may be due to the assembling efficiency of ES-LDP and LDP-ES, which was 83.9% and 27.1%, respectively. These results accord with assembling efficiency, and potential conformational change of the AE binding sites caused by the fusion. In the
in vivo study, mice received tolerated dose of LDM at 0.05 mg/kg showed an inhibition rate of 61.1%. By contrast, ES-LDP-AE and LDP-ES-AE at equivalent doses suppressed the tumor growth by 78.5% and 75.8%, respectively. Furthermore, the ES-LDP-AE-treated group at higher dosage of 0.30 mg/kg showed an inhibition rate of 86.4%. No deaths were found in all treatment groups. As previously mentioned, ES has a unique ability for targeting therapy of cancer [
15]. Endostar is now in clinical use for lung cancers in China, so we investigated the affinity of these ES-based fusion proteins to human lung cancers by tissue-microarray analysis. As shown, the positive percentage of ES and ES-LDP was higher than that of LDP; in addition, ES and ES-LDP share similar binding capability to lung cancer tissue, indicating that the fusion protein ES-LDP retains this capability as of the ES. It is of interest that the integration of LDP into the fusion protein ES-LDP does not compromise ES binding capability, while probably provides a targeting delivery of lidamycin.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
WGJ carried out the cell experiments and was responsible for data analyses, manuscript preparation and editing. XAL constructed the vectors and DFZ helped to provide the fusion proteins. LL helped to prepare the enediyne-energized fusion protein. YL participated in IHC stainings from the TMAs. BYS and SHZ were involved in vivo study. YF helped with the interpretation of the results and with drafting the manuscript. YZL and YSZ designed the overall study, coordinated the study and helped to draft and finalize the manuscript. All authors read and approved the final manuscript.