Chinese Medical Journal 2011;124(4):581-585
Screening of specific binding peptide targeting blood vessel of human esophageal cancer in vivo in mice
ZHI Min, WU Kai-chun, HAO Zhi-ming, GUO Chang-cun , YAO Jia-yin
ZHI Min (Department of Gastroenterology, Sixth Affiliated Hospital of Sun Yat-sen University, Guangzhou, Guangdong 510655, China)
WU Kai-chun (Institute of Digestive Diseases, Xijing Hospital, Fourth Military Medical University, Xi’an, Shaanxi 710032, China)
HAO Zhi-ming (Department of Gastroenterology, First Hospital of Xi’an Jiaotong University, Xi’an, Shaanxi 710068, China)
GUO Chang-cun (Institute of Digestive Diseases, Xijing Hospital, Fourth Military Medical University, Xi’an, Shaanxi 710032, China)
YAO Jia-yin (Department of Gastroenterology, Sixth Affiliated Hospital of Sun Yat-sen University, Guangzhou, Guangdong 510655, China)Correspondence to:WU Kai-chun,Institute of Digestive Diseases, Xijing Hospital, Fourth Military Medical University, Xi’an, Shaanxi 710032, China (Tel: 86-29-83375230. Fax:86-29-82539041. E-mail:firstname.lastname@example.org)
Background Cancer of the esophagus and gastroesophageal junction remains a virulent malignancy with poor prognosis. Rapid progresses were made in chemotherapeutic agents and the development of molecular markers allowed better identification of candidates for targeted therapy. This study aimed to identify the candidate peptides used for anti-angiogenic therapy of esophageal cancer by in vivo screening C7C peptide library for peptides binding specifically to blood vessels of human esophageal cancer.
Methods The phage displayed C7C peptide library was injected intravenously into mice bearing human esophageal tumor xenografts under renal capsule. After 5 rounds of screening, 13 clones were picked up individually and sequenced. During each round of screening, titers of phage recovery were calculated from tumor xenograft and control tissues. Homing of these 9 peptides to tumor vessel was detected by calculating phage titers in the tumor xenograft and control tissues (lung and spleen) after each phage was injected into mice model, and compared with the distribution of phage M13 and VIII-related antigen in tumor xenograft by immunohistochemical staining. Comparisons among groups of data were made using one-way analysis of variance (ANOVA), followed by the Bonferroni multiple comparisons test.
Results The number of phage recovered from tumor tissue of each round increased gradually in tumor group while decreased in control groups (P <0.01 in tumor and spleen, P <0.05 in lung). Immunohistochemical staining showed similar staining pattern with M13 antibody or VIII-related antigen antibody, suggesting that phages displaying the selected peptides could home to blood vessel of human esophageal cancer. According to their DNA, 9 corresponding peptide sequences were deduced. And the homing ability to blood vessel of phages displaying the selected peptides was confirmed by comparing with their recovery in tumor and control tissues. Two motifs, YSXNXW and PXNXXN, were also obtained by analyzing the homology of these peptide sequences. The staining distribution of phage with the sequence of PNPNNST was similar to that of the blood vessel marker factor VIII-related antigen staining. After sequencing, each phage with the selected peptide of PNPNNST with 1.0×1011 pfu/ml was injected intravenously into mice. The homing ability to tumor vessel of these 9 kinds of peptides in the xenograft was higher than control tissues (lung and spleen).
Conclusion Nine peptides obtained from in vivo screening homed to the blood vessel of human esophageal cancer, and the two motifs of YSXNXW and PXNXXN are the possible biochemical recognition units binding to vascular endothelial cells of esophageal cancer.
Phage-displayed library and mice model
The Ph.D.-C7C Phage Display Peptide Library Kit (New England Biolabs, England) was used in screening. The titer of the library is 2×1013 pfu/ml. The library contains a complexity of 1.2×109 individual clones, representing the entire obtainable repertoire of cyclic 7-mer peptide sequences, which expresses random 7-amino acid sequence with a structural constraint imposed by a disulfide bond between two cysteine residues flanking the variable region.
The 2-month-old Balb/c mice were purchased from and raised in the Laboratory Animal Research Centre of the Fourth Military Medical University. The mice were maintained in shoebox cages on a 12-hour light/dark cycle. Immunosuppressed mice were prepared by intraperitoneally injecting the mice with cyclophosphamide of 150 mg/kg by one day before tumor implantation. On the second day, fresh human esophageal squamous carcinoma tissue was obtained from surgical operation and placed in RPMI 1640 as soon as possible to ensure the viability of carcinoma cells. Necrotic tissue was removed from the tumor tissue, and tumor masses were cut into 1 mm × 1 mm × 1 mm pieces in the super clean bench. Subsequently, tumor xenograft was implanted into kidney capsule of the mice by subrenal capsule assay (SRCA).9 The mice were used on the sixth day after operation for in vivo screening.
Screening in vivo
The mice were firstly anesthetized deeply using 70 mg/kg body weight of sodium phenobarbital and then injected intravenously (via tail vein) with 200 µl (1.0×1012 pfu/ml) of the phage display C7C peptide library. Five minutes later (allowing the phages to circulate in the mice), the mice heart was exposed and perfused with 50 ml of 37°C prewarmed RPMI 1640 from the left ventricle. The tumor xenograft and control organs (spleen and lung) were homogenized individually in 1 ml ice-cold RPMI 1640-PI (containing 1 mmol/L PMSF, 20 mg/L aprotinin, and 1 mg/L leupeptin, Sigma, USA), and washed four times with 5 ml ice-cold RPMI 1640-PI-BSA (containing protease inhibitors and 1% BSA, Sigma). Then the phages were rescued by infection with Escherichia coli 2738. Phages homing to tumor xenograft were evaluated by in comparison with blue plaque-forming assay on agar plate containing tetracycline. For each round of in vivo selection, all blue plaques were scraped and grown in 200 ml of Luria-Bertani (LB) nutrient agar together for 4.5 hours at 37°C. After phages were centrifuged and purified, 1.0×1012 pfu/ml of the phage particles diluted in 200 μl of PBS were reinjected into tumor xenograft mice for another round of screening.
Thirteen blue plaques were picked up randomly for sequencing after the fifth round of panning. Single-stranded DNA was extracted from phage supernatant preparations using M13 purification kit (Promega, USA).10 Sequencing reaction was conducted according to the protocol provided in ABI PRISM BigDye Terminator sequencing kit and sequenced on ABI PRISM 377 automatic sequencing equipment (PE Biosystems) with M13 “-96” primer (-96 gIII, 5′-HOCCC TCA TAG TTA GCG TAA CG-3′, provided in C7C kit, New England Biolabs). The amino acid sequences were derived from the nucleotide sequences.
Homing of the phages displaying selected peptides to blood vessel in vivo
After sequencing, each phage with the selected peptide was obtained and grown in 5 ml of LB individually for 4.5 hours at 37°C. After being centrifuged and purified, the selected phages were maintained in 4°C. And then, individual phage with 1.0×1011 pfu/ml was injected intravenously into mice with human esophageal tumor xenograft by SRCA. The blue plaques of tumor and control tissues (spleen and lung) were recovered and calculated. The titers in different tissues were compared.
After the fifth round, the xenograft tumor and organs from mice were removed and fixed in 40 g/L formaldehyde to detect the binding of phage displaying peptide towards vessels by immunohistochemistry.11 These paraffin-embedded sections were dewaxed in xylene, dehydrated in alcohol, washed with PBS and blocked with goat serum. The sections were incubated with antiserum against M13 phage as primary antibody (Pharmacia, USA) in a dilution of 1:100. Then, after the procedure of avidin-biotin complex method kit (Zhongshan, China), we developed the sections using diaminobenzidine and counterstained with hematoxylin. To make sure the distribution of the phage in tumor was similar to it was in the blood vessels, the blood vessels of xenograft tumor were detected by VIII-related antigen staining (Zhongshan, China). The control tissues were still lung and spleen. PBS was used instead of primary antibody for negative control. Both were included in each staining batch.
All statistical analyses were conducted by the statistical program SPSS 17.0 for windows (SPSS Inc, Chicago, USA). Data were expressed as the mean ± standard deviation (SD). Comparisons among groups of data were made using one-way analysis of variance (ANOVA), followed by the Bonferroni multiple comparisons test. Statistical significance for expression analysis was also assessed by ANOVA. A P value <0.05 was considered statistically significant.
Specific binding ability to the blood vessels of phages in vivo
We adopted the method of transplanting fresh human esophageal cancer tissue in immunosuppressed Balb/c mice under subrenal capsule assay to establish a model for screening human vascular specific protein through phage display peptide library. The mice were used on the sixth day. Results reveal that human esophageal squamous tumor tissue grown very well under the subrenal capsule and there were abundant blood vessels around tumor cells in the tumor xenograft of human esophageal squamous tumor tissue (Figure 1A). The immunohistochemical staining of VIII-related antigen also confirmed that (Figure 1B).
|Figure 1. Specific binding to the blood vessels of phages in vivo. A: Xenograft of human esophageal squamous tumor (HE, original magnification ×200). K: the kidney of mice, B: blood vessel, T: tumor cell of human esophageal squamous tumor cells. The distribution of phage VIII-related antigen (B; SABC, original magnification ×400) and M13 (C; SABC, original magnification ×400) in the xenograft tumor by immunohistochemical staining were shown.|
Phage recovery and homing ability of different peptide in vivo
To identify whether phages were selectively homing to esophageal cancer vasculature, we calculated the phages titer of the xenograft tumor and control tissues during each round of panning. The number of phage recovered from tumor tissue of each round increased gradually and the number of the fifth round was about 10.63 folds than it is in the first round (P <0.01). At the same time, the titers of spleen and lung used as control tissues decreased (P <0.01 in spleen and lung) using analysis of variance of repeated measurement data (Figure 2). Thirteen individual phage clones isolated and sequenced from the fifth round of panning. These thirteen clones displayed nine kinds of sequences and two kinds of motifs, YSXNXW and PXNXXN (single letter code, Figure 3) appeared frequently. When these nine phages displayed the peptide were injected into respectively mice to investigate their homing to blood vessels of tumor, the blue plaques numbers obtained from tumor tissue were more by far than these obtained from control tissues. The homing ability to tumor vessel of these 9 kinds of peptides was detected by comparing with the titer in the xenograft and control tissues (lung and spleen) after each kind phage display peptide we obtained was injected into mice model. There was significant statistical difference in sequence YSFNSWM, PNPNNST, YSINDWH, LPAMPNS, YPTPYDI, PMNADNL, SRHDLNS and STVATSQ between tumor group and the two control group as using analysis of variance. There are 7.08 folds in tumor than in lung (P <0.05) or 5.09 folds (P <0.05) than in spleen in sequence YSINDWH; and 2.87 folds in spleen (P <0.05) while 1.47 folds in lung in the sequence PNPNNST (P <0.05). Therefore, the sequence PXNXXN is the object we are to research by using immunohistochemistry in the next step (Figure 4).
|Figure 2. Phage recovery of rounds of screening from various tissues in vivo. The phage recovery from tumor increased gradually and there was a significant difference in round 4th and round 5th than on round 1st (*P <0.01).|
|Figure 3. Amino-acid residues of the phage clones. After the fourth and the fifth screening, 13 phage clones were picked up and sequenced. Based on the deduced amino acid quences after DNA sequencing, 9 kinds of corresponding peptide sequence were obtained. We also obtained two motifs, YSXNXW and PXNXXN, which appeared frenquently in these sequences. In the figure, single letter represents amino acid residues. Italic, bold, or underlined letters indicates amino acid residues which were occurred frequently and indicates motifs of sequence. Letter X represents any amino acid residues.|
|Figure 4. Different homing ability of different phage displaying peptide to blood vessels in vivo. The homing ability to tumor vessel of these 9 kinds of peptides was detected by comparing with the titer in the xenograft and control tissues (lung and spleen) after each kind phage display peptide we obtained was injected into mice model. *P <0.05, †P <0.01.|
Homing ability of PNPNNST peptide in vivo
After sequencing, each phage with the selected peptide with 1.0×1011 pfu/ml was injected intravenously into mice with human esophageal tumor xenograft by SRCA.
The titers in different tissues were compared. It is show that these 9 kinds of peptide all has homing ability toward tumor vessel, the peptides with the highest homing ability are the phages with the sequences of PNPNNST, STVATSQ and YSINDWH. To find the distribution of the binding phage, the phage with the sequence of PNPNNST which occurred twice was injected into mice and the staining distribution of phage was observed. The staining was similar to that of the blood vessel marker factor VIII-related antigen staining (Figure 1B and 1C), which indicated that the selected phage was targeted to blood vessels in the tumor xenograft. There was no significant staining of phage which was observed in the lung and spleen.
The development of a functional vasculature within a solid tumor is essential for its growth and progression. This fact has led to the design of therapies targeting the tumor vasculature, aiming either to prevent the formation of new vessels (anti-angiogenic) or to damage existing vessels (vascular targeting). In addition, the differences between tumor blood vessels and blood vessels associated with normal tissue make the tumor vasculature a unique target, which may allow highly selective treatment of malignant disease.12-14 However, this tumor unique target is not easy to be achieved because the expression level of this unique target is much lower than those seen in the blood vessels of normal tissues and lack of cultured vascular endothelial cells isolated from tumor vasculature makes it even more difficult. Even though this kind of vascular endothelial cell exists, it will lose its characteristics of endothelial cell when it was cultured in vitro for a short period of time.
But now, screening in vivo using phage display peptide library may resolve this problem. Phage display peptide library is a selection technique that has been successfully applied to investigate protein stability. Phage display peptide library offers a number of important advantages. The technology for generating large libraries has been well generated and developed, permitting the simultaneous characterization of a relatively large number of mutants.15 In addition, the high stability of the phage particle permits the use of a wide range of selection conditions. Selected peptides may be used as competitors for natural ligands and for the mapping of binding epitopes. This approach has been exploited for delineation of intracellular signal transduction pathways and for the selection of enzyme substrates and inhibitors.16-19 And using phage library in vivo, a few peptides specifically binding to gastric cancer, prostate cancer and so on were obtained.10,20,21 In this experiment, we also obtained nine kinds of peptide (Figure 3) specifically binding to the vasculature of human esophageal cancer, which may be candidates for anti-angiogenesis therapy.
Since the binding site of an antibody is not unique for a single antigen, several mimotopes with different amino acid sequences can be recognized by binding to different subsites within the binding site.22-26 For example, both erythropoietin (EPO) and one EPO mimetic peptide sequence, even though these ligands share no sequence or structural homology, may represent a minimum epitope on the EPO receptor for productive ligand binding.22,27 In our experiment, although there was no similarity between sequences of these nine selected peptides and the proteins in nature, high affinities toward vasculature were still observed because that phage recovery in xenograft tumor was higher than it is in control tissues. It has been established that at least three amino-acid residues provide the minimal framework for structure formation and protein-protein interaction, such as RGD, LCV and LLG to integrins,28-30 NGR to aminopeptidase.31 Whether the motifs of YS(X)N(X)W or P(X)N(XX)N obtained in the screeing are biochemical recognition units binding to vascular endothelial cells of esophageal cancer have to be further confirmed.
1. Wang Z, Liu XY, Liu FY, Chen JH. Lymph node micrometastasis in patients with esophageal cancer: diagnosis and a prospective study of impact on prognosis. World Chin J Digest (Chin) 2004; 12: 121-124.
2. He YT, Hou J, Qiao CY, Chen ZF, Song GH, Li SS, et al. An analysis of esophageal cancer incidence in Cixian country from 1974 to 1996. World J Gastroenterol 2003; 9: 209-213.
3. Wang GQ, Liu YY, Hao CQ, Lai SQ, Wang GQ, Lu N, et al. A comparative study of endoscopic image stained by iodine and histopathology in early esophageal cancer and precancerous lesions (dysplasia). Chin J Oncol (Chin) 2004; 26: 342-344.
4. Qi FY, Zhang LX, Han CL, Zuo LF, Lin PZ, Guo JW. Up-regulation of cyclooxygenase-2 in carcinogenesis of esophageal epithelia. World Chin J Digest (Chin) 2003; 11: 508-511.
5. Qin HY, Shu Q, Wang D, Ma QF. Study on genetic polymorphisms of DCC gene VNTR in esophageal cancer. World Chin J Digest (Chin) 2000; 8: 782-785.
6. Barbour AP, Lagergren P, Hughes R, Alderson D, Barham CP, Blazeby JM. Health-related quality of life among patients with adenocarcinoma of the gastro-oesophageal junction treated by gastrectomy or oesophagectomy. Br J Surg 2008; 95: 80-84.
7. Crumley AB, McMillan DC, McKernan M, Going JJ, Shearer CJ, Stuart RC. An elevated C-reactive protein concentration, prior to surgery, predicts poor cancer-specific survival in patients undergoing resection for gastro-oesophageal cancer. Br J Cancer 2006; 94: 1568-1571.
8. Quera R, O’Sullivan K, Quigley EM. Surveillance in Barrett’s oesophagus: will a strategy focused on a high-risk group reduce mortality from oesophageal adenocarcinoma? Endoscopy 2006; 38: 162-169.
9. Zhi M, Wu KC, Dong L, Hao ZM, Deng TZ, Hong L, et al. Characterization of a specific phage-displayed peptide binding to vasculature of human gastric cancer. Cancer Biol Ther 2004; 3: 1232-1235.
10. Liang S, Lin T, Ding J, Pan Y, Dang D, Guo C, et al. Screening and identification of vascular-endothelial- cell-specific binding peptide in gastric cancer. J Mol Med 2006; 84: 764-773.
11. Bai F, Liang J, Wang J, Shi Y, Zhang K, Liang S, et al. Inhibitory effects of a specific phage-displayed peptide on high peritoneal metastasis of gastric cancer. J Mol Med 2007; 85: 169-180.
12. Siemann DW, Chaplin DJ, Horsman MR.Vascular-targeting therapies for treatment of malignant disease. Cancer 2004; 100: 2491-2499.
13. Taraboletti G, Giavazzi R. Modelling approaches for angiogenesis. Eur J Cancer 2004; 40: 881-889.
14. Folkman J. Angiogenesis and apoptosis. Semin Cancer Biol 2003; 13: 159-167.
15. Kotz JD, Bond CJ, Cochran AG. Phage-display as a tool for quantifying protein stability determinants. Eur J Biochem 2004; 271: 1623-1629.
16. Romanov VI. Phage display selection and evaluation of cancer drug targets. Curr Cancer Drug Targets 2003; 3: 119-129.
17. Rutledge SE, Volkman HM, Schepartz A. Molecular recognition of protein surfaces: high affinity ligands for the CBP KIX domain. J Am Chem Soc 2003; 125: 14336-14347.
18. Tayapiwatana C, Arooncharus P, Kasinrerk W. Displaying and epitope mapping of CD147 on VCSM13 phages: influence of Escherichia coli strains. J Immunol Methods 2003; 281: 177-185.
19. Sidhu SS, Fairbrother WJ, Deshayes K. Exploring protein-protein interactions with phage display. Chembiochem 2003; 4: 14-25.
20. Christianson DR, Ozawa MG, Pasqualini R, Arap W. Techniques to decipher molecular diversity by phage display. Methods Mol Biol 2007; 357: 385-406.
21. Sergeeva A, Kolonin MG, Molldrem JJ, Pasqualini R, Arap W. Display technologies: application for the discovery of drug and gene delivery agents. Adv Drug Deliv Rev 2006; 58: 1622-1654.
22. Kzhyshkowska J, Workman G, Cardo-Vila M, Arap W, Pasqualini R, Gratchev A, et al. Novel function of alternatively activated macrophages: stabilin-1-mediated clearance of SPARC. J Immunol 2006; 176: 5825-5832.
23. Kolonin MG, Sun J, Do KA, Vidal CI, Ji Y, Baggerly KA, et al. Synchronous selection of homing peptides for multiple tissues by in vivo phage display. FASEB J 2006; 20: 979-981.
24. Han ZY, Wu KC, He FT, Han QL, Nie YZ, Han Y, et al. Screening and identification of mimotope of gastric cancer associated antigen MGb1-Ag. World J Gastroenterol 2003; 9: 1920-1924.
25. Bunn HF. New agents that stimulate erythropoiesis. Blood 2007; 109: 868-873.
26. Wang P, Ding J, Lin T, Han S, Cao SS, Ge FL, et al. MDR-reversing effect of short peptide binding specifically to multidrug-resistant gastric cancer cells. Chin J Oncol (Chin) 2007; 29: 258-261.
27. Sola A, Rogido M, Lee BH, Genetta T, Wen TC. Erythropoietin after focal cerebral ischemia activates the Janus kinase-signal transducer and activator of transcription signaling pathway and improves brain injury in postnatal day 7 rats. Pediatr Res 2005; 57: 481-487.
28. Takagi J. Structural basis for ligand recognition by RGD (Arg-Gly-Asp)-dependent integrins. Biochem Soc Trans 2004; 32: 403-406.
29. Kolachala VL, Bajaj R, Wang L, Yan Y, Ritzenthaler JD, Gewirtz AT, et al. Epithelial-derived fibronectin: expression, signaling and function in intestinal inflammation. J Biol Chem 2007; 282: 32965-32973.
30. Cody V, Davis PJ, Davis FB. Molecular modeling of the thyroid hormone interactions with alpha v beta 3 integrin. Steroids 2007; 72: 165-170.
31. Majhen D, Gabrilovac J, Eloit M, Richardson J, Ambriovic-Ristov A. Disulfide bond formation in NGR fiber-modified adenovirus is essential for retargeting to aminopeptidase N. Biochem Biophys Res Commun 2006; 348: 278-287.