Novel Glypican-3-Binding Peptide for in Vivo ... - ACS Publications

Subscriber access provided by MAHIDOL UNIVERSITY (UniNet)

Article

A novel Glypican-3 (GPC3)-binding Peptide for in vivo Hepatocellular Carcinoma (HCC) Fluorescent Imaging Dongling Zhu, Yushuang Qin, Jingjing Wang, Liwen Zhang, Sijuan Zou, Xiaohua Zhu, and Lei Zhu Bioconjugate Chem., Just Accepted Manuscript • DOI: 10.1021/acs.bioconjchem.6b00030 • Publication Date (Web): 05 Feb 2016 Downloaded from http://pubs.acs.org on February 12, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Bioconjugate Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Bioconjugate Chemistry

A novel Glypican-3 (GPC3)-binding Peptide for in vivo Hepatocellular Carcinoma (HCC) Fluorescent Imaging Dongling Zhu†, Yushuang Qin†, Jingjing Wang‡, Liwen Zhang‡, Sijuan Zou†, Xiaohua Zhu*,†, Lei Zhu*,‡ †

Department of Nuclear Medicine and PET, Tongji Hospital, Tongji Medical College, Huazhong

University of Science and Technology, Wuhan, 430000, China; ‡

State Key Laboratory of Molecular Vaccinology and Molecular Diagnostics and Center for

Molecular Imaging and Translational Medicine, School of Public Health, Xiamen University, Xiamen 361005, China.

Corresponding Author * To whom correspondence should be addressed. Tel: (+) 86-592-2880642, Fax: (+) 86-5922880642; E-mail: [email protected] (L.Z.) and [email protected] (X.Z.)

ACS Paragon Plus Environment

1


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 32

ABSTRACT Glypican-3 (GPC3) is a key member of the glypican family that is expressed on the cell surface by a glycosyl-phosphatidyl-inositol (GPI) anchor. It plays a significant role in hepatocellular carcinoma (HCC) development, angiogenesis and metastasis. Most HCC overexpress GPC3, whereas little GPC3 can be detected in normal adult liver and benign liver lesions. Therefore, it is important to understand the function of GPC3 in HCC tumor development as the GPC3 ligand may facilitate detection of HCC. In this study, a 12-mer peptide with the sequence of DHLASLWWGTEL (denoted as TJ12P1) was identified by screening a phage display peptide library that demonstrated ideal GPC3 binding affinity. We used TJ12P1 conjugated with nearinfrared fluorescent (NIFR) dye Cy5.5 for tumor imaging. After intravenous injection of the imaging agent, TJ12P1, xenografts of high GPC3 expressing hepatocellular carcinoma cell line, HepG2, demonstrated significantly higher tumor accumulation (Tumor/Muscle ratio: 3.98 ± 0.36) than those of low GPC3 expressing prostate cancer cell line, PC3 (Tumor/Muscle ratio: 2.03 ± 0.23). More importantly, GPC3 expression in tumor samples of patients could be visualized using TJ12P1, suggesting the potential use of this peptide as a probe for HCC detection. Our study has successfully identified a promising GPC3-binding peptide ligand for detecting the GPC3 expression in HCC not only in vitro but also in vivo by its non-invasive imaging. KEYWORDS: Phage display; Optical imaging; GPC3; HCC


2

Page 3 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


TOC


3


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 32

INTRODUCTION Hepatocellular carcinoma (HCC) is the sixth most prevalent cancer and the third most frequent cause of cancer-related deaths worldwide, especially for patients with chronic hepatitis B1. The overall 5-year survival rate of HCC is as low as 15% (www.cancer.org) due to late diagnosis and limited treatment options2. Although surgery can effectively prolong the patient survival time, it is only limited to HCC identified at an early stage. Currently, histological examination of the biopsy material, the most widely applied approach to detect HCC3, is invasive and time-consuming4, while the prevalent Alphafetoprotein (AFP) test lacks adequate sensitivity for effective diagnosis of HCC5. A sensitive and specific non-invasive HCC detection approach is therefore needed urgently for improving patient survival and reducing morbidity. Glypicans (GPCs) are a family of heparansulfate proteoglycans that anchor on the cell membrane with a glycosylphosphatidylinositol (GPI) linkage6. This family consists of six members (GPC1-GPC6) in mammals. In particular, many studies have reported that GPC3, a 70 kDa protein, is a promising HCC-specific target as its expression is significantly elevated in HCC, whereas it can be rarely detected in normal adult liver and benign liver lesions7-10. GPC3 is composed of a core protein and two heparansulfate chains and participates in a variety of pathways related to HCC, such as Wnt, Yap and FGF. Through interactions with these signaling pathways, GPC3 can regulate HCC development, metastasis and angiogenesis6,

10-13

. For example, overexpressed GPC3 is

capable of reacting with the Wnt ligand and facilitating Wnt/Frizzled interactions, which are believed to be vital in the progression of many cancers, including HCC6,

11, 13

.

Furthermore, it has been shown that mutations in GPC3 or knockdown of its function can


4

Page 5 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


inhibit HCC growth, reinforcing the important roles of GPC3 in HCC development12, 14, 15

.

A probe targeting GPC3 may facilitate the diagnosis of HCC or even differentiate malignant liver nodules from benign ones. Previously,

89

Zr-conjugated anti-human GPC3

mAb (89Zr-DFO-1G12) was reported to detect HepG2 tumors successfully16 and specifically accumulate in both subcutaneous and orthotopic HepG2 xenografts. αGPC3 IgG1 and the fragments of αGPC3 IgG1 (named as αGPC3 (Fab′)2) were labeled with 89

Zr for identifying orthotopic HepG2 xenografts by positron emission tomography

(PET)17, 18. Orthotopic HepG2 xenografts showed higher uptake of

89

Zr-αGPC3 IgG1 or

89

Zr-αGPC3 (Fab′)2 than the GPC3 negative orthotopic RH-7777 xenografts. Tumor to

liver ratio reached nearly 30.5 at 72 h postinjection of

89

Zr-αGPC3 in the orthotopic

HepG2 xenografts. Although radioisotopes-conjugated GPC3 mAbs demonstrated excellent HCC targetability and specificity, their large size and molecular weight always lead to unsatisfied in vivo pharmacokinetics, suboptimal tumor penetration and elevated immunogenicity18, 19, which results in more radiation to patients from labeling antibody with long half-life radionuclide such as 89Zr. It is therefore important to go on developing shorter half-life HCC diagnostic probes targeting GPC3 with improved sensitivity and targetability as well as reduced immunogenicity. Peptide-based molecular imaging probes have been rapidly evolving due to the advantages of low molecular weight, ease of modification and low cost for scale-up20, 21. Examples include RGD (Arg-Gly-Asp) analogs22 and matrix metalloproteinase (MMPs) substrate-based activatable probes23-26. Phage display peptide library has been proven to be an effective tool to identify protein-binding ligands by screening with purified targets,


5


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 32

cells or in vivo selection methods27-30. In this study, to identify a GPC3-specific peptide, the human recombinant GPC3 protein was utilized as a target for screening the phage display peptide library. After five rounds of screening, a 12-mer peptide (denoted as TJ12P1) was identified that bound to GPC3 with tight binding ability and specificity. The binding ability and specificity between TJ12P1 and GPC3 were analyzed and confirmed by fluorescent cell staining, enzyme-linked immunosorbent assay (ELISA), dot blotting, and Western blotting. The apparent binding affinity of TJ12P1 to GPC3 was calculated to be 280.4 ± 33.51 nM. In vivo and ex vivo tests also confirmed the specific targeting of TJ12P1 against HepG2 tumors. Importantly, by specifically targeting GPC3, TJ12P1 was able to differentiate between tumor tissues from patients and cholangiocarcinoma (CCA) as well as normal adult liver tissues, highlighting a great translational potential for HCC diagnosis. RESULTS Peptide characterization To obtain a specific and high affinity GPC3 peptide binder, the recombinant human GPC3 protein was immobilized as a target for phage display screening. Phage clones binding with GPC3 protein gradually concentrated over the five rounds screening (Table S1). To get the clones with high binding affinity and specificity for GPC3, Tween-20 concentration in the washing steps was increased stepwise (0.1%-0.5%). In the last round of screening, clones binding with GPC3 target were competitively washed off by GPC3 protein to ascertain the specificity of clones. Peptides displayed on the phage clones were sequenced and revealed that a peptide with the sequence of DHLASLWWGTEL was present at a frequency of 70 % among the 40 sequencing samples (Table 1, Figure 1A).


6

Page 7 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


We named this peptide as TJ12P1 and synthesized it for further applications. The purification and characterization of TJ12P1 are presented in Figure S1. Before moving forward, the stability of TJ12P1 was studied by incubation it with mouse serum for 24 hours. As Figure S2 shown, TJ12P1 is stable in the first 6 hours when incubated with mouse serum that 86.44±8.02% TJ12P1 is left. However, the intact peptide amount decreased after 6 hours and 62.81 ± 7.11% left at 24 h after incubation, implying a rapid metabolism of TJ12P1. Next, binding affinity between TJ12P1 and the human recombinant GPC3 protein was determined by the ELISA test. Scrambled TJ12P1 with the sequence of WLSHLGDLTWEA was used as control to confirm the binding affinity and specificity between TJ12P1 and the human recombinant GPC3 protein. As shown in Figure 1B, the apparent Kd value between biotin-TJ12P1 and the human recombinant GPC3 protein was 280.4 ± 33.51 nM. The biotin-TJ12P1-scr, on the other hand, did not show any binding affinity against the human recombinant GPC3 protein, indicating that the tight binding ability of TJ12P1 with GPC3 was sequence dependent. In order to detect the GPC3 expression in vivo, a near-infrard dye, Cy5.5 (ex/em: 670/690 nm) was purposely conjugated onto TJ12P1 and binding affinity was analyzed and calculated as 390.03 ± 27.47 nM, suggesting the conjugation of dye molecule onto TJ12P1 did not signficantly affect its binding affinity to GPC3 (Figure 1B). The chemical structure and characterization of Cy5.5-TJ12P1 is shown in Figure S2. The exact mass weight for Cy5.5-TJ12P1 was measured as 1990.99.


7


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 32

Table 1. Ratio of different peptide sequence after five rounds screening Peptide sequence

Numbers

Recurrence (%)

DHLASLWWGTEL

28

70

DWSSWVYRDPQT

9

22.5

EHFSLWWNPPLV

2

5

NHFSYEFWFSLP

1

2.5


8

Page 9 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 1. (A) The peptide sequence of TJ12P1: the sequence was identified by five rounds of screening of a phage display library. (B) The human GPC3 protein-binding affinity of biotinTJ12P1 and biotin-TJ12P1-scr: The apparent Kd value between biotin-TJ12P1 and the human recombinant GPC3 protein was 280.4 ± 33.51 nM. The apparent Kd value between Cy5.5TJ12P1 and the human recombinant GPC3 protein was 390.03 ± 27.47 nM. No concentration dependent binding between biotin-TJ12P1-scr and the human recombinant GPC3 protein was found.


9


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 32

Immunoblotting analysis To further confirm the binding specificity of TJ12P1 to GPC3, the peptide was used for in vitro detection of the GPC3 expression by blotting tests. The GPC3 expression in the HepG2 and PC3 cells was confirmed by Western blotting. As shown in Figure S4A, a high expression of GPC3 was observed in the HepG2 cells, whereas little GPC3 was detected in the PC3 cells. When normalized with β-actin, a nearly five times overexpression of GPC3 in HepG2 than in PC3 was calculated by analyzing the band intensity in Western blotting (48% ± 8% versus 11% ± 1.2, Figure S4B). We therefore used the HepG2 and PC3 cell lines as high GPC3 expressing and low GPC3 expressing controls, respectively, in our study. Subsequently, biotinylated TJ12P1 was used instead of the primary antibody to detect GPC3 (Figure S5A). As expected, GPC3, specifically detected by TJ12P1, was comparable to that detected by the antibody in both HepG2 and PC3 cells. The band intensities of GPC3 in the HepG2 and PC3 cells detected by TJ12P1 were 55% ± 7.6 and 14% ± 2.5, respectively (Figure S5B) which is very close to that of antibody. Our results implied that TJ12P1 is capable of binding not only with pure GPC3 protein, but, like antibodies, also can detect GPC3 in cells (Figure S5).

Fluorescent cell staining The binding specificity of TJ12P1 to GPC3 was further analyzed by fluorescent cell staining. As shown in Figure 2, strong fluorescence was observed in HepG2 cells treated with Cy5.5-TJ12P1 suggesting that TJ12P1 can bind with GPC3 on the HepG2 cell membrane. It is of note that the binding of TJ12P1 to HepG2 was time dependent. At 4 h after incubation, more fluorescent dye can be observed than at 2 h due to the cell


10

Page 11 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


internalization of Cy5.5-TJ12P1. The binding could be effectively blocked when an excess amount of free TJ12P1 (1 mg) was used before Cy5.5-TJ12P1 was applied, implying the specificity of binding between TJ12P1 and the GPC3 protein. Also, few fluorescent signals could be detected on the low GPC3 expressing PC3 cells even 4 h after incubation with Cy5.5-TJ12P1.

Figure 2. Specific imaging of Cy5.5-TJ12P1 targeting GPC3 in HepG2 cells: Red color is from Cy5.5-TJ12P1 (ex/em=690/724 nm). Blue color is from DAPI for visualizing cell nuclei. After 2 h incubation, Cy5.5-TJ12P1 mainly bound to HepG2 cell surface, and was gradually internalized after 4 h. A weak signal was observed when GPC3 expressed on HepG2 cells was blocked by


11


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 32

TJ12P1 prior to labeling with Cy5.5-TJ12P1. Quite few fluorescent signals were detected in PC3 cultures. Scale bar equals 10 µm.

In vivo imaging of GPC3 using Cy5.5-TJ12P1 Optical imaging modality with low cost, high resolution, and ideal sensitivity has been used widely for the preclinical evaluation of imaging probes. Although limited by the penetration depth, the near infrared dye Cy5.5 allows a visualization of deep tissue in rodents with low background, providing a possibility of preclinical evaluation of the molecular imaging probe. In our study, Cy5.5 was conjugated on the N-terminal of TJ12P1 for in vivo GPC3 targeting. To evaluate the specificity of TJ12P1 against GPC3, HepG2 tumor models with high GPC3 expression and PC3 tumor models with low GPC3 expression were established (Figure S4). Cy5.5-TJ12P1 was intravenously injected into tumor-bearing mouse models at a concentration of 10 nmol. As the results presented in Figure 3A, there was a higher accumulation of Cy5.5-TJ12P1 in the HepG2 tumors (n=3) than in the PC3 tumors (n=3). The signals peaked at 4 h postinjection (p.i.) and decreased after 4 h. At 24 h p.i., fluorescent signal could only be observed in the tumor area because of the strong binding of the TJ12P1 with GPC3 expressed on the tumor cells. When an excess amount of 4 mg free TJ12P1 was injected into the mouse, the GPC3 binding sites were saturated and therefore binding of Cy5.5-TJ12P1 in the tumor area was insignificant demonstrating the specificity of Cy5.5-TJ12P1 binding to GPC3 (Figure 3A). Some fluorescent signal could be seen in the tumor areas when Cy5.5-TJ12P1 scrambled peptide or free Cy5.5 were injected. However, the signal was significantly weak and was probably caused by the rich blood vessels distribution in the tumor.


12

Page 13 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Quantification of optical imaging was carried out in the regions of interest (ROI). Fluorescent intensity in each tumor at 4 h p.i. is shown in Figure 3B. The results indicate that HepG2 tumors in the experimental groups had a higher accumulation of Cy5.5TJ12P1 than the control groups at 4 h time point. To analyze the capability of Cy5.5TJ12P1 targeting high GPC3 expressing HepG2 tumors, tumor to muscle ratios were calculated and are shown in Figure 3C. Fluorescent tumor/muscle ratio at 4 h p.i. for HepG2 tumor xenografts was 3.98 ± 0.36, while the ratios were 2.24 ± 0.26, 1.14 ± 0.12 and 2.03± 0.23 in the Cy5.5-TJ12P1-scr, free Cy5.5 and PC3 control groups, respectively. Also, the tumor accumulation can be significantly inhibited by administering excess amount of the free TJ12P1 peptide prior to the injection of Cy5.5-TJ12P1. Our in vivo data suggest that Cy5.5-TJ12P1 specifically binds to GPC3 in vivo and can detect HepG2 tumors.


13


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 32

Figure 3. In vivo optical imaging of Cy5.5-TJ12P1 targeting GPC3 in various tumor xenografts: (A) Representative sagittal images of HepG2, PC3 tumor-bearing nude mice and HepG2 tumor-bearing nude mice at different time points after intravenous injection of 10 nmol of imaging agent. White arrows indicate tumors. HepG2 tumors had higher accumulations of Cy5.5-TJ12P1 than PC3 tumors, and fluorescent intensity had a significant decrease when HepG2 tumors were blocked with TJ12P1. Less fluorescence was found in the Cy5.5-TJ12P1-scr and free Cy5.5 control groups. (B) Fluorescent intensity (ROI analysis) at 4 h p.i. in each tumor. (C) Tumor/Muscle Ratio at 4 h p.i. (mean ± SD, n=3/group). * P< 0.05.


14

Page 15 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Biodistribution To confirm the in vivo high tumor uptake of Cy5.5-TJ12P1, mice were sacrificed at 24 h p.i. and major organs were collected for ex vivo fluorescent imaging. As shown in Figure 4, tumors dissected from the HepG2 xenografts in the experimental group had a higher uptake of Cy5.5-TJ12P1 than the blocking control, Cy5.5-TJ12P1-scr, or the PC3 control group. Besides the tumor, kidney also showed high uptake of Cy5.5-TJ12P1. This could be due to the fact that degraded peptide fragments of Cy5.5-TJ12P1 are likely cleared out by the kidneys. Figure 5 shows the histological analysis of tumor cryosections with a predominant presence of Cy5.5-TJ12P1 in HepG2 tumors, and very little of Cy5.5TJ12P1 in the blocking groups. Few signals from Cy5.5 were found either in PC3 tumor models or with the negative control with Cy5.5-TJ12P1-scr (Figure 5), further confirming the specificity of binding of Cy5.5-TJ12P1 to GPC3.

Figure 4. Fluorescent images of organs and tissues: the images were taken at 24 h p.i. of 10 nmol imaging agent per mouse. (A) HepG2 xenograft, (B) HepG2 blocked with TJ12P1, (C) HepG2 treated with Cy5.5-TJ12P1-scr, (D) PC3 xenograft. 1, heart; 2, liver; 3, kidney; 4, muscle; 5, spleen; 6, lung; 7, tumor; 8, intestine. (E) Biodistribution in tumor xenografts at 24 h


15


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 32

p.i. of 10 nmol imaging agent. * P< 0.05.

Figure 5. Histochemical localization of fluorescence in tumor slices after 4 h intravenous injection. Tumor uptake of Cy5.5-TJ12P1 in HepG2 was higher than that in PC3. Fluorescence intensity in the HepG2 was much reduced when blocked with TJ12P1 or when Cy5.5-TJ12P1-scr was used. Red color is from Cy5.5 (ex/em=690/724 nm). Blue color is from DAPI for nuclei visualization. Scale bar equals 50 µm.

Tumor histochemistry


16

Page 17 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


To investigate its translational potential in the clinic, TJ12P1 was labeled with Cy5.5 for the detection of GPC3 expression in liver tissues from patients and healthy individuals. Proteins were extracted from tumor tissues of HCC and CCA patients and normal adult liver tissues (denoted as HCC-PTT, CCA-PTT and NA-LT, respectively). As confirmed by Western blotting, GPC3 was highly expressed in HCC-PTT, but it was not detectable in CCA-PTT and NA-LT (Figure S6). The same samples were incubated with Cy5.5TJ12P1 and the results are displayed in Figure 6A. Fluorescent Western blotting shows binding of Cy5.5-TJ12P1 with GPC3 in HCC-PTT samples, but not in CCA-PTT and NA-tissue samples (Figure 6A). We also performed the fluorescent Western blotting on a variety of samples including tumor cell lines as well as tissues from human tumors and healthy individuals. Figure 6B clearly shows more intense staining of Cy5.5TJ12P1 in HepG2 tumor slices and HCC-PTS samples than in PC3 tumor slices, CCA-PTS and NALTS. A similar fluorescent intensity was detected when these tumor slices were stained with anti-GPC3 primary antibody. Thus our experiments confirmed that TJ12P1 is capable of binding with GPC3 protein in cell lines as well as in clinical samples from patients.


17


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 32

Figure 6. Clinical application of TJ12P1 targeted to GPC3. (A) GPC3 was detected by Cy5.5TJ12P1 on the PVDF membrane only in HCC-PTT but not in NA-LT and CCA-PTT. (B) Both Cy5.5-TJ12P1 and the anti-GPC3 primary antibody detected GPC3 in HepG2 tumor slices and HCC-PTS but there was little detectable fluorescence in PC3 tumor slices, CCA-PTS and NALTS. Green color is from FITC (ex/em=488/530 nm). Red color is from Cy5.5 (ex/em=690/724 nm). Blue color is from DAPI. Scale bar equals 50 µm.

DISCUSSION


18

Page 19 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Several studies have employed intact monoclonal antibodies and their fragments against GPC3 for the multimodality HCC imaging16-18,

31

. However, large protein-based

radionuclide molecular imaging probes, when used in orthotopic tumor xenografts and detected by PET and single-photon emission computed tomography (SPECT), always result in unsatisfactory tumor uptake and pharmacokinetics. This is, in large part, due to the metabolism of these imaging probes in the liver. Therefore, small molecules, such as peptides, will be of great benefit for tumor detection22, 32-35. Previously, Lee et al used peptide ligands against GPC3 for proteomic mass spectrometry23. Peptide L5 and L5-2 showed strong targeting capability for GPC3 in cell culture studies. However, the exact binding affinity between the peptides and GPC3 was not determined, and in vivo GPC3 targetability had not been explored. Therefore, it remained challenging to develop GPC3 targeting ligands. In this study, we identified the peptide TJ12P1 by phage display technology that showed tight binding affinity against GPC3. The apparent Kd value between TJ12P1 and the human recombinant GPC3 protein was determined to be 280.4 ± 33.51 nM (Figure 1B). The TJ12P1 peptide showed specific binding to GPC3 in cell lysates, intact cells as well as tumor tissues (Figure S5, Figure 2 and 6B). Most significantly, the NIR dye-labeled TJ12P1 was able to detect GPC3 in tumors from the HCC patients by simple incubation, whereas no specific signals could be detected in GPC3-negative tumors such as in tumor tissues from the CCA patients (Figure 6). Although poor tissue penetration of fluorescent dyes is certainly a disadvantage, optical imaging with multiple fluorescent dyes enables multichannel imaging and relatively rapid and cost-effective preclinical evaluation compared with radionuclide-based imaging. It is


19


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 32

therefore playing an increasingly important role for preclinical tumor detection36-38 as well as in vivo evaluations of the probes. Herein, we confirmed by optical imaging that NIR-labeled TJ12P1 specifically targets high GPC3 expressing HepG2 tumors (Figure 3A). The tumor/muscle ratios reached a peak 3.98 ± 0.36 at 4 h postinjection (Figure 3C). It should be mentioned that there is some uptake of Cy5.5-TJ12P1 by the liver which may affect HCC detection in the future. It is possible that the liver uptake is caused by the hydrophobicity of Cy5.5-TJ12P1 attributed to the insolubility of Cy5.5. Several approaches may be helpful to overcome this problem, for example conjugation of TJ12P1 with more hydrophilic chemical groups will be able to reduce the normal liver uptake of TJ12P and thus providing better imaging of HCC. Additionally, TJ12P1 can be cyclized or multimerized to alter its hydrophobicity, stability, and even affinity. Currently, we are investigating the potency of replacing hydrophobic amino acids in TJ12P1 to reduce its normal liver uptake with computer assisted modeling. Recent studies have elucidated close interactions of GPC3 with Wnt and other signaling pathways6, 11, 13. Through these interactions, GPC3 is believed to promote tumor growth and development. We hypothesize that interruption of the interactions between GPC3 and various signaling pathways may induce tumor cells death and ablation of tumor growth. In this respect, the specific binding affinity of TJ12P1 to GPC3 can be exploited by incorporating chemotherapeutic drugs in nanoplatforms using TJ12P1 for HCC-targeted therapy. In conclusion, a peptide with strong binding affinity against GPC3 was identified in vitro and in vivo. Our study has furnished a reliable peptide-based imaging agent, Cy5.5-TJ12P1, which can be used to monitor GPC3, with high specificity. TJ12P1 will not only be valuable for GPC3-targeted HCC detection and therapy, but it may also


20

Page 21 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


facilitate understanding the role of GPC3 in HCC development. The related investigations targeting GPC3 by TJ12P1 are currently under way. EXPERIMENTAL PROCEDURES In vitro panning of GPC3 binding peptide The Ph.D.™-12TM phage display peptide library was screened according to the commercial kit manual (New England Biolabs Inc. USA). Briefly, the human recombinant GPC3 protein (Sino Biological Inc. Beijing, China) was dissolved in 0.1 M NaHCO3 (pH 8.6) at a concentration of 100 µg/mL and immobilized on a 35 mm polystyrene dish in a humidified incubator and stored at 4 °C overnight. Next day, the human recombinant GPC3 protein was blocked by the blocking buffer (0.1 M NaHCO3, pH 8.6, 5 mg/mL BSA, 0.02% NaN3) for 1 h at 37 °C. Subsequently, 10 µL phage display peptide library (4×1010 phages) diluted in 1 mL TBST were exposed to the dish for 1 h. Unbound phages were then washed off with TBST ten times and bound phages were eluted by adding 2 mL of 0.2 M Glycine-HCl (pH 2.2) containing 1 mg/mL BSA to the plate for 10 min. The solution was then neutralized with 150 µL of 1 M Tris-HCl (pH 9.1). 1 µL of selected binding phages were collected for phage tittering. The remaining phages were mixed with 20 mL of ER2738 culture (at early log stage) with moderate shaking at 37 °C for 4.5 h. The culture was centrifuged at 10,000 rpm at 4 °C for 10 min, and then the upper 80% of the supernatant was transferred to a fresh tube and added to 1/6 volume of NaCl/PEG (2.5 M NaCl with 20% [w/v] PEG-8000). The phages were allowed to precipitate overnight at 4 °C, and then were centrifuged for 10 min at 10,000 rpm, at 4 °C. The amplified phages were collected and pipetted to 200 µL of TBS buffer (50 mM Tris and 150 mM NaCl, pH 7.5). Subsequently, the titer was determined on


21


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 32

LB/IPTG/Xgal (Solarbio) plates, and the quantified amplified phage clones were used for the next round of panning. This panning procedure was repeated three more times. The fifth round of panning procedure was slightly altered by using human recombinant GPC3 protein (100 µg/mL) for competing off the GPC3 binding peptide. During the five rounds of screening, the concentration of Tween-20 was increased stepwise in the washing procedures (0.1-0.5%). Finally, appropriate DNA regions of the fifth round panning phages were sequenced, and corresponding peptide sequences were determined using − 96 gIII sequencing primer. Peptide synthesis and labeling TJ12P1 peptide with the sequence of DHLASLWWGTEL was prepared by using a peptide synthesizer (CS Bio Co. Inc.) and purified by high performance liquid chromatography (HPLC) on a C18 column. For purification, the flow rate was set as 10 mL/min, with a linear gradient of 10% to 55% acetonitrile/water (0.1% trifluoroacetic acid) over 30 min (C18 column, 5 µm, 250 × 20 mm). The collected fractions were analyzed by analytical RP-HPLC, using 5% to 65% acetonitrile containing 0.1% TFA versus distilled water containing 0.1% TFA over 30 minutes at a flow rate of 1 mL/min (C18 column, 5 µm, 120 Å, 250 × 4.6 mm). Using electrospray mass spectrometry, the exact mass weight of the peptide was determined. The scrambled sequence WLSHLGDLTWEA of TJ12P1 (denoted as TJ12P1-scr) for a comparative study was synthesized and purified in the same way as above. For biotin conjugated TJ12P1 (named as biotin-TJ12P1), TJ12P1 (1.4 µmol) in 200µL of N, N-dimethylformamide (DMF) was mixed with 1.4 µmol biotin N-hydroxysuccinimide (Seebio Biotech, Inc. Shanghai, China) containing 2% diisopropylethyalamine (DIPEA).


22

Page 23 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


The reaction was sustained for 1 h (pH 8.0) at room temperature, and then stopped by adding 20 µL acetic acid. Next, the solution was purified through HPLC, and the desired compound was collected and lyophilized. At the same time, biotin-TJ12P1-scr as a control was obtained as described above. To synthesize Cy5.5 (ex/em: 670/690 nm) labeled with TJ12P1 (named as Cy5.5TJ12P1), 1 mg of TJ12P1 dissolved in 200 µL of DMF was mixed with 0.5 mg of Cy5.5 N-hydroxysuccinimide ester (Seebio Biotech, Inc. Shanghai, China) containing 2% DIPEA. The reaction was carried out for 1 h at 37 °C with moderate stirring and then stopped by adding 20 µL of acetic acid. Subsequently, the mixture was purified by reversed-phase HPLC on a C18 column, and purified solution of Cy5.5-TJ12P1 was collected and lyophilized. Cy5.5-TJ12P1-scr was also synthesized for in vivo and ex vivo comparative studies. Measurement of binding affinity To achieve the apparent Kd value between the human recombinant GPC3 protein and TJ12P1, the human recombinant GPC3 protein (100 ng/well) was immobilized on a 96well plate at 4 °C overnight. Next day, the human recombinant GPC3 protein was incubated with 8 gradient diluted concentrations of biotin-TJ12P1 at 37 °C for 1 h (n=3/concentration). The horseradish peroxidase (HRP)-conjugated avidin (AB11037, Life Science Product & Services, Inc. 1:3000) was used as the second antibody in this test. After the reaction between biotin-TJ12P1 and avidin-HRP for 1 h at 37 °C, 100 µL TMB (PW026, Sangon Biotech, Inc.) was added to each well, and the plate was kept in dark for 15 min at 37 °C. Fifty µL of H2SO4 was then added to each well to stop the reaction. At the end of the test, the absorbance at 450 nm of the 96-well plate was


23


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 32

measured in microtiter plate reader (Thermo Scientific, Beijing, China). BiotinWLSHLGDLTWEA (biotin scrambled-TJ12P1, denoted as biotin-TJ12P1-scr) was used as a control to confirm the binding affinity between TJ12P1 and the human recombinant GPC3 protein. Cell culture and animal model HCC cell line HepG2 and prostate cancer cell line PC3 (ATCC, Manassas, VA) were cultured in DMEM medium (Thermo Scientific, Beijing, China) containing 10% fetal bovine serum (PAA, Chalfont St Giles, U.K.) supplemented with 0.1% penicillin (100 µg/mL) and streptomycin (100 µg/mL) at 37 °C with 5% CO2. Immunoblotting analysis For Western blotting, the proteins were extracted from HepG2 and PC3 cell lines, and protein concentration was first measured roughly by ultraviolet spectrophotometry. Total protein (50 µg) was respectively separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis and transferred to polyvinylidene difluoride (PVDF) membranes. In this study, β-actin was used as the internal reference. Membranes were blocked by 5% BSA for 1 h at room temperature, and then incubated with the first rabbit anti-GPC3 antibody (sc-11395, Santa Cruz Biotechnology, Inc. 1:1000) and mouse anti-β-actin antibody (sc1615, Santa Cruz Biotechnology, Inc. 1:1000) overnight at 4 °C. All membranes were then washed three times with TBST for a total of 30 min. The membranes were subsequently incubated with the second HRP-conjugated goat anti-rabbit antibody (sc2004, Santa Cruz Biotechnology, Inc. 1:5000) and HRP-conjugated goat anti-mouse antibody (sc-2005, Santa Cruz Biotechnology, Inc. 1:5000) for 1 h at room temperature. The membranes were washed again as above, and all the incubation and washing


24

Page 25 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


conditions were carried out at a moderate shaking speed. Finally, expression level of GPC3 and β-actin in HepG2 and PC3 cells were detected under ECL chemiluminescence (molecular imager, ChemiDoc XRS+, Bio-RAD). In our study, biotin-TJ12P1 and HRPconjugated avidin were replaced as the first and second antibodies, respectively for Western blotting to confirm the binding affinity between TJ12P1 and GPC3 and determining the capability of TJ12P1 to recognize GPC3. For fluorescent dot blotting, protein was extracted from HepG2 and PC3 cells. The ratio of protein concentration between HepG2 and PC3 was the same as in western blotting, but test procedures were somewhat simplified. Briefly, total protein from HepG2 and PC3 cells was added to a methanol-pretreated polyvinylidene difluoride (PVDF) membrane (n=3/sample), and then the membrane was incubated with biotin-TJ12P1/streptavidinCy5 (S000-06, Rockland, Inc. 1:5000) as

per Western blotting procedure. The free

streptavidin-Cy5 group was used as a control in this test. Finally, optical images (Carestream FX Pro) of the membrane were acquired to show the difference in fluorescence staining between HepG2 and PC3 cells. To confirm the capability that TJ12P1 can monitor GPC3 effectively is suitable for clinical

application.

Protein

extracted

from

HCC

patient

tumor

tissues,

cholangiocarcinoma (CCA) patient tumor tissues (GPC3 negative) and normal adult liver tissues (peri-tissues from benign liver lesions, GPC3 negative) were selected to react with GPC3 ligands (tissues were obtained from Zhangshan Hospital, affiliated to Xiamen University, Xiamen, Fujian, China). Detail test procedures of Western blotting and fluorescent Western blotting were as described above. To analyze the binding specificity of TJ12P1 against GPC3, we directly applied Cy5.5-TJ12P1 to human tissues for


25


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 32

fluorescent Western blotting and observed with fluorescent imaging system (IVIS Lumina II, Caliper). Fluorescent cell staining For cell labeling with Cy5.5-TJ12P1, HepG2 cells were seeded in eight-well chambers at a concentration of 3×104 cells/well. Next day, the HepG2 cells were fixed by 90% cold alcohol for 20 min and blocked by 10% BSA at 37 °C for 1 h and subsequently incubated with 10 nM Cy5.5-TJ12P1 for 2 and 4 h. To confirm the specificity of binding of Cy5.5TJ12P1 to GPC3 in the HepG2 cells, 1 mg of TJ12P1 was applied before adding Cy5.5TJ12P1 as a blocking control. For a comparative study, PC3 cells were also incubated with 10 nM Cy5.5-TJ12P1 for 2 h. The well chambers were washed with PBS three times.

Cells

were

then

mounted

with

4′,

6-diamidino-2-phenylindole

(DAPI,

Bioengineering Co., Ltd, Shanghai, China) for nuclei visualization and detected using confocal microscope (LSM780, Carl Zeiss). Whole body small animal optical imaging Animal experiments were conducted under protocols approved by Animal Care and Use Committee (CC/ACUCC) of Xiamen University. HepG2 and PC3 tumor models were established by injecting 6 × 106 cells to the right flank of the female athymic nude mice (6 weeks old, female, 19-21 g). When the tumor volume was reached 150 mm3, the mice were used for optical imaging. In vivo imaging system (Carestream FX Pro) was used to conduct optical imaging acquisition in HepG2 and PC3 tumor models (n=3/group). Ten nmol Cy5.5-TJ12P1 was injected to the tumor models via tail veins, and the acquisition was executed at different time points (4 h, 6 h and 24 h, ex/em=630/700 nm). During the acquisition process, nude mice were anesthetized with isoflurane. All acquired images of


26

Page 27 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


the whole body small animals were shown at the same scale. After the acquisition, regions of interest (ROI) were measured to show different fluorescent intensities in HepG2, PC3 tumor models and HepG2 tumor models with different treatments (TJ12P1 blocking, Cy5.5-TJ12P1-scr and free Cy5.5 control groups). For Cy5.5-TJ12P1-scr and free Cy5.5 control groups, 10 nmol of each imaging agent was injected via tail vein to show the strong tumor uptake of Cy5.5-TJ12P1. In the blocking group, 4 mg TJ12P1 was intravenously injected to occupy the GPC3 sites to confirm the specific tumor uptake of Cy5.5-TJ12P1. Nude mice were sacrificed, and main organs and tissues were dissected from tumor xenograft models after injecting 10 nmol imaging agents for 24 h blood circulation to confirm the targeting specificity ex vivo (IVIS Lumina II, Caliper). Fluorescent intensity was compared in each of the HepG2, PC3 tumor models and HepG2 tumor models with different treatments (TJ12P1 blocking and Cy5.5-TJ12P1-scr groups) by taking the average scaled signal from the organs and tumor tissues. Tumor histochemistry and immunohistochemistry For tumor histochemistry, HepG2 and PC3 tumor xenografts were sacrificed at 4 h postinjection of 10 nmol imaging agents. The tumors were frozen and subsequently sectioned with a thickness of 5 µm. The slices were imaged after mounting with DAPI under a confocal microscope (LSM780, Carl Zeiss) to detect the localization of Cy5.5TJ12P1/Cy5.5-TJ12P1-scr in tumor tissues. For immunohistochemistry, HepG2 and PC3 tumors, tumor tissues from HCC and CCA patients, and normal adult liver tissues were sliced into 5 µm cryosections. Before staining, tumor slices were dried in the air and fixed by cold acetone for 20 min, and then


27


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 32

dried again for 30 min at room temperature. After blocking by 10% BSA, 10 µg/mL rabbit anti-GPC3 mAb (sc-11395, Santa Cruz Biotechnology, Inc. 1:1000) and Cy5.5TJ12P1 were applied to the cryosections for 60 min at room temperature. Unbound ligands were washed off by PBS three times and sections stained by rabbit anti-GPC3 mAb followed by incubation with FITC conjugated goat anti-rabbit antibody (ZF-0311, 1:500) for 1 h at room temperature. The slices were washed again with PBS three times. After mounting with DAPI, all samples were visualized under a confocal microscope (LSM780, Carl Zeiss). Statistical analysis Results were expressed as mean ± SD. Differences within groups and between groups were checked by two-tailed paired and unpaired Student’s tests respectively. The results were considered significant with P values

Novel Glypican-3-Binding Peptide for in Vivo ... - ACS Publications

Recommend Documents